From zmailer-owner@nic.funet.fi Thu Jan 2 11:09:47 1997 Received: from taurus-1.siol.net ([193.189.160.11]) by nic.funet.fi with ESMTP id <65223-17184>; Thu, 2 Jan 1997 11:08:57 +0200 Received: from localhost by mail.siol.net with ESMTP id <292616-15787>; Thu, 2 Jan 1997 10:08:45 +0100 To: zmailer@nic.funet.fi Subject: is it OK that daemon@MAILHOST is the Sender? Date: Thu, 02 Jan 1997 10:08:36 +0100 From: tomaz@mail.siol.net Message-Id: <97Jan2.100845+0100_met.292616-15787+1@mail.siol.net> Return-Path: Hello! Is it OK that daemon is the Sender and in Return-path? See this: >From daemon@mail.siol.net Thu Jan 2 09:55:21 1997 Message-ID: <32CB775F.410C@siol.net> Date: Thu, 02 Jan 1997 09:52:47 +0100 Sender: From: Tomaz Borstnar X-Sender: Tomaz Borstnar X-Mailer: Mozilla 4.0b1 (Win95; I) MIME-Version: 1.0 To: primoz@siol.net, tomaz.test@siol.net, tomaz@mail.siol.net Subject: test X-Priority: Normal Content-Type: multipart/alternative; boundary="----------36C83F820940" Return-Path: Lines: 20 ------------36C83F820940 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii test ------------36C83F820940 Content-Transfer-Encoding: 7bit Content-Type: text/html; charset=us-ascii
 test
------------36C83F820940-- ------------- I use 2.99.44-patch1, btu had same problem on 2.99.38. I use only routes, localnames, aliases and mail.conf from old configuration. Here is my /etc/zmailer.conf: # Do not edit this file, instead edit /disk1/mail/work/zmailer-2.99.44/SiteConfig.in file ZCONFIG=/etc/zmailer.conf MAILBIN=/usr/local/mail/bin MAILSHARE=/usr/local/mail/share MAILVAR=/usr/local/mail/lib MAILBOX=/var/mail POSTOFFICE=/usr/local/mail/postoffice ROUTERDIRS=router1:router2:router3:router4 LOGDIR=/usr/local/mail/log MANDIR=${prefix}/man LIBRARYDIR=/usr/local/lib INCLUDEDIR=/usr/local/include SMTPOPTIONS="-asve -l /usr/local/mail/log/smtpserver" SCHEDULEROPTIONS="-l /usr/local/mail/log/scheduler.perflog -Q -S -H" NROUTERS=4 NOBODY=nobody LOGLEVEL="deferred: file: header_defer:" NNTPSERVER=nntp INEWSBIN=inews SENDMAILPATH=/usr/lib/sendmail RMAILPATH=/usr/bin/rmail SELFADDRESSES= DBTYPE=ndbm DBEXT=pag From zmailer-owner@nic.funet.fi Thu Jan 2 11:28:11 1997 Received: from taurus-1.siol.net ([193.189.160.11]) by nic.funet.fi with ESMTP id <64723-12146>; Thu, 2 Jan 1997 11:27:34 +0200 Received: from localhost by mail.siol.net with ESMTP id <292616-16048>; Thu, 2 Jan 1997 10:27:27 +0100 To: zmailer@nic.funet.fi Subject: Re: is it OK that daemon@MAILHOST is the Sender? In-reply-to: Your message of "Thu, 02 Jan 1997 10:08:36 +0100." <97Jan2.100845+0100_met.292616-15787+1@mail.siol.net> Date: Thu, 02 Jan 1997 10:27:18 +0100 From: tomaz@mail.siol.net Message-Id: <97Jan2.102727+0100_met.292616-16048+2@mail.siol.net> Return-Path: Silly me - adding daemon to zmailer group solved the problem. Tomaz From zmailer-owner@nic.funet.fi Thu Jan 2 12:03:22 1997 Received: from lutra.sztaki.hu ([192.84.227.1]) by nic.funet.fi with ESMTP id <65236-303>; Thu, 2 Jan 1997 12:02:52 +0200 Received: from localhost by lutra.sztaki.hu with SMTP id <142437-5494>; Thu, 2 Jan 1997 11:02:44 +0100 Date: Thu, 2 Jan 1997 11:02:38 +0100 (MET) Sender: Gabor Kiss From: Gabor Kiss To: zmailer@nic.funet.fi Subject: Re: header rewrite function for sender and recipient In-Reply-To: <96Jun26.170431hkt.18881-13330+134@uxmail.ust.hk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 26 Jun 1996, Lai Yiu Fai wrote: > I have encountered a problem that our configuration require two different > header rewriting functions for sender and recipient address. However, Zmailer > crossbar procedure only return one function and apply for both address > header. Is it possible to make it the same as sendmail that ruleset 1 > and ruleset 2 applies to sender and recipient respectively ? (Browsing the list archive I found that no one answered this question. Is it true?) zmailer-2.99.43/router/shliaise.c:crossbar() contains a comment: /* * We expect to see something like * (rewrite (fc fh fu) (tc th tu)) or ----> * ((address-rewrite header-rewrite) (fc fh fu) (tc th tu)) * back from the crossbar function. */ This could be something similar: 'crossbar' zmsh function might return a list of rewriting functions. I've tried to analyze the function calling crossbar() [zmailer-2.99.43/router/rfc822.c:sequencer()] but I couldn't found this feature implemented. However a generalized return value of 'crossbar' might be (rewrite-function (fc fh fu fa) (tc th tu ta)) [for compatibility reasons] as well as ((sender-rewrite-func recipient-rewrite-func) ((fc fh fu fa) (tc th tu ta)) The function should get the header name as argument, so it could apply different rewrite strategies on different header elements (To/Resent-To/... and From/Sender/Resent-From/Reply-To/...) if necessary. Regards Gabor (newbie zmailerist) From zmailer-owner@nic.funet.fi Tue Jan 7 05:16:46 1997 Received: from wit393101.student.utwente.nl ([130.89.235.11]) by nic.funet.fi with ESMTP id <68351-16177>; Mon, 6 Jan 1997 02:10:04 +0200 Received: from localhost ([127.0.0.1]) by wit393101.student.utwente.nl with SMTP id <23837-20287>; Mon, 6 Jan 1997 01:10:35 +0100 Date: Mon, 6 Jan 1997 01:10:30 +0100 (MET) From: Florian Overkamp Reply-To: sim@wit393101.student.utwente.nl To: zmailer@nic.funet.fi Subject: lists with 2.99.44 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi I've just upgraded my system to Zmailer 2.99.44 with patch1 applied (from .27) and stumbled upon a problem. I use a Zmailers internal maillist feature in combination with a heavily modified ListServ on my Linux machine. In the old Zmailer this worked fine, just create a file in the lists directory and off we go. But with the new version I get this : : address: sim@wit393101.student.utwente.nl : file: 23837-20286 => : address: testlist@wit393101.student.utwente.nl Usage: listexpand [ -e error-address ] [ -E errors-to-address ] [-p privilege] [ -c comment ] [ -N notarystring ] $attribute $localpart $origaddr < /file/path No routed addresses -> deferred I've been skimming through the router files but cannot find anything really different from the old version. I first thought it might be a problem with the fact that I used some configuration files from the old version, because I've defined a couple of special channels (fidonet etc.) but a friend of mine was installing the same zmailer clean, and has the same problem. I've had to fall back to the listexpand.c program but I find the internal server more elegant and I have no need for such 'heavy armor'. I'd like to figure out why the lists are not working now. Any idea's ?? Thanx in advance. /* Florian Overkamp */ /* PGP Key-ID 233FC67D */ /* PGP Fingerprint 00 3F F5 63 DD FB 5D B7 0A 16 0A EC F9 BD 01 97 */ From zmailer-owner@nic.funet.fi Tue Jan 7 05:16:47 1997 Received: from waste.org ([38.225.68.2]) by nic.funet.fi with ESMTP id <68041-19013>; Mon, 6 Jan 1997 08:36:32 +0200 Received: from localhost by waste.org with SMTP id <4155-11235>; Mon, 6 Jan 1997 00:36:12 -0600 Date: Mon, 6 Jan 1997 00:36:01 -0600 (CST) From: Oliver Xymoron To: Matti Aarnio cc: Zmailer Subject: .43b/.44 problems (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I'm having serious troubles with Zmailer this evening. I had been running .43b with a couple patches for most of December, until earlier this evening a message on one of our lists (~1000 remote subscribers) seems to have caused router to explode. Since I could find no sign of what the trouble was, beyond that it ate enough system resources to prevent console logins, and I had the archive for 2.99.44 handy, I did a clean install, including the patch to functions.c from the 28th. All seemed well, until I tried to expand one of my majordomo outgoing aliases and discovered this: Usage: listexpand [ -e error-address ] [ -E errors-to-address ] [-p privilege] [ -c comment ] [ -N notarystring ] $attribute $localpart $origaddr < /file/path which suggested the following fix: --- aliases.cf~ Sun Jan 5 21:40:36 1997 +++ aliases.cf Sun Jan 5 22:44:50 1997 @@ -122,8 +122,11 @@ priv=$(get $attr privilege) [ -f "\1" ] && priv=$(getpriv "644" $priv "\1" include) && - nattr=($newattribute $attr privilege $priv) && + nattr=$(newattribute $attr privilege $priv) && defer='' return $(listexpand -E "$user-owner@$mydomain" -p $priv \ -e root -c "\1 file expansion" \ $nattr "\1" "$user" < "\1") This fixed, I proceeded to try and send a message to a test list, which resulted in these lines at the end of router's log: : address: test-outgoing s_apply() ran out of svstack[]! (svstop=5) And router promptly disappears without leaving a core(?!). My experimental fix: --- router/shliaise.c~ Sun Jan 5 21:11:06 1997 +++ router/shliaise.c Mon Jan 6 00:02:32 1997 @@ -11,7 +11,7 @@ struct conscell **return_valuep = 0; struct conscell *s_value = 0; int svstop = -1; -#define SVSTOPMAX 5 /* Is this too low ? */ +#define SVSTOPMAX 20 /* Is this too low ? */ struct conscell *svstack[SVSTOPMAX] = { 0 }; extern int l_apply __((u_char *fname, struct conscell *l)); ...and it's still running after 20 minutes and a couple mailing list messages, even though I have no idea what the above module is for. Anyone seen similar problems with .43b or .44? -- "Love the dolphins," she advised him. "Write by W.A.S.T.E.." From zmailer-owner@nic.funet.fi Tue Jan 7 05:16:49 1997 Received: from sid.microtec.net ([204.50.80.60]) by nic.funet.fi with ESMTP id <66362-15203>; Mon, 6 Jan 1997 18:38:57 +0200 Received: from localhost (asid@localhost [127.0.0.1]) by sid.microtec.net (8.8.4/8.8.4) with SMTP id LAA00158 for ; Mon, 6 Jan 1997 11:38:51 -0500 Date: Mon, 6 Jan 1997 11:38:51 -0500 (EST) From: Alexandre Sidorenko To: zmailer@nic.funet.fi Subject: router instability Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi, we're running zmailer-2.99.44p1 on Linux, kernel 2.0.27 libc-5.4.17. All the components of Zmailer are quite stable save router. Namely, when I restart the router (-dkn 4) with zmailer router sometimes the 4 router processes just die in 10-30 seconds, delivering some mail and without any nasty diagnostics. After I issue 'zmailer router' several times (sometimes it's more than 10) one of the router processes turns to zombie, but the remaining 3 become quite stable and after that Zmailer runs without any problems for many days (2.99.43 demonstarted the same behaviour, but with 1 zombie router it was successfully working in heavy traffic conditions for more that 2 weeks). Removing '*.fc' files and restarting the router does not help - the number of attempts to obtain in this 1-zombie+3router state is random, typically 3-10. It looks just like there's some memory initialization problem and it starts successfully when malloc grabs memory filled by something. Has anybody seen the similar behaviour ? Thanks in advance, ------------------------------------------------------------- Alexandre Sidorenko Senior System Administrator asid@microtec.net Internet Microtec, Montreal, QC ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Thu Jan 16 19:28:15 1997 Received: from media.netmediatech.com ([207.34.208.130]) by nic.funet.fi with ESMTP id <67459-3670>; Thu, 16 Jan 1997 19:27:00 +0200 Received: from net ([207.34.208.137]) by media.netmediatech.com with SMTP id <43258-7297>; Thu, 16 Jan 1997 12:28:19 -0500 Sender: pzi@netmediatech.com Message-ID: <32DE652C.41C6@netmediatech.com> Date: Thu, 16 Jan 1997 12:28:12 -0500 From: Peter Ziobrzynski Organization: netMedia Technology Inc. X-Mailer: Mozilla 3.01Gold (X11; I; IRIX 6.2 IP22) MIME-Version: 1.0 To: zmailer mailing list Subject: how to deliver mail to IMAP mailboxes Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: I noticed that recent zmailer configurations (scheduler.conf and sm.conf) have support for Cyrus IMAP server mail delivery. I eanbled the local cyrus mail delivery in scheduler.conf and all works fine for a simple (toplevel) mailboxes like joe@foo.com. Now I would like to deliver mail to mailboxes deeper in the user mailbox hierarchy like joe.work@foo.com or joe.test@foo.com. Does anybody know how to make zmailer to accept mail for this type of addresses? I guess the changes would have to be done on the smtpserver level so it accepts extensions in RCPT TO command. -- Peter Ziobrzynski, netMedia Technology Inc. 204 Richmond St. #300, Toronto Ontario, Canada, M5V-1V6 tel.(416) 596-8520x242, fax.(416) 596-8610 From zmailer-owner@nic.funet.fi Fri Jan 17 17:12:27 1997 Received: from media.netmediatech.com ([207.34.208.130]) by nic.funet.fi with ESMTP id <67798-9460>; Fri, 17 Jan 1997 17:11:22 +0200 Received: from net ([207.34.208.137]) by media.netmediatech.com with SMTP id <43258-9255>; Fri, 17 Jan 1997 10:12:37 -0500 Sender: pzi@netmediatech.com Message-ID: <32DF96E3.41C6@netmediatech.com> Date: Fri, 17 Jan 1997 10:12:35 -0500 From: Peter Ziobrzynski Organization: netMedia Technology Inc. X-Mailer: Mozilla 3.01Gold (X11; I; IRIX 6.2 IP22) MIME-Version: 1.0 To: zmailer@nic.funet.fi, info-cyrus@andrew.cmu.edu Subject: Re: how to deliver mail to IMAP mailboxes References: <32DE652C.41C6@netmediatech.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: Peter Ziobrzynski wrote: > > I noticed that recent zmailer configurations (scheduler.conf and > sm.conf) > have support for Cyrus IMAP server mail delivery. > I eanbled the local cyrus mail delivery in scheduler.conf and all works > fine for a simple (toplevel) mailboxes like joe@foo.com. > Now I would like to deliver mail to mailboxes deeper in the user mailbox > hierarchy like joe.work@foo.com or joe.test@foo.com. > > Does anybody know how to make zmailer to accept mail for this type of > addresses? > I guess the changes would have to be done on the smtpserver level so > it accepts extensions in RCPT TO command. Just to follow up on my own question: I was in the middle of IMAP RFC reading and IMAP server installation and did not understand IMAP when writing my previous message. Now I know that hierarchy of folders on the IMAP server is not supposed to be exposed for mail delivery by MTA. Only the top level folder is. But think that a feature like this that would allow IMAP users to open various folders inside their hierarchy of folders for mail delivery would be a very useful. IMAP access control could be used to make some folders visible for the MTA and behave like a valid mailboxes during SMTP exchanges. I guess this could go on the IMAP wish list... -- Peter Ziobrzynski, netMedia Technology Inc. 204 Richmond St. #300, Toronto Ontario, Canada, M5V-1V6 tel.(416) 596-8520x242, fax.(416) 596-8610 From zmailer-owner@nic.funet.fi Fri Jan 17 19:56:37 1997 Received: from THOR.INNOSOFT.COM ([192.160.253.66]) by nic.funet.fi with ESMTP id <68273-9460>; Fri, 17 Jan 1997 19:55:20 +0200 Received: from eleanor.innosoft.com by INNOSOFT.COM (PMDF V5.1-6 #8694) with SMTP id <01IEBD632468A8CURZ@INNOSOFT.COM> for zmailer@nic.funet.fi; Fri, 17 Jan 1997 09:53:23 PST Date: Fri, 17 Jan 1997 09:54:22 -0800 (PST) From: Chris Newman Subject: Re: how to deliver mail to IMAP mailboxes In-reply-to: <32DF96E3.41C6@netmediatech.com> To: Peter Ziobrzynski Cc: zmailer@nic.funet.fi, info-cyrus@andrew.cmu.edu Message-id: MIME-version: 1.0 Content-type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 17 Jan 1997, Peter Ziobrzynski wrote: > I was in the middle of IMAP RFC reading and IMAP server installation > and did not understand IMAP when writing my previous message. > Now I know that hierarchy of folders on the IMAP server is not supposed > to be exposed for mail delivery by MTA. Only the top level folder is. > > But think that a feature like this that would allow IMAP users to open > various folders inside their hierarchy of folders for mail delivery > would be a very useful. IMAP access control could be used to make > some folders visible for the MTA and behave like a valid mailboxes > during SMTP exchanges. I guess this could go on the IMAP wish list... The IMAP ACL extension allows control of the "p" (post) flag to allow mail delivery to a folder other than INBOX. The convention that the Cyrus sendmail configuration uses is that mail to username+foo@cyrus will be filed into user..foo if "p" rights are available, otherwise it goes into INBOX (user.). From zmailer-owner@nic.funet.fi Fri Jan 17 20:00:47 1997 Received: by nic.funet.fi id <68294-5042>; Fri, 17 Jan 1997 19:59:22 +0200 Subject: Re: how to deliver mail to IMAP mailboxes From: Matti Aarnio To: pzi@netmediatech.com (Peter Ziobrzynski) Date: Fri, 17 Jan 1997 19:59:19 +0200 (GMT) Cc: zmailer@nic.funet.fi, info-cyrus@andrew.cmu.edu In-Reply-To: <32DF96E3.41C6@netmediatech.com> from "Peter Ziobrzynski" at Jan 17, 97 10:12:35 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1378 Message-Id: <97Jan17.195922+0200_eet.68294-5042+212@nic.funet.fi> Return-Path: Peter Ziobrzynski wrote: > Just to follow up on my own question: > > I was in the middle of IMAP RFC reading and IMAP server installation > and did not understand IMAP when writing my previous message. > Now I know that hierarchy of folders on the IMAP server is not supposed > to be exposed for mail delivery by MTA. Only the top level folder is. If CYRUS allows injection to foouser.barfolder. or even: user.name.foofolder, then it "just" needs some changes at how the "dotfull localparts" are detected, and processed at ZMailer. Right now such ones will never go to local delivery on standard ZMailer. If such functionality is desired, we can talk about it in the ZMailer list. > But think that a feature like this that would allow IMAP users to open > various folders inside their hierarchy of folders for mail delivery > would be a very useful. IMAP access control could be used to make > some folders visible for the MTA and behave like a valid mailboxes > during SMTP exchanges. I guess this could go on the IMAP wish list... It is partly Cyrus issue -- IMAP protocol does not care about what those dots mean -- and partly a mailer issue. > -- > Peter Ziobrzynski, netMedia Technology Inc. > 204 Richmond St. #300, Toronto Ontario, Canada, M5V-1V6 > tel.(416) 596-8520x242, fax.(416) 596-8610 /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Jan 17 20:17:48 1997 Received: from media.netmediatech.com ([207.34.208.130]) by nic.funet.fi with ESMTP id <66236-9460>; Fri, 17 Jan 1997 20:16:07 +0200 Received: from net ([207.34.208.137]) by media.netmediatech.com with SMTP id <43258-9265>; Fri, 17 Jan 1997 13:17:37 -0500 Sender: pzi@netmediatech.com Message-ID: <32DFC238.2781@netmediatech.com> Date: Fri, 17 Jan 1997 13:17:28 -0500 From: Peter Ziobrzynski Organization: netMedia Technology Inc. X-Mailer: Mozilla 3.01Gold (X11; I; IRIX 6.2 IP22) MIME-Version: 1.0 To: Matti Aarnio CC: zmailer@nic.funet.fi, info-cyrus@andrew.cmu.edu Subject: Re: how to deliver mail to IMAP mailboxes References: <97Jan17.195922+0200_eet.68294-5042+212@nic.funet.fi> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: Matti Aarnio wrote: > > Peter Ziobrzynski wrote: > > Just to follow up on my own question: > > > > I was in the middle of IMAP RFC reading and IMAP server installation > > and did not understand IMAP when writing my previous message. > > Now I know that hierarchy of folders on the IMAP server is not supposed > > to be exposed for mail delivery by MTA. Only the top level folder is. > > If CYRUS allows injection to foouser.barfolder. or even: > user.name.foofolder, then it "just" needs some changes at > how the "dotfull localparts" are detected, and processed at > ZMailer. Right now such ones will never go to local delivery > on standard ZMailer. > > If such functionality is desired, we can talk about it in the > ZMailer list. I guess the sendmail behavior/feature would be a good choice for zmailer as Chris Newman mentioned in the previous message. We could hack rrouter.cf to do that - am I right? [...] -- Peter Ziobrzynski, netMedia Technology Inc. 204 Richmond St. #300, Toronto Ontario, Canada, M5V-1V6 tel.(416) 596-8520x242, fax.(416) 596-8610 From zmailer-owner@nic.funet.fi Fri Jan 17 20:37:31 1997 Received: by nic.funet.fi id <67561-9212>; Fri, 17 Jan 1997 20:35:57 +0200 Subject: Re: how to deliver mail to IMAP mailboxes From: zmailer@nic.funet.fi To: pzi@netmediatech.com (Peter Ziobrzynski) Date: Fri, 17 Jan 1997 20:35:53 +0200 (GMT) Cc: zmailer@nic.funet.fi, info-cyrus@andrew.cmu.edu Reply-To: zmailer@nic.funet.fi In-Reply-To: <32DFC238.2781@netmediatech.com> from "Peter Ziobrzynski" at Jan 17, 97 01:17:28 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 857 Message-Id: <97Jan17.203557+0200_eet.67561-9212+265@nic.funet.fi> Return-Path: > I guess the sendmail behavior/feature would be a good choice for zmailer > as Chris Newman mentioned in the previous message. A more generic approach would allow: full.name+folder.subfolder to be mapped to: userid+folder.subfolder at the mailer, and then inject it to CYRUS. How you can do this with various systems (e.g. with sendmail, and ZMailer) is another topic. An even more complex problem is to map: full.name@domain1 -> userid1 full.name@domain2 -> userid2 and still be able to do those same Andrew-style "+"-maps: full.name+folder@domain1 -> userid1+folder ... however that is not CYRUS issue either ... > We could hack rrouter.cf to do that - am I right? [ ZMailer ] yes - but it isn't sufficient. See above. > -- > Peter Ziobrzynski, netMedia Technology Inc. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Jan 17 22:07:18 1997 Received: from insosf1.netins.net ([167.142.225.5]) by nic.funet.fi with ESMTP id <65711-9460>; Fri, 17 Jan 1997 22:05:28 +0200 Received: (from root@localhost) by insosf1.netins.net id OAA30251 for zmailer@nic.funet.fi; Fri, 17 Jan 1997 14:05:13 -0600 (CST) Received: (from rtucker@localhost) by crasher2.ttgcitn.com (8.8.3/960507.rtucker) id NAA26797; Fri, 17 Jan 1997 13:54:14 -0600 Date: Fri, 17 Jan 1997 13:54:14 -0600 From: Ryan Tucker Message-Id: <199701171954.NAA26797@crasher2.ttgcitn.com> X-Loop: rtucker@puck.nether.net To: zmailer@nic.funet.fi Subject: Re: how to deliver mail to IMAP mailboxes References: <97Jan17.203557+0200_eet.67561-9212+265@nic.funet.fi> In-Reply-To: <97Jan17.203557+0200_eet.67561-9212+265@nic.funet.fi> Precedence: junk X-Mailer: TTGCITN Communications AutoResponder version 1.0/970106 X-Loop: rtucker@ttgcitn.com Return-Path: [This is an automatic response. No need to resend anything. Cool? Cool.] Your e-mail has been received. Right now, I am *really* behind with e-mail. It may take a few days to respond. Don't panic and don't try to send your e-mail again, for it's already here. If you have an urgent request, resend your message with the Priority set to high or with URGENT: in the subject line, and I'll find it rather quickly. If you are getting this and you are a list owner, make sure your mailing list software is setting the Precedence: header to one of the following: Precedence: list Precedence: bulk Precedence: junk If your list sets those, and you're still getting this, then e-mail me with URGENT: in the subject line, because I broke the danged recipe again. Also, if you didn't specifically e-mail me, then something is also broken. E-mail me with the famous URGENT: subject line and I'll get back with you. This recipe is a work in progress ;-) Apologies for any inconvenience, and thanks for your patience. -rt -- TTGCITN Communications Vacation Auto-Responding Daemon version 1.0 You should receive only one copy of this message per some random time period. If you have problems, e-mail postmaster@ttgcitn.com. And now for a random Linux cookie: ================================== "How should I know if it works? That's what beta testers are for. I only coded it." (Attributed to Linus Torvalds, somewhere in a posting) From zmailer-owner@nic.funet.fi Sat Jan 18 05:51:19 1997 Received: by nic.funet.fi id <66294-2990>; Sat, 18 Jan 1997 05:47:18 +0200 Subject: Re: .43b/.44 problems (fwd) From: Matti Aarnio To: oxymoron@waste.org (Oliver Xymoron) Date: Sat, 18 Jan 1997 05:46:22 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Oliver Xymoron" at Jan 6, 97 00:36:01 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1233 Message-Id: <97Jan18.054718+0200_eet.66294-2990+78@nic.funet.fi> Return-Path: (Olivier did heroic debugging run, and found parametrization error on aliases.cf for listexpand call. The fix is correct.) .... > : address: > test-outgoing > s_apply() ran out of svstack[]! (svstop=5) > > And router promptly disappears without leaving a core(?!). It calls abort() -- doesn't abort() drop core in your system ? It is a self-recursion caller return value storage pointer.... (mumble mumble...) With a little bit of thought I rewrote the whole static svstack[]- array away. The goal in there is to save the return_valuep -indirect pointer, reset it to some local pointer variable, get the result there, and then restore the old pointer. ... > ...and it's still running after 20 minutes and a couple mailing list > messages, even though I have no idea what the above module is for. > > Anyone seen similar problems with .43b or .44? For some reason you had several recursed calls of $(listexpand ... ) which exceeded that stack of yours. A side-result from having some addresses that expanded to more addresses via $(listexpand ... ), I guess. > -- > "Love the dolphins," she advised him. "Write by W.A.S.T.E.." /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Jan 18 05:59:09 1997 Received: by nic.funet.fi id <65478-5459>; Sat, 18 Jan 1997 05:55:18 +0200 Subject: Re: router instability From: Matti Aarnio To: asid@microtec.net (Alexandre Sidorenko) Date: Sat, 18 Jan 1997 05:49:20 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Alexandre Sidorenko" at Jan 6, 97 11:38:51 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1094 Message-Id: <97Jan18.055518+0200_eet.65478-5459+115@nic.funet.fi> Return-Path: > Hi, > > we're running zmailer-2.99.44p1 on Linux, kernel 2.0.27 libc-5.4.17. All > the components of Zmailer are quite stable save router. Namely, when I > restart the router (-dkn 4) with > > zmailer router > > sometimes the 4 router processes just die in 10-30 seconds, delivering > some mail and without any nasty diagnostics. After I issue 'zmailer > router' several times (sometimes it's more than 10) one of the router > processes turns to zombie, but the remaining 3 become quite stable and > after that Zmailer runs without any problems for many days (2.99.43 > demonstarted the same behaviour, but with 1 zombie router it was > successfully working in heavy traffic conditions for more that 2 weeks). PROPABLY this is the svstack[] limit thing that Olivier did notice too. I still wonder, why you don't get cores ? ... > Has anybody seen the similar behaviour ? > > Thanks in advance, > ------------------------------------------------------------- > Alexandre Sidorenko Senior System Administrator > asid@microtec.net > Internet Microtec, Montreal, QC /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Jan 18 07:54:14 1997 Received: by nic.funet.fi id <65689-18189>; Sat, 18 Jan 1997 07:53:37 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.45beta0 is available Message-Id: <97Jan18.075337+0200_eet.65689-18189+155@nic.funet.fi> Date: Sat, 18 Jan 1997 07:53:37 +0200 Return-Path: Hello, I have returned from USENIX, and finally had some time to look at the ZMailer. Now most of the reports that I got about 43/44 have been addressed, and "it works for me" -- at my workstation. (Which is not very much testing, thus "beta0") The autoconfiguration may have been broken as I tried to accomodate detecting a situation where gethostbyname() routine is in the libraries, but we really need separate resolver library (Nick: -l44bsd library is NOT added automagically, nor I have good idea on how to do it :( ) /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jan 21 10:30:18 1997 Received: from gaston.pivoine.septgie.fr ([194.199.139.2]) by nic.funet.fi with ESMTP id <65823-385>; Tue, 21 Jan 1997 10:29:49 +0200 Received: from samsagace.sept.fr ([194.199.139.2]) by gaston.pivoine.septgie.fr with ESMTP id <313385-12006>; Tue, 21 Jan 1997 08:29:32 +0100 Message-Id: <97Jan21.082932+0100_met.313385-12006+1@gaston.pivoine.septgie.fr> From: To: unlisted-recipients:; (no To-header on input) Date: Tue, 21 Jan 1997 08:29:30 +0100 Return-Path: Hello, I have a problem whith list expansion in zmailer 2.99.44 (idem .45) I checked the list file protection and that the owner is trusted. Zmailer is running on a SUN Ultra with Solaris 2.5.1. Any clue? Thanks . z# router deug-a-caen-84 <.interactive@gaston.pivoine.septgie.fr>: address: deug-a-caen-84 Usage: listexpand [ -e error-address ] [ -E errors-to-address ] [-p privilege] [ -c comment ] [ -N notarystring ] $attribute $localpart $origaddr < /file/path ((error expansion '":include:/opt/local/mail/lists/deug-a-caen-84"')) Jean-Luc Tardivel ям  From zmailer-owner@nic.funet.fi Tue Jan 21 19:29:49 1997 Received: by nic.funet.fi id <65610-3678>; Tue, 21 Jan 1997 19:29:00 +0200 Subject: Re: your mail From: Matti Aarnio To: tardivel@pivoine.septgie.fr Date: Tue, 21 Jan 1997 19:28:54 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <97Jan21.082932+0100_met.313385-12006+1@gaston.pivoine.septgie.fr> from "tardivel@pivoine.septgie.fr" at Jan 21, 97 08:29:30 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 739 Message-Id: <97Jan21.192900+0200_eet.65610-3678+159@nic.funet.fi> Return-Path: > Hello, > I have a problem whith list expansion in zmailer 2.99.44 (idem .45) > I checked the list file protection and that the owner is trusted. > Zmailer is running on a SUN Ultra with Solaris 2.5.1. > Any clue? Thanks . You need to pick aliases.cf from 2.99.45 package. There are actually a few more fixes in there for the router as well. > z# router deug-a-caen-84 > <.interactive@gaston.pivoine.septgie.fr>: address: deug-a-caen-84 > Usage: listexpand [ -e error-address ] [ -E errors-to-address ] [-p privilege] [ -c comment ] [ -N notarystring ] $attribute $localpart $origaddr < /file/path > ((error expansion '":include:/opt/local/mail/lists/deug-a-caen-84"')) > > Jean-Luc Tardivel /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jan 23 01:03:40 1997 Received: from sid.microtec.net ([204.50.80.60]) by nic.funet.fi with ESMTP id <68045-2441>; Thu, 23 Jan 1997 01:03:11 +0200 Received: from localhost (asid@localhost [127.0.0.1]) by sid.microtec.net (8.8.4/8.8.4) with SMTP id SAA02112; Wed, 22 Jan 1997 18:04:12 -0500 Date: Wed, 22 Jan 1997 18:04:11 -0500 (EST) From: Alexandre Sidorenko To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: router instability In-Reply-To: <97Jan18.055518+0200_eet.65478-5459+115@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="852245584-942562541-853974251=:1980" Return-Path: This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --852245584-942562541-853974251=:1980 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sat, 18 Jan 1997, Matti Aarnio wrote: > > Hi, > > > > we're running zmailer-2.99.44p1 on Linux, kernel 2.0.27 libc-5.4.17. All > > the components of Zmailer are quite stable save router. Namely, when I > > restart the router (-dkn 4) with > > > > zmailer router > > > > sometimes the 4 router processes just die in 10-30 seconds, delivering > > some mail and without any nasty diagnostics. After I issue 'zmailer > > router' several times (sometimes it's more than 10) one of the router > > processes turns to zombie, but the remaining 3 become quite stable and > > after that Zmailer runs without any problems for many days (2.99.43 > > demonstarted the same behaviour, but with 1 zombie router it was > > successfully working in heavy traffic conditions for more that 2 weeks). > > PROPABLY this is the svstack[] limit thing that Olivier did > notice too. > > I still wonder, why you don't get cores ? > ... Hello again, I have found that this bug can be reproduced (on Linux at least) by injecting the attached 2 files into the ~/rooter spooling directory. Obviously addresses in these messages are malformed, but it should not crash the router all the same... The first file (100000) makes router die. It was in the router directory since beginning of Dec.96 and was causing this 'zombie' problem on 2.99.43, 44 and now 45 patchlevels of Zmailer. The second file (100001) makes router eat memory till machine crashes ! Files are MIME-attached as routerbug.tar.gz Would you please look at that ? Thanks in advance, ------------------------------------------------------------- Alexandre Sidorenko Senior System Administrator asid@microtec.net Internet Microtec, Montreal, QC ------------------------------------------------------------- --852245584-942562541-853974251=:1980 Content-Type: APPLICATION/octet-stream; name="routerbug.tar.gz" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: H4sIAOWY5jIAA+2VXVPbOBSGudavOJteALu1azsfTkzCR4PbZiYBNsm0ndn2 QrFPiIotuZJMNv31K4fsBCiUK7bTHT0X9owk67yy5Ue+V7HzrPie1/I82IGK +3cAP6iHAK0g9BvNRqPpm6aw5dV34Jlj3VAqTSXAjhRC/2jcU/2/KPi3Rslp RmRync6lyOE1Td138fDcHfCi1LD3V+A13KbnhqHrNzuf98mS6QVMRtMLwlLk mvG5gLNzZ3Aan02dSTx+P+jHx4UQWcfxm65IV0ohuhzNXIWQGoJ622/sk3Wx brIoWXFFj28NOyRapIrD+bh/Me3JedIOgoNat2aaoWvuhwT5tYM8JSNUil6i KR1Btx6cNDzf67hhPfTuzndKNUYwLfEleHU4xQT8TqcFvhcFnagewB/VP0De mEARbAI9kuyjM6IsQxnBSHxjWUah7npvRZbC3ogmjGuhFgcwOICLi/4+GQ1G sfMepWKCR+C7HpmKCGqfaodvJXJmOmqbFU3K2RdMdARYzQ9vfvv9BThlQfqC a/OOnemqMCvQ5mu9KjLK+IHJSaVC3SuVQ1XC2HaopFzNUToxT0TK+GUE4Yxp QlI0G30usitC2BxWooQFvcabiphCqYBx0AuEjCoNS8QroDyFpeApSjNguViZ 1vVDfFeTpWTaFIQZTa5eAtO7CmaY0FKhGbUrzVWi0pkJAOsdI0q5WZ2JDWXh AnxAs5myzDy3ngXMB65iXaNcgRKCu+Rn/x7/e/xq6/vPXuMJ/3v3/d8Mm9b/ /wXf+3943j8Znp2M4rX6m25Qb7lBI3QDv3VH/jf+TlELTvVxzhIpNCaPClwv ysojx19LNJrQmLmJyNdKf7BnK/mPzgSr/ggeqnXLyB8YT8VSQVymQlIYssuF ho19jXzNSsiI5XhXyE8LtsaUcNrtpjnNaj+S7NfSZEqdQppTgM4yXLu+L/Kc VulhLGYodSX8h9fb/TIX8rJEtV1eQg83h9KYVWlSGNKcysQo+pH3/u8x0jva hj7682jcizuF4AqPeoTAPfaKaKiN829m3ifbEROGPEEQ8zlLzGEF1EiZQyGU RlBMupOEcl75neW5qZAyqk2v0bwWpSLmEJGC9+J2YfIi55gzdDcz9wKPfJ/k dufP/i8sFovFYrFYLBaLxWKxWCwWi8VisVgsFovlV+Mfh9sEawAoAAA= --852245584-942562541-853974251=:1980-- From zmailer-owner@nic.funet.fi Mon Jan 27 09:17:52 1997 Received: from netcom10.netcom.com ([192.100.81.120]) by nic.funet.fi with SMTP id <67938-3468>; Mon, 27 Jan 1997 09:17:15 +0200 Received: (from alcocer@localhost) by netcom10.netcom.com (8.6.13/Netcom) id XAA28557; Sun, 26 Jan 1997 23:17:00 -0800 Date: Sun, 26 Jan 1997 23:17:00 -0800 From: alcocer@netcom.com (Dario Alcocer) Message-Id: <199701270717.XAA28557@netcom10.netcom.com> To: zmailer Subject: File locking broken with 2.99.45 on FreeBSD 2.1 Return-Path: Is anyone using ZMailer on FreeBSD? I've installed it on my PC, 486DX80 with 32MB. After installation, I tried testing it by sending a message to another local address, but I did not see the message delivered to the local mailbox, nor did I recieve a bounce message. I went rummaging in $POSTOFFICE/queue to see if I could figure out what went wrong, and I found this in one of the message files: % channel error % todsn ORCPT=rfc822;postoffice % to % env-end % To: dummy:; (error trapped source) % From: The Post Office % Sender: mailer-daemon % Subject: Double-fault: Delivery reports about your email % Cc: The Post Office % MIME-Version: 1.0 % Content-Type: multipart/report; report-type=delivery-status; % boundary="A41C6D020.854311499=_/pino.connectnet.com" % % --A41C6D020.854311499=_/pino.connectnet.com % Content-Type: text/plain % % This is a collection of reports about email delivery % process concerning a message you originated: % % : can't dotlock "/var/spool/mail/alcocer": Permission deni % ed (this is abnormal, investigate!) How should file locking be configured on FreeBSD? In case it matters, I *was* running 2.99.26 with HOSTENV (in Config) set to FreeBSD-2.0, and I was not encountering file locking problems before. Any advice/suggestions appreciated. -- Dario Alcocer alcocer@netcom.com From zmailer-owner@nic.funet.fi Mon Jan 27 09:17:54 1997 Received: from netcom10.netcom.com ([192.100.81.120]) by nic.funet.fi with SMTP id <67743-4090>; Mon, 27 Jan 1997 09:17:30 +0200 Received: (from alcocer@localhost) by netcom10.netcom.com (8.6.13/Netcom) id XAA28639; Sun, 26 Jan 1997 23:17:24 -0800 Date: Sun, 26 Jan 1997 23:17:24 -0800 From: alcocer@netcom.com (Dario Alcocer) Message-Id: <199701270717.XAA28639@netcom10.netcom.com> To: zmailer Subject: Upgrading from 2.99.26 to 2.99.45 Return-Path: After installing 2.99.45 on my FreeBSD machine, I noticed that single-letter directories A-Z were created in the $POSTOFFICE/queue directory; I hadn't noticed this with 2.99.26. Will my message files survivie the upgrade, or will I have to somehow re-submit the messages after upgrading to 2.99.45? Thanks in advance. -- Dario Alcocer alcocer@netcom.com From zmailer-owner@nic.funet.fi Mon Jan 27 09:30:40 1997 Received: from misery.sdf.com ([204.244.210.193]) by nic.funet.fi with SMTP id <67540-3468>; Mon, 27 Jan 1997 09:30:05 +0200 Received: from misery.sdf.com [204.244.213.33] by misery.sdf.com with smtp (Exim 1.59 #1) id 0vodmj-0000kB-00; Sun, 26 Jan 1997 15:14:49 -0800 Date: Sun, 26 Jan 1997 15:14:49 -0800 (PST) From: Tom Samplonius To: Dario Alcocer cc: zmailer Subject: Re: File locking broken with 2.99.45 on FreeBSD 2.1 In-Reply-To: <199701270717.XAA28557@netcom10.netcom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sun, 26 Jan 1997, Dario Alcocer wrote: > Is anyone using ZMailer on FreeBSD? I've installed it on my PC, > 486DX80 with 32MB. After installation, I tried testing it by sending > a message to another local address, but I did not see the message > delivered to the local mailbox, nor did I recieve a bounce message. > > I went rummaging in $POSTOFFICE/queue to see if I could figure out > what went wrong, and I found this in one of the message files: > > % channel error > % todsn ORCPT=rfc822;postoffice > % to > % env-end > % To: dummy:; (error trapped source) > % From: The Post Office > % Sender: mailer-daemon > % Subject: Double-fault: Delivery reports about your email > % Cc: The Post Office > % MIME-Version: 1.0 > % Content-Type: multipart/report; report-type=delivery-status; > % boundary="A41C6D020.854311499=_/pino.connectnet.com" > % > % --A41C6D020.854311499=_/pino.connectnet.com > % Content-Type: text/plain > % > % This is a collection of reports about email delivery > % process concerning a message you originated: > % > % : can't dotlock "/var/spool/mail/alcocer": Permission deni > % ed (this is abnormal, investigate!) > > > How should file locking be configured on FreeBSD? In case it matters, > I *was* running 2.99.26 with HOSTENV (in Config) set to FreeBSD-2.0, > and I was not encountering file locking problems before. Don't do dot locking when delivering to the standard BSD mboxes. flock() locking is quite safe on FreeBSD (and faster than dotlocking). The way Zmailer does mailbox dotlocking it appears that the mail spool directory needs to be mode 1777, which I would not recommend. > Any advice/suggestions appreciated. > > -- > Dario Alcocer > alcocer@netcom.com > > Tom From zmailer-owner@nic.funet.fi Mon Jan 27 09:59:03 1997 Received: by nic.funet.fi id <67717-28739>; Mon, 27 Jan 1997 09:57:40 +0200 Subject: Re: Upgrading from 2.99.26 to 2.99.45 From: Matti Aarnio To: alcocer@netcom.com (Dario Alcocer) Date: Mon, 27 Jan 1997 09:57:39 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199701270717.XAA28639@netcom10.netcom.com> from "Dario Alcocer" at Jan 26, 97 11:17:24 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1127 Message-Id: <97Jan27.095740+0200_eet.67717-28739+725@nic.funet.fi> Return-Path: > After installing 2.99.45 on my FreeBSD machine, I noticed that > single-letter directories A-Z were created in the $POSTOFFICE/queue > directory; I hadn't noticed this with 2.99.26. Will my message files > survivie the upgrade, or will I have to somehow re-submit the messages > after upgrading to 2.99.45? Those directories are there for hashing the scheduler work-directory into smaller bits, and thus reducing the work the system has to do on directory related operations -- like when opening a file, and needing to read thru the directory at first to locate the i-node number... >From README.UPGRADING: ZMailer-2.99.37: Mailbox locks are now runtime configurable, see the manpage of mailbox.8 and also SiteConfig(.in) The scheduler is capable to spread its working set into subdirectories to shorten the system directory access time, however the setup REQUIRES matching transporters. The change is in the library function ctlopen(), which means that your custom transporters need just be relinked. > Thanks in advance. > -- > Dario Alcocer > alcocer@netcom.com /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jan 27 12:18:53 1997 Received: by nic.funet.fi id <67641-3468>; Mon, 27 Jan 1997 12:17:52 +0200 Subject: Re: File locking broken with 2.99.45 on FreeBSD 2.1 From: Matti Aarnio To: alcocer@netcom.com (Dario Alcocer) Date: Mon, 27 Jan 1997 12:17:39 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199701270717.XAA28557@netcom10.netcom.com> from "Dario Alcocer" at Jan 26, 97 11:17:00 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2953 Message-Id: <97Jan27.121752+0200_eet.67641-3468+824@nic.funet.fi> Return-Path: > Is anyone using ZMailer on FreeBSD? I've installed it on my PC, > 486DX80 with 32MB. After installation, I tried testing it by sending > a message to another local address, but I did not see the message > delivered to the local mailbox, nor did I recieve a bounce message. I have now thought about it for a few minutes (and read Tom Samplonius' comment too..) There is a runtime configurable way to fix this; From README.UPGRADING: ZMailer-2.99.37: Mailbox locks are now runtime configurable, see the manpage of mailbox.8 and also SiteConfig(.in) The default is compiled into the mailbox -- if system has flock(), the default sequence is: dotlock + flock(), else it is: dotlock + lockf() (if lockf() is not available, then it is only: dotlock) Setting the ZCONFIG parameter MBOXLOCKS=F and on scheduler.conf, set the local/* to look something like following: --------------- # Local delivery: user mail, files, processes local/* interval=60s idlemax=90s expiry=3h # want 20 channel slots, but only one HOST maxchannel=10 maxring=5 --> # We want to run mail-delivery in a sort of 'set-gid' mode --> group=mail # # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! command="mailbox -8" --------------- Albeit that might not be absolutely necessary.. (.. and onwards he mumbles about strange looking UID/GID setting codes inside the mailbox program...) I wish there were some clear and simple way to autoconfigure this aspect about the system. > I went rummaging in $POSTOFFICE/queue to see if I could figure out > what went wrong, and I found this in one of the message files: > > % channel error > % todsn ORCPT=rfc822;postoffice > % to > % env-end > % To: dummy:; (error trapped source) > % From: The Post Office > % Sender: mailer-daemon > % Subject: Double-fault: Delivery reports about your email > % Cc: The Post Office > % MIME-Version: 1.0 > % Content-Type: multipart/report; report-type=delivery-status; > % boundary="A41C6D020.854311499=_/pino.connectnet.com" > % > % --A41C6D020.854311499=_/pino.connectnet.com > % Content-Type: text/plain > % > % This is a collection of reports about email delivery > % process concerning a message you originated: > % > % : can't dotlock "/var/spool/mail/alcocer": Permission denied (this is abnormal, investigate!) > > How should file locking be configured on FreeBSD? In case it matters, > I *was* running 2.99.26 with HOSTENV (in Config) set to FreeBSD-2.0, > and I was not encountering file locking problems before. Right. It apparently uses FLOCK -only. (Definitely not using MAILLOCK) > Any advice/suggestions appreciated. > > -- > Dario Alcocer > alcocer@netcom.com /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jan 28 11:26:16 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <66817-9089>; Tue, 28 Jan 1997 11:25:06 +0200 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id OG26386; Tue, 28 Jan 1997 10:24:22 +0100 Date: Tue, 28 Jan 1997 10:24:22 +0100 From: Gabor Kiss Message-Id: <199701280924.OG26386@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: 2.99.45 'configure' problem X-Charset: US X-Char-Esc: 0 Return-Path: Changelog: * configure.in, config.h.in: ... - Test for res_init() instead of gethostbyname() to see if the system provided libraries are enough, or if more are needed (like -lresolv) This causes libnsl not to link under Solaris (2.5). At least at me. My resolver is what came with BIND 4.9.4. A quick and dirty workaround: After running 'configure' edit config.status. Replace s%@LIBSOCKET@%-lsocket%g with s%@LIBSOCKET@%-lsocket -lnsl%g Then run config.status again. Gabor From zmailer-owner@nic.funet.fi Thu Feb 6 21:00:29 1997 Received: from cs.dal.ca ([129.173.4.5]) by nic.funet.fi with SMTP id <55541-16413>; Thu, 6 Feb 1997 20:58:04 +0200 Received: by cs.dal.ca via suspension id <28121(3)>; Thu, 6 Feb 1997 14:55:44 -0400 Received: by cs.dal.ca id <28120(2)>; Thu, 6 Feb 1997 14:54:58 -0400 Date: Thu, 6 Feb 1997 14:54:56 -0400 From: David Trueman To: zmailer@nic.funet.fi Subject: 2.99.45 router core dump In-Reply-To: <96Sep14.181542+0300_eet_dst.73767-30569+60@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: We are trying to get 2.99.45 running on Linux system (debian 1.1 with kernel 2.0.28), all previous zmailer experience being with the UofT version. The router almost immediately dumps core on the following message: ------------------------------------------------------------------------- to csuite-tech-approval@csuite.chebucto.ns.ca from Majordomo-Owner@csuite.chebucto.ns.ca env-end To: csuite-tech-approval@csuite.chebucto.ns.ca From: majordomo@csuite.chebucto.ns.ca Subject: APPROVE csuite-tech Reply-To: majordomo@csuite.chebucto.ns.ca -- David Trueman requests that you approve the following: subscribe csuite-tech David Trueman If you approve, please send a message such as the following back to majordomo@csuite.chebucto.ns.ca (with the appropriate PASSWORD filled in, of cou rse): approve PASSWORD subscribe csuite-tech David Trueman If you disapprove, do nothing. Thanks! majordomo@csuite.chebucto.ns.ca ------------------------------------------------------------------------- The traceback from gdb follows: #0 find_errto (info=0x0) at rfc822.c:2171 2171 if (STRING(info)) (gdb) (gdb) bt #0 find_errto (info=0x0) at rfc822.c:2171 #1 0x800ec42 in sequencer (e=0x804df70, file=0x80653d4 "481324-10257") at rfc822.c:1680 #2 0x800b666 in run_rfc822 (argc=2, argv=0x80653e4) at rfc822.c:158 #3 0x80269e3 in execute (c=0xbfffec50, caller=0xbffff580, oretcode=0, name=0x803d326 "rfc822") at execute.c:394 #4 0x801a4b2 in runcommand (c=0xbfffec50, pc=0xbffff580, retcodep=0xbffff57c) at interpret.c:686 #5 0x801c10b in interpret (code=0x8077c60 "\r", eocode=0x8077f73 "\016H", entry=0x8077c6f "2#file", caller=0xbffff580, retcodep=0xbffff57c, cdp=0x80714cc) at interpret.c:1638 #6 0x801e266 in lapply (fname=0x803d32d "process", l=0x804df40) at interpret.c:2594 #7 0x801e31b in apply (argc=2, argv=0xbffff874) at interpret.c:2618 #8 0x800a1da in s_apply (argc=2, argv=0xbffff874) at shliaise.c:45 #9 0x8005038 in rd_doit (filename=0x8078bc2 "481324-7382", dirs=0x8055f3c "") at functions.c:660 #10 0x80052a7 in rd_stability (dirp=0x8079bb8, dirs=0x8055f3c "") at functions.c:746 #11 0x800569f in run_daemon (argc=1, argv=0xbffffd98) at functions.c:913 #12 0x8002e7f in main (argc=3, argv=0xbffffe08) at router.c:301 #13 0x8002654 in _start () #14 0x32fa2626 in _malloc_statsbuf () #15 0x4f9b800 in ?? () Cannot access memory at address 0x1408. (gdb) up #1 0x800ec42 in sequencer (e=0x804df70, file=0x80653d4 "481324-10257") at rfc822.c:1680 1680 errto = find_errto(gg); /* Possible 'ERR' valu e */ I would appreciate any hints on how to go about tracking down the problem, having never debugged a newer zmailer (and not having done a UofT one in a long time). Thanks in advance, David Trueman, Systems Manager, Dalhousie Math, Stats and Computing Science Technical Chair, Chebucto Community Net From zmailer-owner@nic.funet.fi Thu Feb 6 21:07:24 1997 Received: by nic.funet.fi id <55277-31139>; Thu, 6 Feb 1997 21:05:04 +0200 Subject: Re: 2.99.45 router core dump From: Matti Aarnio To: david@cs.dal.ca (David Trueman) Date: Thu, 6 Feb 1997 21:05:03 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "David Trueman" at Feb 6, 97 02:54:56 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1378 Message-Id: <97Feb6.210504+0200_eet.55277-31139+266@nic.funet.fi> Return-Path: > We are trying to get 2.99.45 running on Linux system (debian 1.1 with > kernel 2.0.28), all previous zmailer experience being with the UofT > version. > > The router almost immediately dumps core on the following message: Yes, I have seen that symptom too -- and fixed it. ... > The traceback from gdb follows: > > #0 find_errto (info=0x0) at rfc822.c:2171 > 2171 if (STRING(info)) > (gdb) > (gdb) bt > #0 find_errto (info=0x0) at rfc822.c:2171 > #1 0x800ec42 in sequencer (e=0x804df70, file=0x80653d4 "481324-10257") > at rfc822.c:1680 > #2 0x800b666 in run_rfc822 (argc=2, argv=0x80653e4) at rfc822.c:158 ... > (gdb) up > #1 0x800ec42 in sequencer (e=0x804df70, file=0x80653d4 "481324-10257") > at rfc822.c:1680 > 1680 errto = find_errto(gg); /* Possible > 'ERR' valu > e */ > > I would appreciate any hints on how to go about tracking down the problem, > having never debugged a newer zmailer (and not having done a UofT one > in a long time). ftp://ftp.funet.fi/pub/unix/mail/zmailer/zmailer-2.99.46.tar.gz Stored into archive 2 minutes ago :-) (And no, my responce time is NOT fast, it just happens sometimes ;-) ) > Thanks in advance, > > David Trueman, > Systems Manager, Dalhousie Math, Stats and Computing Science > Technical Chair, Chebucto Community Net /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Feb 6 21:17:49 1997 Received: by nic.funet.fi id <55455-16413>; Thu, 6 Feb 1997 21:15:39 +0200 Subject: released 2.99.46 From: Matti Aarnio To: zmailer@nic.funet.fi Date: Thu, 6 Feb 1997 21:15:26 +0200 (GMT) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 8174 Message-Id: <97Feb6.211539+0200_eet.55455-16413+258@nic.funet.fi> Return-Path: Hello, I just released version 2.99.46 with a lot of things, many half-cooked, several bugfixes, and some extremely weirdo things :) Lets see... - Transport-agents do log "final diagnostics" into the transport specification file by themselves ("ok2", and "error2" cases), thus releaving a bit of the load of the scheduler - The scheduler does not anymore pick whole transport- workfile into its core -- into a malloced buffer. This should shrink dramatically the size of the scheduler, when it has huge queues inside itself. - FQDNaliases -- this@user: that@address (this for my employer ;-) ) - Easier support for handling white-space in the local-part of addresses -- and doing proper quotations while sending things out to the transporter. - White-spaces in the alias key-side: "test space": somebody-else Keep sending me email/questions, I do try to respond, though I am damn busy doing some other things.. /Matti Aarnio -------- ChangeLog ----------- Thu Feb 6 19:00:24 1997 Matti Aarnio * Makefile.in: patchlevel 46 -- time for a release.. * router/functions.c, proto/aliases.cf, proto/crossbar.cf, proto/canon.cf: New router-zmsh routines: $(dequote .. ) $(condquote .. ) To provide support for having an alias entry like this: "The Space Head": st-tng Wed Feb 5 10:02:20 1997 Caldera Desktop User * utils/makendbm/makendbm.c: Runtime option (switch) -a to process alias-like input (keys are RFC-821-quoted strings) that follows format: keyname: LWSP value,value2,value3 Thus allowing feeding the output of printaliases routine to makendbm for creating some real dbase, instead of indirectly indexed ordered flat ascii file.. * libsh/sh.ssl: A new way to code white-space into '*sift' patterns by using paired quotes: 'text with " dbl quote'"' and single quote" Strings: -111111111111111111111--222222222222222222- There '-' are ignored, and chars of '1', and '2' are taken in. * proto/cf/aliases.cf, proto/cf/crossbar.cf: Recognize (with special pattern) that the local address contains spaces, and do add quotes to it! (Similarly for 'internet' rewrite rule..) * INSTALL, proto/cf/fqdnalias.cf, proto/newfqdnaliases.in, proto/db/fqdnaliases: A new routeing mechanism: fqdnalias * Overview: Added one new performance benchmark result * README.UPGRADING, scheduler/update.c, transports/libta/ctlopen.c, transport/libta/diagnostic.c: Due to a change at transports/libta/diagnostic.c, and its counterpart within scheduler/update.c, new transport agents might not work with old scheduler -- it is a matter of linking the TA with the new library. * compat/sendmail/sendmail.c, smtpserver/smtpserver.c: Do proper RFC821->RFC822 syntax quoting of possible backslashfull addresses: \rfc821@domain --> "\rfc822"@domain * config.h.in, include/syslog.h, libc/syslog.c HAVE_SYSLOG_H -- mark also if (or if not) we have in the system. * libident/identuser.c, include/identuser.h: A thorough rewamp of things -- to make sure no buffer overflows will occur due to a hostile data.. * include/libz.h, lib/rfc822scan.c, router/functions.c, router/rfc822.c, router/rfc822hdrs.c: When doing address list expansions, process each LINE of the input as a sole container of the address -- if there is some address like: "@foo.domain" in a line, previously this became catenated with the next line as: "@foo.domain:faa@fii" * include/ta.h, transports/libta/tasyslog.c: New generic tasyslog() to log out transport-agent report in similar manner to that of the sendmail(8). * proto/Makefile.in: Build/don't build some new/old dirs. Especially builds the hash-subdirs for the scheduler! * lib/Makefile.in, lib/taspoolid.c: New generic tasyslog() support routine to build unique spoolids by catenating a base-64 encoded mtime of the email contents file, and the inode number of that file. * lib/prversion.c: Year change -- Copyright 1992-1997 * proto/cf/aliases.cf: Handle all variants of list owner-, -owner, and -request address injections neatly. Does not set list outbound sender to be owner-listname, though! (See aliases-vger.cf) * proto/cf/rrouter.cf: - recognize addresses that start with a pipe character, and DON'T mess with them with the focus testings! - Handle backquote-prefix properly (i.e. strip it..) - Detect feed to files (or X.400 address..) - If the `fullnames' database, nor `newsgroup' database found anything, try handling it in aliases database. * router/libdb/headers.c: Moved two headers into "unknown" category, where we won't be checking on them: keywords, references * router/libdb/ordered.c, router/libdb/unordered.c: scan over the first token with skip821address() -- now I can do aliases with quotes on the left-side, and spaces in them! * router/rfc822.c: Bugfix: wrong pointer picked for recipient attribute causing more or less mysterious crash.. * router/rfc822hdrs.c: New form of the generated Message-Id: It is slightly shorter than the previous one, and definitely a lot more sortable.. * scheduler/msgerror.c: - Slight improvement on handling spacefull 'host' object (and 'user' object too..) - Threw away a 'dnsrecipient' information pointer, and updated error log writers/parsers for it. * scheduler/scheduler.c, scheduler/scheduler.h, scheduler/update.c: Will not keep 'contents' of the TA-spec file in memory for very long -- only up to the initial scheduling input of the message. * smtpserver/relaytest.h smtpserver/relaytest.c: Some more code -- "flesh on the bones", but not yet usable.. * smtpserver/smtpserver.c: Tried to alter the code to carry around a 'content' package so that perhaps someday the smtp-server can be threaded (or some such), and each thread runs with its own content. * transports/libta/diagnostic.c, transports/mailbox/mailbox.c, transports/smtp/smtp.c: Doing 'sendmail(8)' like syslog data preparation Sat Jan 25 15:18:08 1997 Matti Aarnio * libsh/interpret.c, libsh/tregexp.c: Trace SSIFT/TSIFT statements with clearly differentiated "trace compare/trace match" statements -- namely report "tcomparing 'pattern' and 'string'" for the TSIFT, and "scomparing 'pattern' and 'string'" for the SSIFT (and "smatched"/"tmatched" for matches likewise..) * proto/cf/rrouter.cf: Where that trace change was motivated... was something that I had implemented sloppily in a ssift statement... Namely I noticed, that the "fullname expansion traps ':include:/path/to.file' into itself, and fails". Then I had decided to implement a test for RE "^[^.:]+\.[^.]+$" (where the starting/ending limiters are implicite in ssift!) but that RE has a fault of matching only ONE dot -- so no "foo.bar.bat" is matched with it, while it correctly does not match cases where the part before that dot has a colon (from ":include:"). Now I test at first for explicite case of ":include:", and only then test for dotfull names. I am still uncertain if we should not rewrite it entirely differently -- to move also fullname/newsgroup map tests inside the $(routeruser ..) routine, and thus be able to allow sendmail -like aliases where dotfull names can be in the standard aliases as well. * smtpserver/smtpserver.c, proto/smtpserver.conf, smtpserver/Makefile.in, smtpserver/relaytest.h, smtpserver/relaytest.c: Some more coding of multi-db-format query lookups of service limitation mechanisms. * utils/makendbm/ndbmlook.c: Yet another parametrization error in the DB code.. (A non-fatal one, I presume.. affected only GDBM) From zmailer-owner@nic.funet.fi Fri Feb 7 10:04:40 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <55594-20417>; Fri, 7 Feb 1997 10:04:11 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86809-11611>; Fri, 7 Feb 1997 03:03:52 -0500 Date: Fri, 7 Feb 1997 03:03:46 -0500 (EST) From: Andy Poling To: zmailer@nic.funet.fi Subject: Multiple Routers Processing A Single Message In-Reply-To: <97Feb6.211539+0200_eet.55455-16413+258@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: We're running version 2.99.38 on IRIX 5.3 systems (using EFS filesystems) with NROUTERS set to 10. The (I think) relevant piece of code in router/functions.c hasn't changed as of 2.99.46 though. I've witnessed instances of multiple routers apparently somehow renaming the queue file and then processing it (redundantly), resulting in the the mail being delivered as multiple discrete messages. As an example, the following entries (split and wrapped for readability) from the router log show three different routers grabbing the same message and processing it independantly (right?): <97Feb6.051600-0500_est.37331-2693+78@summit.request.net>: file: 37331-2693 statistics@request.com => webmaster@bakersound.com <97Feb6.051600-0500_est.37331-2693+78@summit.request.net>: address: webmaster@bakersound.com <97Feb6.051600-0500_est.37331-2693+78@summit.request.net>: recipient: cdsmail bakersound.com webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2695+60@summit.request.net>: file: 37331-2695 statistics@request.com => webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2695+60@summit.request.net>: address: webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2688+65@summit.request.net>: file: 37331-2688 statistics@request.com => webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2688+65@summit.request.net>: address: webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2688+65@summit.request.net>: recipient: cdsmail bakersound.com webmaster@bakersound.com <97Feb6.051601-0500_est.37331-2695+60@summit.request.net>: recipient: cdsmail bakersound.com webmaster@bakersound.com I checked, and HAVE_RENAME is set to "1" in config.h . I even traced (using the IRIX utility "par") a router process and it is indeed rename()ing the files: 0mS router( 5519): END-sginap() = 0 0mS router( 5519): open(., O_RDONLY, 0) = 5 1mS router( 5519): fcntl(5, F_SETFD, 1) OK 1mS router( 5519): fstat(5, 0x7fff2ae0) OK 1mS router( 5519): getdents(5, 0x1004758c, 4096) = 4092 2mS router( 5519): stat(39208, 0x7fff2ae0) OK 2mS router( 5519): stat(39208, 0x7fff2608) OK 2mS router( 5519): rename(39208, 39208-5519) OK 35mS router( 5519): open(39208-5519, O_RDONLY, 0666) = 6 The IRIX man page for rename(2) doesn't actually say that it's an atomic operation. I'm thinking that maybe it's not. :-( Would it perhaps be appropriate to put some sort of locking around the call to eqrename() to force single-threading in that section? It seems like the proper paranoid thing to do anyhow... especially for systems that don't have rename() and must therefore make do with a link() followed by an unlink(). I'm trying this approach (using lockf() on the original file, which I know isn't portable) just to see if it works. I've attached a context diff (just cut-n-pasted, so any TABs may be destroyed) to router/functions.c in 2.99.46 below just so you can see what I'm talking about. -Andy ============================================================================== *** functions.c.orig Thu Feb 6 12:18:00 1997 --- functions.c Fri Feb 7 02:17:41 1997 *************** *** 596,601 **** --- 596,602 ---- char *sh_memlevel = getlevel(MEM_SHCMD); int thatpid; struct stat stbuf; + int lockfd; *pathbuf = 0; if (*dirs) { /* If it is in alternate dir, move to primary one, *************** *** 656,664 **** --- 657,680 ---- sprintf(buf, "%ld-%d", (long)stbuf.st_ino, router_id); + /* + * we lock it before the rename()... late-comers will be blocked on + * the lock until it's too late and the file is long gone + */ + if ((lockfd = open(pathbuf, O_RDWR, 0)) < 0 + || lockf(lockfd, F_LOCK, 0)) + { + close(lockfd); + return 0; + } + if (eqrename(pathbuf, buf, 1) < 0) + { + close(lockfd); return 0; /* something is wrong, erename() complains. (some other process picked it ?) */ + } + close(lockfd); filename = buf; /* message file is now "file-#" and belongs to this process */ } From zmailer-owner@nic.funet.fi Fri Feb 7 10:18:26 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <55238-20417>; Fri, 7 Feb 1997 10:18:03 +0200 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id JM24142; Fri, 7 Feb 1997 09:17:48 +0100 Date: Fri, 7 Feb 1997 09:17:48 +0100 From: Gabor Kiss Message-Id: <199702070817.JM24142@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: Re: 2.99.45 router core dump X-Charset: US X-Char-Esc: 0 Return-Path: > The router almost immediately dumps core on the following message: > > ------------------------------------------------------------------------- > to csuite-tech-approval@csuite.chebucto.ns.ca > from Majordomo-Owner@csuite.chebucto.ns.ca > env-end > To: csuite-tech-approval@csuite.chebucto.ns.ca ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This is an alias, isn't it? I've experienced the same thing. Some aliases make abort the router. Gabor > From: majordomo@csuite.chebucto.ns.ca > Subject: APPROVE csuite-tech > Reply-To: majordomo@csuite.chebucto.ns.ca From zmailer-owner@nic.funet.fi Fri Feb 7 18:45:40 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <55652-9744>; Fri, 7 Feb 1997 18:44:21 +0200 Received: by utu.fi id <31217-732>; Fri, 7 Feb 1997 18:43:26 +0200 Subject: Re: 2.99.45 router core dump From: Matti Aarnio To: kissg@zeronull.aszi.sztaki.hu (Gabor Kiss) Date: Fri, 7 Feb 1997 18:43:16 +0200 (EET) Cc: zmailer@nic.funet.fi In-Reply-To: <199702070817.JM24142@zeronull.aszi.sztaki.hu> from "Gabor Kiss" at Feb 7, 97 09:17:48 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Length: 648 Message-Id: <97Feb7.184326+0200_eet.31217-732+2308@utu.fi> Return-Path: > > The router almost immediately dumps core on the following message: > > > > ------------------------------------------------------------------------- > > to csuite-tech-approval@csuite.chebucto.ns.ca > > from Majordomo-Owner@csuite.chebucto.ns.ca > > env-end > > To: csuite-tech-approval@csuite.chebucto.ns.ca > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > This is an alias, isn't it? > I've experienced the same thing. Some aliases make abort the router. Yes, that was another bug - and a third one, also fixed in the 2.99.46. But I just found a few bugs at the revised smtpserver -- except patch1 tonight. > Gabor /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 7 21:28:13 1997 Received: by nic.funet.fi id <55805-25725>; Fri, 7 Feb 1997 21:26:47 +0200 Subject: Re: Multiple Routers Processing A Single Message From: Matti Aarnio To: andy@realbig.com (Andy Poling) Date: Fri, 7 Feb 1997 21:26:43 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Andy Poling" at Feb 7, 97 03:03:46 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1924 Message-Id: <97Feb7.212647+0200_eet.55805-25725+263@nic.funet.fi> Return-Path: > We're running version 2.99.38 on IRIX 5.3 systems (using EFS filesystems) > with NROUTERS set to 10. The (I think) relevant piece of code in > router/functions.c hasn't changed as of 2.99.46 though. It hasn't. I am more and more convinced that it should revert back to the old link()/unlink() method, which behaviour apparently can be made completely deterministic, and most definitely it will not behave in this SysVr4 manner that we so much see... Your locking code is an interesting idea, though :-) > I've witnessed instances of multiple routers apparently somehow renaming the > queue file and then processing it (redundantly), resulting in the the mail > being delivered as multiple discrete messages. > > As an example, the following entries (split and wrapped for readability) > from the router log show three different routers grabbing the same message > and processing it independantly (right?): ... > The IRIX man page for rename(2) doesn't actually say that it's an atomic > operation. I'm thinking that maybe it's not. :-( Yes, if it yields an error, the target name may already exist, but the origin has not yet been removed... Umm.. EBUSY error it was ? (At least with Solaris) > Would it perhaps be appropriate to put some sort of locking around the call > to eqrename() to force single-threading in that section? It seems like the > proper paranoid thing to do anyhow... especially for systems that don't have > rename() and must therefore make do with a link() followed by an unlink(). Perhaps not -- those are safe from this rename() fallacy :-( > I'm trying this approach (using lockf() on the original file, which I know > isn't portable) just to see if it works. I've attached a context diff (just > cut-n-pasted, so any TABs may be destroyed) to router/functions.c in 2.99.46 > below just so you can see what I'm talking about. > > -Andy /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 7 21:33:27 1997 Received: by nic.funet.fi id <56134-20417>; Fri, 7 Feb 1997 21:32:11 +0200 Subject: Re: released 2.99.46 From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Fri, 7 Feb 1997 21:32:07 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at Feb 6, 97 02:16:43 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2017 Message-Id: <97Feb7.213211+0200_eet.56134-20417+218@nic.funet.fi> Return-Path: > On Thu, 6 Feb 1997, Matti Aarnio wrote: > > I just released version 2.99.46 with a lot of things > > I just picked up your release, but it would not compile on either a Debian > Linux 1.2 system or a Sun Solaris 2.5 system. Both are using GCC > 2.7.2. The error was: smtpserver/relaytest.h: It has one #ifdef HAVE_DB_H, which should be #ifdef HAVE_DBM_H > gcc -O2 -DHAVE_CONFIG_H -I.. -I../include -c relaytest.c > relaytest.c: In function `reldefine': > relaytest.c:37: union has no member named `_db' > *** Error code 1 > make: Fatal error: Command failed for target `relaytest.o' > Current working directory .../zmailer-2.99.46/smtpserver > *** Error code 1 > make: Fatal error: Command failed for target `all' > > Also, while I'm at it, I ran into a minor annoyance in the compilation for > the Linux system which is I get the following error: > > cc -O2 -DHAVE_CONFIG_H -I../.. -I../../include -I../../libsh -c bind.c > dbm.c gdbm.c header.c hostsfile.c incore.c ndbm.c ordered.c unordered.c > yp.c bsdbtree.c > dbm.c:31: conflicting types for `dbmclose' > /usr/include/dbm.h:52: previous declaration of `dbmclose' Ok, that I can't repeat myself -- don't have dbm.h in my systems. > make[3]: *** [../../libs/libzmdb.a] Error 1 > make[3]: Leaving directory `./zmailer-2.99.46/router/libdb' > make[2]: *** [libzmdb.a] Error 2 > make[2]: Leaving directory `./zmailer-2.99.46/libs' > make[1]: *** [../libs/libtag] Error 2 > make[1]: Leaving directory `./zmailer-2.99.46/smtpserver' > make: *** [all] Error 1 > > I can fix this one by commenting out the line in the 'config.h.' file > which says: > > #define HAVE_DBM_H 1 > > This was also present in the 2.99.45 release. I suspect it may be I get > this because I have both NDBM and GDBM installed on the system. And either of them has their own DBM compability package also installed into the system, don't they ? > Thanks, > Roy Bixler > The University of Chicago Press > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 7 22:05:32 1997 Received: by nic.funet.fi id <55947-12279>; Fri, 7 Feb 1997 22:03:26 +0200 Subject: Re: released 2.99.46 From: Matti Aarnio To: bruce@cs.ualberta.ca (Bruce Wm Folliott) Date: Fri, 7 Feb 1997 22:03:20 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <97Feb6.155455-0700_mst.13062-23535+158@scapa.cs.ualberta.ca> from "Bruce Wm Folliott" at Feb 6, 97 03:54:49 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2864 Message-Id: <97Feb7.220326+0200_eet.55947-12279+202@nic.funet.fi> Return-Path: I am copying privately received email to the list -- a breach of etiquette, but I think you all will benefit from these replies -- and I don't need to repeat them all... /Matti Aarnio > > Keep sending me email/questions, I do try to respond, though > > I am damn busy doing some other things.. > > zmailer-2.99.46 > Reading specs from /usr/gnu/lib/gcc-lib/sparc-sun-solaris2.5/2.7.2/specs > gcc version 2.7.2 > SunOS sedgewick 5.5 Generic_103093-06 sun4u sparc SUNW,Ultra-1 > > gcc -I../include -I../libresolv -g -O2 -DUSELOOPBACK -c gethnamaddr.c > In file included from gethnamaddr.c:73: > ../include/syslog.h:153: parse error before `__' > ../include/syslog.h:154: parse error before `__' > ../include/syslog.h:158: parse error before `__' ?:-O -- no idea... Unless there are some macros defining openlog, syslog, and closelog.. Umm.. can't find ? > gcc -g -O2 -DHAVE_CONFIG_H -I.. -I../include -I../libresolv -c > relaytest.c > relaytest.c: In function `reldefine': > relaytest.c:37: union has no member named `_db' One missing character at reltest.h > gcc -g -O2 -DHAVE_CONFIG_H -I.. -I../include -I../libsh -I../libresolv -c rfc822.c > rfc822.c: In function `mkTrace': > rfc822.c:947: warning: assignment discards `const' from pointer target type I can silence it, but don't like it. Usually that pointer is not used to refer on anything, but in a couple of cases it refers to object that needs a change runtime.. Perhaps a cast could do it ? I will try. > gcc -g -O2 -DHAVE_CONFIG_H -I../.. -I../../include -c sendmail.c > sendmail.c: In function `check_and_print_to': > sendmail.c:695: warning: initialization makes integer from pointer > without a cast > sendmail.c:714: warning: assignment makes integer from pointer without a cast > sendmail.c:714: warning: assignment makes pointer from integer without a cast > sendmail.c:714: warning: assignment makes pointer from integer without a cast > sendmail.c:732: warning: passing arg 1 of `free' makes pointer from integer without a cast Oops. 'char newcp' which should have been 'char *newcp' ... > Got a few of these: > gcc -g -O2 -DHAVE_CONFIG_H -I../.. -I../../include -o hold hold.o > version.o -L../../libs -lta -lz -lzc -lmalloc -lresolv -lsocket > Undefined first referenced > symbol in file > gethostbyname ../../libs/libzc.a(myhostname.o) > (symbol belongs to implicit dependency /usr/lib/libnsl.so.1) > ld: fatal: Symbol referencing errors. No output written to hold Yes, a result from trying to accomodate another case of using Solaris, and having non-standard libresolver in it. I pulled back into previous style, which does not support autodetection of non-standard libresolv. > Got similiar things on SunOS 4.1.4 Except this last one -- I hope.. > Bruce /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 7 22:18:44 1997 Received: from butler.cc.tut.fi ([130.230.102.10]) by nic.funet.fi with ESMTP id <55810-25725>; Fri, 7 Feb 1997 22:16:11 +0200 Received: (from mjs@localhost) by butler.cc.tut.fi (8.8.4/8.8.4) id WAA03648; Fri, 7 Feb 1997 22:16:04 +0200 (EET) From: Matti Saarinen Message-Id: <199702072016.WAA03648@butler.cc.tut.fi> Subject: Re: released 2.99.46 To: mea@nic.funet.fi (Matti Aarnio) Date: Fri, 7 Feb 1997 22:16:04 +0200 (EET) Cc: zmailer@nic.funet.fi In-Reply-To: <97Feb7.213211+0200_eet.56134-20417+218@nic.funet.fi> from "Matti Aarnio" at Feb 7, 97 09:32:07 pm X-Mailer: ELM [version 2.4 PL25] Content-Type: text Return-Path: > > smtpserver/relaytest.h: It has one #ifdef HAVE_DB_H, which > should be #ifdef HAVE_DBM_H > I had the same problem and after I made the change suggested above the error messages changed (The operating system is Digital UNIX 3.2D-1): gmake[2]: Leaving directory `/usr/local/src/mail/zmailer/zmailer-2.99.46/libs' cc -migrate -O -g3 -std1 -I/usr/local/include -L/usr/local/lib -DHAVE_CONFIG_H -I.. -I../include -I../libresolv -Iyes -c relaytest.c -o relaytest.o cc: Error: relaytest.h, line 25: Missing type specifier or type qualifier. DB *_db; ----^ gmake[1]: *** [relaytest.o] Error 1 gmake[1]: Leaving directory `/usr/local/src/mail/zmailer/zmailer-2.99.46/smtpserver' gmake: *** [all] Error 1 I cannot solve that problem so I hope that someone can. -- - Matti - From zmailer-owner@nic.funet.fi Fri Feb 7 22:28:54 1997 Received: by nic.funet.fi id <56102-25725>; Fri, 7 Feb 1997 22:27:25 +0200 Subject: Re: released 2.99.46 From: Matti Aarnio To: mjs@butler.cc.tut.fi (Matti Saarinen) Date: Fri, 7 Feb 1997 22:27:18 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199702072016.WAA03648@butler.cc.tut.fi> from "Matti Saarinen" at Feb 7, 97 10:16:04 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 636 Message-Id: <97Feb7.222725+0200_eet.56102-25725+275@nic.funet.fi> Return-Path: > > smtpserver/relaytest.h: It has one #ifdef HAVE_DB_H, which > > should be #ifdef HAVE_DBM_H > > > > I had the same problem and after I made the change suggested above the > error messages changed (The operating system is Digital UNIX 3.2D-1): Yes, you too don't have BSD DB in your system. (and I do have a think skull at the times...) It is a 4 line patch - one away, and 3 in. I will this time try to compile these on DEC UNIX 4.0, and also on Solaris 2.5.1. Then it should work far better than with Linux alone.. > I cannot solve that problem so I hope that someone can. > > -- > - Matti - /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Feb 8 01:18:08 1997 Received: by nic.funet.fi id <55553-15834>; Sat, 8 Feb 1997 01:16:54 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: zmailer-2.99.46.patch1 Message-Id: <19970207231654Z55553-15834+3@nic.funet.fi> Date: Sat, 8 Feb 1997 01:16:54 +0200 Return-Path: With that thing I get it to compile, and run at my Linuxes (RedHat 4.0+, and RedHat/Caldera at home), DEC UNIX 4.0, and Sun Solaris 2.5.1 with Sun's cc. Now things will be much better, and I will hear less bug reports ? ... even I don't believe that last one :-) Tell me your experiences, I will perhaps be around during the weekend, and if not, next week then among running between project meetings.. (being a Project Manager is never easy for a UNIX Hacker ;) ) /Matti Aarnio -- heading for home for tonight From zmailer-owner@nic.funet.fi Mon Feb 10 14:56:50 1997 Received: from uxmail.ust.hk ([143.89.14.30]) by nic.funet.fi with ESMTP id <56865-10765>; Mon, 10 Feb 1997 14:55:28 +0200 Received: from ccsu51.ust.hk ([143.89.103.20]) by uxmail.ust.hk with SMTP id <102247-29351>; Sat, 8 Feb 1997 13:25:48 +0800 Received: by ccsu51.ust.hk (SMI-8.6/SMI-SVR4) id NAA21089; Sat, 8 Feb 1997 13:25:39 +0800 From: ccyflai@uxmail.ust.hk (Lai Yiu Fai) Message-Id: <199702080525.NAA21089@ccsu51.ust.hk> Subject: Re: Multiple Routers Processing A Single Message To: mea@nic.funet.fi (Matti Aarnio) Date: Sat, 8 Feb 1997 13:25:39 +0800 (HKT) Cc: andy@realbig.com, zmailer@nic.funet.fi In-Reply-To: <97Feb7.212647+0200_eet.55805-25725+263@nic.funet.fi> from "Matti Aarnio" at Feb 7, 97 09:26:43 pm X-Mailer: ELM [version 2.4 PL24alpha3] Content-Type: text Return-Path: > > > We're running version 2.99.38 on IRIX 5.3 systems (using EFS filesystems) > > with NROUTERS set to 10. The (I think) relevant piece of code in > > router/functions.c hasn't changed as of 2.99.46 though. Problem not only found on IRIX but also on Solaris 2.5. I'm running 2.99.38 also and limit NROUTERS to 1 for temporary fix. > > It hasn't. I am more and more convinced that it should > revert back to the old link()/unlink() method, which > behaviour apparently can be made completely deterministic, > and most definitely it will not behave in this SysVr4 > manner that we so much see... I've tried to uncomment HAVE_RENAME and use old link/unlink for locking in 2.99.38 but seems the problem still presists. Besides this, any fix in 2.99.46p1 to address the problem? > > Your locking code is an interesting idea, though :-) > I'll try Andy's locking code and just to see whether it helps in Solaris. ======================================================================= Lai Yiu Fai | Tel.: (852) 2358-6202 Centre of Computing Services | Fax.: (852) 2358-0967 & Telecommunications | E-mail: ccyflai@uxmail.ust.hk | The Hong Kong University of | Clear Water Bay, Science & Technology | Kowloon, Hong Kong. From zmailer-owner@nic.funet.fi Mon Feb 10 15:12:09 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <56420-12978>; Mon, 10 Feb 1997 15:11:00 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <87146-20372>; Sat, 8 Feb 1997 02:06:47 -0500 Date: Sat, 8 Feb 1997 02:06:41 -0500 (EST) From: Andy Poling To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: Multiple Routers Processing A Single Message In-Reply-To: <97Feb7.212647+0200_eet.55805-25725+263@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 7 Feb 1997, Matti Aarnio wrote: > > We're running version 2.99.38 on IRIX 5.3 systems (using EFS filesystems) > > with NROUTERS set to 10. The (I think) relevant piece of code in > > router/functions.c hasn't changed as of 2.99.46 though. > > It hasn't. I am more and more convinced that it should > revert back to the old link()/unlink() method, which > behaviour apparently can be made completely deterministic, > and most definitely it will not behave in this SysVr4 > manner that we so much see... Just to get a sense of history, I retrieved and looked at the version 2.2.1 router code and it's essentially identical, except that it renames files to filename(router_number). > > The IRIX man page for rename(2) doesn't actually say that it's an atomic > > operation. I'm thinking that maybe it's not. :-( > > Yes, if it yields an error, the target name may already exist, > but the origin has not yet been removed... > Umm.. EBUSY error it was ? (At least with Solaris) [...] > > Would it perhaps be appropriate to put some sort of locking around the call > > to eqrename() to force single-threading in that section? It seems like the > > proper paranoid thing to do anyhow... especially for systems that don't have > > rename() and must therefore make do with a link() followed by an unlink(). > > Perhaps not -- those are safe from this rename() fallacy :-( Just in case you didn't catch it (in retrospect I wan't very explicit), the problem is that while one router process is trying to rename() (or link/unlink) a file, one or more other router processes can also rename() or link() it - yielding more than one router process that each think they have an exclusive copy. I looked at the alternative link()/unlink() code and it is even more vulnerable to the same race condition. Between the link() and unlink(), another router process could also perform a link() (to a different filename). I just cannot see any way to avoid having to use some locking method to avoid rename()/link() races between competing router processes. As I see it, you have a choice of forcibly single-threading the entire directory scanning process (sub-optimal) or forcibly single-threading just the little section of code that renames the file. If you don't like the idea of locking the msg file itself, then perhaps some "central" synchronization method could be used. E.g. have the router processes lock a single (always present) file in that (or some other) directory. I'm trying to consider portability here... a semaphore would probably be the best solution, but who wants to deal with portably using semaphores... > > I'm trying this approach (using lockf() on the original file, which I know > > isn't portable) just to see if it works. FYI, I have received no further complaints of redundant messages since adding the locking code, but I won't feel safe for a couple more days. -Andy From zmailer-owner@nic.funet.fi Mon Feb 10 21:36:06 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <74461-20119>; Mon, 10 Feb 1997 21:23:58 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <87025-31406>; Mon, 10 Feb 1997 14:23:43 -0500 Date: Mon, 10 Feb 1997 14:23:42 -0500 (EST) From: Andy Poling To: Eugene Crosser cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: Multiple Routers Processing A Single Message In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 10 Feb 1997, Eugene Crosser wrote: > > Just in case you didn't catch it (in retrospect I wan't very explicit), the > > problem is that while one router process is trying to rename() (or > > link/unlink) a file, one or more other router processes can also rename() > > or link() it - yielding more than one router process that each think they > > have an exclusive copy. > > Sorry guys, maybe I am missing something, but I always thought that link/ > unlink/rename are all `atomic' on any normal UNIX. So, if you renamed a > file to something private, no one would ever be able to rename it too. > If you are using link/unlink, you must check the result of unlink as well > as the result of link to be safe, if unlink returned ENOENT that means that > someone else has grabbed the file already and you should just drop your copy. Some systems apparently do not have a rename() system call, or it may not actually be atomic (as is apparently the case with IRIX 5.3) . You're right about checking the unlink(). I hadn't thought about being able to use unlink()'s atomicity to prevent a race. Is unlink() *always* atomic though? I heard some time ago that link() is not properly atomic in some NFS cases (specifically, a VMS NFS server was the one I heard about). -Andy From zmailer-owner@nic.funet.fi Tue Feb 11 01:24:33 1997 Received: by nic.funet.fi id <74362-20876>; Tue, 11 Feb 1997 01:17:31 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: Infinite recursion in 2.99.46(p1) proto/cf/aliases.cf Message-Id: <19970210231731Z74362-20876+41@nic.funet.fi> Date: Mon, 10 Feb 1997 18:02:05 +0200 Return-Path: You may have noticed strange deaths of the routers with the lattest release. I certainly did -- and spent hours tracking for the reason, which was obvious -- once I realized what I was looking at... [mea@mea zmailer-2.99.46p1]$ diff -u proto/cf/aliases.cf{~,} --- proto/cf/aliases.cf~ Thu Feb 6 19:25:45 1997 +++ proto/cf/aliases.cf Mon Feb 10 17:41:37 1997 @@ -85,8 +85,11 @@ a=$quad #mxh=$(setf $(last $a) ()) - #key="$(elements $quad)" - key="$(channel $quad)$(user $quad)$(attributes $quad)" + # For .forward (et.al.) the ATTRIBUTE may change, therefore + # we must not use it to the key -- in case we do a forwarding... + # That said, I don't think we need 'host'-part either.. + #key="$(channel $quad)$(host $user)$(user $quad)" + key="$(channel $quad)$(user $quad)" didexpand=$(expansions "$key") al=() This is cut&paste, so it might not go in peacefully, but you should get a pretty good idea about how it is fixed... /Matti Aarnio PS: Queue at nic.funet.fi router: 3200, and decreasing :-) From zmailer-owner@nic.funet.fi Tue Feb 11 01:47:00 1997 Received: from sid.microtec.net ([204.50.80.60]) by nic.funet.fi with ESMTP id <55948-7337>; Tue, 11 Feb 1997 01:42:51 +0200 Received: from localhost (asid@localhost [127.0.0.1]) by sid.microtec.net (8.8.5/8.8.5) with SMTP id NAA15070; Mon, 10 Feb 1997 13:08:38 -0500 Date: Mon, 10 Feb 1997 13:08:38 -0500 (EST) From: Alexandre Sidorenko Reply-To: Alexandre Sidorenko To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Router serious bug Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="852245584-783715493-855598118=:15040" Return-Path: This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --852245584-783715493-855598118=:15040 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I am sending this message for the second time - it seems like first time it has lost somewhere, I have not seen a copy of it in the mailing-list. This problem still exists for the newest (2.99.46p1) version of Zmailer ! On Sat, 18 Jan 1997, Matti Aarnio wrote: > > Hi, > > > > we're running zmailer-2.99.44p1 on Linux, kernel 2.0.27 libc-5.4.17. All > > the components of Zmailer are quite stable save router. Namely, when I > > restart the router (-dkn 4) with > > > > zmailer router > > > > sometimes the 4 router processes just die in 10-30 seconds, delivering > > some mail and without any nasty diagnostics. After I issue 'zmailer > > router' several times (sometimes it's more than 10) one of the router > > processes turns to zombie, but the remaining 3 become quite stable and > > after that Zmailer runs without any problems for many days (2.99.43 > > demonstarted the same behaviour, but with 1 zombie router it was > > successfully working in heavy traffic conditions for more that 2 weeks). > > PROPABLY this is the svstack[] limit thing that Olivier did > notice too. > > I still wonder, why you don't get cores ? > ... Hello again, I have found that this bug can be reproduced (on Linux at least) by injecting the attached 2 files into the ~/rooter spooling directory. Obviously addresses in these messages are malformed, but it should not crash the router all the same... The first file (100000) makes router die. It was in the router directory since beginning of Dec.96 and was causing this 'zombie' problem on 2.99.43, 44 and now 45 patchlevels of Zmailer. The second file (100001) makes router eat memory till machine crashes ! Files are MIME-attached as routerbug.tar.gz Would you please look at that ? Thanks in advance, ------------------------------------------------------------- Alexandre Sidorenko Senior System Administrator asid@microtec.net Internet Microtec, Montreal, QC ------------------------------------------------------------- --852245584-783715493-855598118=:15040 Content-Type: APPLICATION/octet-stream; name="routerbug.tar.gz" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Files to crash router H4sIAOWY5jIAA+2VXVPbOBSGudavOJteALu1azsfTkzCR4PbZiYBNsm0ndn2 QrFPiIotuZJMNv31K4fsBCiUK7bTHT0X9owk67yy5Ue+V7HzrPie1/I82IGK +3cAP6iHAK0g9BvNRqPpm6aw5dV34Jlj3VAqTSXAjhRC/2jcU/2/KPi3Rslp RmRync6lyOE1Td138fDcHfCi1LD3V+A13KbnhqHrNzuf98mS6QVMRtMLwlLk mvG5gLNzZ3Aan02dSTx+P+jHx4UQWcfxm65IV0ohuhzNXIWQGoJ622/sk3Wx brIoWXFFj28NOyRapIrD+bh/Me3JedIOgoNat2aaoWvuhwT5tYM8JSNUil6i KR1Btx6cNDzf67hhPfTuzndKNUYwLfEleHU4xQT8TqcFvhcFnagewB/VP0De mEARbAI9kuyjM6IsQxnBSHxjWUah7npvRZbC3ogmjGuhFgcwOICLi/4+GQ1G sfMepWKCR+C7HpmKCGqfaodvJXJmOmqbFU3K2RdMdARYzQ9vfvv9BThlQfqC a/OOnemqMCvQ5mu9KjLK+IHJSaVC3SuVQ1XC2HaopFzNUToxT0TK+GUE4Yxp QlI0G30usitC2BxWooQFvcabiphCqYBx0AuEjCoNS8QroDyFpeApSjNguViZ 1vVDfFeTpWTaFIQZTa5eAtO7CmaY0FKhGbUrzVWi0pkJAOsdI0q5WZ2JDWXh AnxAs5myzDy3ngXMB65iXaNcgRKCu+Rn/x7/e/xq6/vPXuMJ/3v3/d8Mm9b/ /wXf+3943j8Znp2M4rX6m25Qb7lBI3QDv3VH/jf+TlELTvVxzhIpNCaPClwv ysojx19LNJrQmLmJyNdKf7BnK/mPzgSr/ggeqnXLyB8YT8VSQVymQlIYssuF ho19jXzNSsiI5XhXyE8LtsaUcNrtpjnNaj+S7NfSZEqdQppTgM4yXLu+L/Kc VulhLGYodSX8h9fb/TIX8rJEtV1eQg83h9KYVWlSGNKcysQo+pH3/u8x0jva hj7682jcizuF4AqPeoTAPfaKaKiN829m3ifbEROGPEEQ8zlLzGEF1EiZQyGU RlBMupOEcl75neW5qZAyqk2v0bwWpSLmEJGC9+J2YfIi55gzdDcz9wKPfJ/k dufP/i8sFovFYrFYLBaLxWKxWCwWi8VisVgsFovlV+Mfh9sEawAoAAA= --852245584-783715493-855598118=:15040-- From zmailer-owner@nic.funet.fi Tue Feb 11 02:01:12 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <55308-28788>; Tue, 11 Feb 1997 01:56:55 +0200 Received: from ariel.sovam.com ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-6811>; Mon, 10 Feb 1997 21:53:47 +0300 Date: Mon, 10 Feb 1997 21:53:35 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: Multiple Routers Processing A Single Message To: Andy Poling cc: Matti Aarnio , zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > Just in case you didn't catch it (in retrospect I wan't very explicit), the > problem is that while one router process is trying to rename() (or > link/unlink) a file, one or more other router processes can also rename() > or link() it - yielding more than one router process that each think they > have an exclusive copy. Sorry guys, maybe I am missing something, but I always thought that link/ unlink/rename are all `atomic' on any normal UNIX. So, if you renamed a file to something private, no one would ever be able to rename it too. If you are using link/unlink, you must check the result of unlink as well as the result of link to be safe, if unlink returned ENOENT that means that someone else has grabbed the file already and you should just drop your copy. Eugene From zmailer-owner@nic.funet.fi Tue Feb 11 02:53:50 1997 Received: from weaver-gw.netapp.com ([198.95.224.2]) by nic.funet.fi with ESMTP id <73978-7337>; Tue, 11 Feb 1997 02:46:36 +0200 Received: from netapp.com ([192.9.200.1]) by weaver.netapp.com with SMTP id <15978-3888>; Mon, 10 Feb 1997 13:27:36 -0000 Received: by netapp.com (4.1/SMI-4.1) id AA23840; Mon, 10 Feb 97 13:26:51 PST From: sterling@netapp.com (Bruce Sterling Woodcock) Message-Id: <9702102126.AA23840@netapp.com> Subject: Re: Multiple Routers Processing A Single Message To: andy@realbig.com (Andy Poling) Date: Mon, 10 Feb 1997 13:26:50 -0800 (PST) Cc: crosser@online.ru, mea@nic.funet.fi, zmailer@nic.funet.fi In-Reply-To: from "Andy Poling" at Feb 10, 97 02:23:42 pm X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 802 Return-Path: > Is unlink() *always* atomic though? I heard some time ago that link() is > not properly atomic in some NFS cases (specifically, a VMS NFS server was > the one I heard about). Andy, I think we must have read the same thread... said VMS implementation also had other problems (like updating the timestamp on the parent directory) that broke "traditional" NFS server semantics. I think the final consensus was that it was not behaving correctly and needed to be fixed... I believe they did that, although I don't know if the unlink() problem was specifically addressed. I believe unlink() is generally always atomic over NFS as well. Bruce -- Bruce Sterling Woodcock ------ Systems Administrator ][ sterling@netapp.com Network & Systems Administration - Network Appliance ][ sterling@netcom.com From zmailer-owner@nic.funet.fi Thu Feb 13 01:57:10 1997 Received: from sys4.cambridge.uk.psi.net ([154.32.106.14]) by nic.funet.fi with ESMTP id <56160-30111>; Thu, 13 Feb 1997 01:55:27 +0200 Received: from nadt.org.uk by sys4.cambridge.uk.psi.net (8.7.5/SMI-5.5-UKPSINet) id PAA10875; Wed, 12 Feb 1997 15:11:16 GMT Received: from infodev.nadt.org.uk (infodev.nadt.org.uk [194.155.224.205]) by charlie.nadt.org.uk (8.6.12/8.6.12) with SMTP id OAA04141 for ; Wed, 12 Feb 1997 14:02:41 GMT Date: Wed, 12 Feb 1997 14:02:41 GMT Posted-Date: Wed, 12 Feb 1997 14:02:41 GMT Message-Id: <199702121402.OAA04141@charlie.nadt.org.uk> X-Website: http://www.innotts.co.uk/~nadt X-Sender: robmel@wrcmail X-Mailer: Windows Eudora Light Version 1.5.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: zmailer@nic.funet.fi From: Robin Melville Subject: Configuration help please. Return-Path: Dear Zmailer folks, I am reviewing MTAs with a view to replacing the use of sendmail on our mailhub. Rather than start in at the deep end of what seems a fairly steep learning curve on Zmailer config, I wondered if any other users had a similar configuration to ourselves and could provide us with a template. We have no direct connection to the internet. All mail to unknown hosts is batched via UUCP to our ISP with the source machine name hidden (ie. robmel@dev.nadt.org.uk appears as robmel@nadt.org.uk). All incoming mail comes to our hub via SMTP (or UUCP if it's inbound from the Internet) where it is passed by SMTP to one of a pair of POP3 server hosts according to the addressee. At the moment this is done via the sendmail aliases file. eg. joebloggs: joebloggs@pop1 marybloggs: marybloggs@pop2 We do run BIND but only for local host lookup. We will also soon want to accept incoming internet mail with a different domain and pass this through via SMTP to a MS Mail server or possibly CC: Mail (Another department). All our hosts are FreeBSD 2.x. (if that makes any difference). I'd be very grateful for your help. Best Regards Robin. -------------------------------------------------------- Robin Melville, Addiction & Forensic Information Service Nottingham Alcohol & Drug Team (Extn. 49178) Vox: +44 (0)115 952 9478 Fax: +44 (0)115 952 9421 Email: robmel@nadt.org.uk WWW: http://www.innotts.co.uk/nadt/ --------------------------------------------------------- From zmailer-owner@nic.funet.fi Thu Feb 13 06:01:27 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <55600-4540>; Thu, 13 Feb 1997 05:58:34 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <87233-17847>; Wed, 12 Feb 1997 22:57:16 -0500 Date: Wed, 12 Feb 1997 22:57:11 -0500 (EST) From: Andy Poling To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: Multiple Routers Processing A Single Message (patch enclosed) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Matti- As we have discussed recently, one can enforce single-threadedness of renaming of the message files in $POSTOFFICE/router by using link() and unlink() and checking the return from unlink(). Participants seemed to be of the opinion that unlink() should always be an atomic operation, whereas rename() is apparently not always atomic. I've implemented this change after testing the theory using file locks for several days. The duplicate messages ceased. The diff below changes lib/esyslib.c to never use rename(), but to use link() and unlink() instead and check the outcome of the unlink(). If the unlink() of "from" fails, then "to" is also unlink()ed and we return -1 to indicate failure - either due to an error or because another router process link()ed and unlink()ed the file before we could. This should enforce single-threaded renaming in the routers. I submit the following diff (against v2.99.46) for your consideration. I have placed the resultant router into production use on several IRIX 5.3 systems. -Andy ============================================================================== *** esyslib.c.dist Thu Dec 12 11:35:35 1996 --- esyslib.c Wed Feb 12 22:27:36 1997 *************** *** 207,230 **** erename(from, to) char *from, *to; { ! int r; ! ! #ifdef HAVE_RENAME ! ! if ((r = rename(from, to)) < 0) { ! #else /* !HAVE_RENAME */ ! ! if (((r = unlink(to)) < 0 && errno != ENOENT) ! || (r = link(from, to)) < 0) { ! #endif /* HAVE_RENAME */ fprintf(stderr, "%s: rename(%s,%s): %s\n", progname, from, to, strerror(errno)); } ! #ifndef HAVE_RENAME ! else ! unlink(from); ! #endif /* !HAVE_RENAME */ ! return r; } int --- 207,227 ---- erename(from, to) char *from, *to; { ! if (((unlink(to)) < 0 && errno != ENOENT) ! || (link(from, to)) < 0) { fprintf(stderr, "%s: rename(%s,%s): %s\n", progname, from, to, strerror(errno)); + return(-1); + } + if (unlink(from)) + { + if (errno != ENOENT) + fprintf(stderr, "%s: unlink(%s): %s\n", progname, + to, strerror(errno)); + unlink(to); + return(-1); } ! return(0); } int *************** *** 231,250 **** eqrename(from, to) char *from, *to; { ! int r; ! ! #ifdef HAVE_RENAME ! ! if ((r = rename(from, to)) < 0) { ! #else /* !HAVE_RENAME */ ! ! if (((r = unlink(to)) < 0 && errno != ENOENT) ! || (r = link(from, to)) < 0) { ! #endif /* HAVE_RENAME */ } ! #ifndef HAVE_RENAME ! else ! unlink(from); ! #endif /* !HAVE_RENAME */ ! return r; } --- 228,241 ---- eqrename(from, to) char *from, *to; { ! if (((unlink(to)) < 0 && errno != ENOENT) ! || (link(from, to)) < 0) { ! return(-1); ! } ! if (unlink(from)) ! { ! unlink(to); ! return(-1); } ! return(0); } From zmailer-owner@nic.funet.fi Fri Feb 14 15:53:49 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <73869-1676>; Fri, 14 Feb 1997 15:52:14 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109544-21682>; Fri, 14 Feb 1997 14:52:05 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Fri, 14 Feb 1997 14:51:43 +0100 Date: Fri, 14 Feb 1997 14:51:31 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: What does this mean ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I really have trouble to get zmailer 2.99.46 going on HP-UX 9.05. What do these messages mean in the router log ? <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: address: goldt <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: file: 26678-6941 Sven Goldt => goldt@localhost <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: address: goldt@l ocalhost <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: recipient: local - goldt <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: info: uid 28113 gid 15 size 50764 headersize 34 bodysize 50730 now 855927845 delay 11 resent no trusted no The mail is stuck in the queue and will not be delivered. Greets, Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Fri Feb 14 16:25:26 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <74019-7011>; Fri, 14 Feb 1997 16:24:32 +0200 Received: by utu.fi id <31441-21891>; Fri, 14 Feb 1997 16:23:33 +0200 Subject: Re: What does this mean ? From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Fri, 14 Feb 1997 16:23:29 +0200 (EET) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Feb 14, 97 02:51:31 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Length: 1016 Message-Id: <19970214142333Z31441-21891+3417@utu.fi> Return-Path: > I really have trouble to get zmailer 2.99.46 going on HP-UX 9.05. > What do these messages mean in the router log ? > > <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: address: goldt > <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: file: 26678-6941 > Sven Goldt => goldt@localhost > <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: address: goldt@l > ocalhost > <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: recipient: local > - goldt > <19970214134405Z26678-6941+1@hprbg1.informatik.tu-muenchen.de>: info: uid 28113 > gid 15 size 50764 headersize 34 bodysize 50730 now 855927845 delay 11 resent no > trusted no > > The mail is stuck in the queue and will not be delivered. From that "recipient: local - goldt" I would guess you have configuration scripts all the way from the dawn... :-) What do you mean with 'stuck in the queue' ? Stuck on the scheduler queue, or still sitting in the router ? > Greets, Sven /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 14 16:54:49 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <56016-7011>; Fri, 14 Feb 1997 16:54:00 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109552-21686>; Fri, 14 Feb 1997 15:53:43 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Fri, 14 Feb 1997 15:53:32 +0100 Date: Fri, 14 Feb 1997 15:53:26 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: What does this mean ? In-Reply-To: <19970214142333Z31441-21891+3417@utu.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 14 Feb 1997, Matti Aarnio wrote: > From that "recipient: local - goldt" I would guess you have > configuration scripts all the way from the dawn... :-) Well, in deed these scripts should be superior. These script e.g. check the .forward file to see if you pipe to a trusted prg like 'filter' or 'procmail' and not to a selfwritten mail-multiplexer. They have been developed from 2.99.25 and adapted to 2.99.38... they also do some other things like checking domains and checking if FUZZYMATCH is wanted. If you are interested in them i can send them to you. > What do you mean with 'stuck in the queue' ? Stuck on the > scheduler queue, or still sitting in the router ? The mail is in the 'queue' directory and a transport file exists aswell. It retries to send the mail, but somehow it fails. The real problem is it is very difficult to debug the problem. Can i manually run 'mailbox' with gdb to see what it is doing ? Greets, Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Fri Feb 14 17:45:44 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <74050-14648>; Fri, 14 Feb 1997 17:44:35 +0200 Received: by utu.fi id <31447-21891>; Fri, 14 Feb 1997 17:43:49 +0200 Subject: Re: What does this mean ? From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Fri, 14 Feb 1997 17:43:46 +0200 (EET) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Feb 14, 97 03:53:26 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Length: 1342 Message-Id: <19970214154349Z31447-21891+3484@utu.fi> Return-Path: > On Fri, 14 Feb 1997, Matti Aarnio wrote: > > > From that "recipient: local - goldt" I would guess you have > > configuration scripts all the way from the dawn... :-) > > Well, in deed these scripts should be superior. These script e.g. > check the .forward file to see if you pipe to a trusted > prg like 'filter' or 'procmail' and not to a selfwritten mail-multiplexer. > They have been developed from 2.99.25 and adapted to 2.99.38... > they also do some other things like checking domains and checking > if FUZZYMATCH is wanted. > If you are interested in them i can send them to you. > > > What do you mean with 'stuck in the queue' ? Stuck on the > > scheduler queue, or still sitting in the router ? > > The mail is in the 'queue' directory and a transport file exists > aswell. It retries to send the mail, but somehow it fails. > > The real problem is it is very difficult to debug the problem. > Can i manually run 'mailbox' with gdb to see what it is doing ? Yes you can. It might need some "directory" commands at first to locate all the source files, but othervice: -> # cd $POSTOFFICE/transport/ -> # gdb $MAILBIN/ta/mailbox -> (gdb) break process -> (gdb) run ... (options you use for the local channel) <- #hungry -> 12345-678 ... > Greets, Sven > -- > The only limits are in your mind ! /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 14 19:18:00 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <56162-7011>; Fri, 14 Feb 1997 19:17:08 +0200 Received: by utu.fi id <31050-21891>; Fri, 14 Feb 1997 19:16:14 +0200 Subject: Re: Configuration help please. From: Matti Aarnio To: robmel@nadt.org.uk (Robin Melville) Date: Fri, 14 Feb 1997 19:16:03 +0200 (EET) Cc: zmailer@nic.funet.fi In-Reply-To: <199702121402.OAA04141@charlie.nadt.org.uk> from "Robin Melville" at Feb 12, 97 02:02:41 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Content-Length: 5102 Message-Id: <19970214171614Z31050-21891+3542@utu.fi> Return-Path: > Dear Zmailer folks, > > I am reviewing MTAs with a view to replacing the use of sendmail on our > mailhub. Rather than start in at the deep end of what seems a fairly steep > learning curve on Zmailer config, I wondered if any other users had a > similar configuration to ourselves and could provide us with a template. > > We have no direct connection to the internet. All mail to unknown hosts > is batched via UUCP to our ISP with the source machine name hidden (ie. > robmel@dev.nadt.org.uk appears as robmel@nadt.org.uk). I haven't specifically written a "mail hub" configuration to rewrite addresses as to hide away actual hosts -- hadn't any need myself for it.. All my setups are generally speaking mailservers which consider some set of addresses as belonging to the local machine. That is, the local-delivery spool is at the local machine, and all my systems use that same spool -- they do share it either via NFS, or by reading from it eg. via POP or IMAP. Now to come to think of it, it is likely that building a "mboxmap" can point to individual hosts for actual local delivery at them in per user basis. Following is "merely" just thought out about how to do it, without any actual systems doing it this way. (That I know of.) > All incoming mail comes to our hub via SMTP (or UUCP if it's inbound from > the Internet) where it is passed by SMTP to one of a pair of POP3 server > hosts according to the addressee. At the moment this is done via the > sendmail aliases file. > > eg. > joebloggs: joebloggs@pop1 > marybloggs: marybloggs@pop2 Unlike aliases, the mboxmap looks quite a bit different. The mboxmap database has lines like following: joebloggs pop1:home-of-joebloggs:joebloggs marybloggs pop2:home-of-marybloggs:marybloggs mea pop3:/var/spool:mea Left side: key without ':' terminating it. Rest: data. The first two lines are according to the standard approach, however you can be a bit "sneaky", and store to the local /var/spool/PObox/ directory -- yes, you need to add one symlink there into even more commonly used path ;-) ( or edit $MAILSHARE/cf/aliases.cf to change value of POBOX=PObox to be: POBOX=mail, and then things should be even simpler :-) ) Do note that the same mboxmap files should reside at each and every of the mail accepting servers (symmetric configuration there!) and the mailer's idea of $hostname should be written into the FIRST field of the result. (My workstation is mea.tmt.tele.fi, not plain "mea".) Actually each machine can have identical databases, and configration files sans ONE file. The different file is usually residing at a path of: /etc/mail.conf and it should claim unique $hostname, but common $mydomain for all machines involved in the mboxmap-cloud. That way the local-delivery is actually fully symmetric setup. At the `mail.conf' you should also define SMARTHOST to be a targer for the outbound deliveries. At all machines except the one with UUCP it should point to the gateway, and there it should point to the remote uucp target. The mail.conf files: At the relay: orgdomain=nadt.org.uk mydomain=$orgdomain hostname=relay.$orgdomain SMARTHOST=that.uucp-gw.uk # Faked name At all others: orgdomain=nadt.org.uk mydomain=$orgdomain hostname=popX.$orgdomain SMARTHOST=relay.$orgdomain The protocols you need to support ( list at router.cf ) are: protocols="routes" (I don't think you would need: protocols="routes smtp", but I could be wrong there.. You should not need.) The localnames database should list all variants about the hostname that you are willing to accept, including all the hostnames in the "mboxmap"-cloud: dev.nadt.org.uk nadt.org.uk foo.nadt.org.uk nadt.org.uk nadt.org.uk nadt.org.uk And the routes database should have lines: .nadt.org.uk error!nosuchdomain that.uucp-gw.uk uucp!gwuucpname That is, all domains not matching your local identities (as listed in 'localnames'), but within your $orgdomain are routed thru an error channel to yield an error report. However you need to add that error-form file to the system $MAILSHARE/forms/ -directory. > We do run BIND but only for local host lookup. > > We will also soon want to accept incoming internet mail with a different > domain and pass this through via SMTP to a MS Mail server or possibly CC: > Mail (Another department). Just add more "routes" entries pointing to those, like: ccmail.nadt.org.uk smtp! msmail.nadt.org.uk smtp! some.domain.uk smtp! > All our hosts are FreeBSD 2.x. (if that makes any difference). It does not make any difference. > I'd be very grateful for your help. > > Best Regards > > Robin. > -------------------------------------------------------- > Robin Melville, Addiction & Forensic Information Service > Nottingham Alcohol & Drug Team (Extn. 49178) > Vox: +44 (0)115 952 9478 Fax: +44 (0)115 952 9421 > Email: robmel@nadt.org.uk > WWW: http://www.innotts.co.uk/nadt/ /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 14 20:58:01 1997 Received: by nic.funet.fi id <55400-7011>; Fri, 14 Feb 1997 20:56:41 +0200 Subject: Re: smtp; 500 (nameserver data inconsistency ... From: Matti Aarnio To: bruce@cs.ualberta.ca (Bruce Wm Folliott) Date: Fri, 14 Feb 1997 20:56:33 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <97Feb14.102451-0700_mst.13519-3565+175@scapa.cs.ualberta.ca> from "Bruce Wm Folliott" at Feb 14, 97 10:24:43 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 4678 Message-Id: <19970214185641Z55400-7011+125@nic.funet.fi> Return-Path: > Matti, (Again I am sending privately addressed mail to the list, but this issue has wider interest, I believe.) > I get this bounce (zmailer 2.99.45): > > > : ...\ > > expired after 3 days, problem was: > > smtp; 500 (nameserver data inconsistency. No MX, no address: 'isg_nw.cs.uni-magdeburg.de', errno=Error 0) > > but there is an A record for isg_nw.cs.uni-magdeburg.de: UNDERSCORE in the hostname ? Gee... You see, the lattest resolver codes by Paul Vixie et.al. make that instantly rejected target... The gethostbyname() routine yields NULL. (I am presuming you are using 'bundled libresolv') (The RFCs of the DNS tell that the character set is: ".", "-", "A"-"Z", "a"-"z", "0"-"9" and it definitely does not contain underscore..) Testing that underscorefull address at DEC UNIX 4.0 configured "--with-bundled-libresolv" yields the same result you are seeing, and my "for the comparison" test at a Linux machine without that library SENDS OUT JUST FINE :-/ .... > How do I get more information on why this message bounces? > Mail to deussen@monet.cs.uni-magdeburg.de seems to work > fine. `telnet isg_nw.cs.uni-magdeburg.de smtp` seems to > work fine also. The telnet works ok until it too begins to use 'tightly by the specs' version of the resolver routines. I have talked with my local collegues about ZMailer development issues, and one of the topics is SMTP's reactions to various cases of "No MX, no address"-like situations. I do agree to add some more tests, and different reports -- so that cases where there simply are no DNS data at some zones (that always yield 0 replies) will abort immediately, and those where we are 'the best mx', we be giving a better error message. Giving this host an instant reply of "Hostname with illegal syntax" might speed things up, though. (And explain why the resolver rejected it..) Have a diff (for 2.99.46p1 actually) at the end. Hmm.. Oh damn, now I recall why I made it to be slow at the rejection.. Yes, it should still be slow for the case of "we be the best mx", and fast for all else. As to why ? We plan to do a few ugly tricks at Telecom Finland, and put ZMailer into a HighAvailability-cluster where the machines speak out thru their normal ethernets, and they can all the sudden get be assigned IP-alias of their backup machine. That way we may all the sudden appear as being 'the best mx' for that other machine, (even have its address!), and users/lists in it. At the end there is a diff for improved error messages. > Bruce Wm Folliott /Matti Aarnio --- transports/smtp/smtp.c~ Wed Feb 5 01:38:50 1997 +++ transports/smtp/smtp.c Fri Feb 14 20:26:54 1997 @@ -1853,7 +1853,7 @@ else fprintf(SS->verboselog," rc=%d, mxh[0].host=%s (host=%s)\n", rc,mxh[0].host,host); - hp = NULL; /* Ruined HP within getmxrr() */ + hp = NULL; /* Ruined 'hp' datas within getmxrr() */ switch (rc) { /* remotemsg is generated within getmxrr */ case EX_SOFTWARE: @@ -1885,14 +1885,28 @@ return EX_TEMPFAIL; } - if (noMX) + r = EX_UNAVAILABLE; /* This gives instant rejection */ + + if (strchr(host,'_') != NULL) { sprintf(remotemsg, - "smtp; 500 (configuration inconsistency. MX usage forbidden, no address in DNS: '%s', errno=%s)", - host,strerror(errno)); - else + "smtp; 500 (Hostname with illegal [to the DNS] underscore in it: '%s')", + host); + } else if (noMX) { sprintf(remotemsg, - "smtp; 500 (nameserver data inconsistency. No MX, no address: '%s', errno=%s)", + "smtp; 500 (configuration inconsistency. MX usage forbidden, no address in the DNS: '%s', errno=%s)", host,strerror(errno)); + } else { + if (mxcnt > 0) { + sprintf(remotemsg, + "smtp; 500 (nameserver data inconsistency. All MXes rejected (we be best?), no address: '%s', errno=%s)", + host,strerror(errno)); + r = EX_TEMPFAIL; /* This gives delayed rejection (after a timeout) */ + } else { + sprintf(remotemsg, + "smtp; 500 (nameserver data inconsistency. No MX, no address: '%s', errno=%s)", + host,strerror(errno)); + } + } time(&endtime); notary_setxdelay((int)(endtime-starttime)); notaryreport(NULL,FAILED,"5.4.4 (nameserver data inconsistency)", @@ -1901,7 +1915,7 @@ fprintf(SS->verboselog,"%s\n",remotemsg+6); /* it was: EX_UNAVAILABLE, but such blocks retrying, thus current EX_TEMPFAIL, which will cause timeout latter on.. */ - return EX_TEMPFAIL; + return r; } hp_addr_randomize(hp); { From zmailer-owner@nic.funet.fi Fri Feb 14 21:26:44 1997 Received: by nic.funet.fi id <55609-7011>; Fri, 14 Feb 1997 21:24:54 +0200 Subject: Re: Multiple Routers Processing A Single Message (patch enclosed) From: Matti Aarnio To: andy@realbig.com (Andy Poling) Date: Fri, 14 Feb 1997 21:24:52 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Andy Poling" at Feb 12, 97 10:57:11 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1598 Message-Id: <19970214192454Z55609-7011+128@nic.funet.fi> Return-Path: > Matti- > > As we have discussed recently, one can enforce single-threadedness of > renaming of the message files in $POSTOFFICE/router by using link() and > unlink() and checking the return from unlink(). Participants seemed to be of > the opinion that unlink() should always be an atomic operation, whereas > rename() is apparently not always atomic. > > I've implemented this change after testing the theory using file locks for > several days. The duplicate messages ceased. There were dupes even with file-locks ? > The diff below changes lib/esyslib.c to never use rename(), but to use link() > and unlink() instead and check the outcome of the unlink(). If the unlink() > of "from" fails, then "to" is also unlink()ed and we return -1 to indicate > failure - either due to an error or because another router process link()ed > and unlink()ed the file before we could. This should enforce single-threaded > renaming in the routers. I accepted it in a bit different form. For erename() the rename(2) is ok, but for eqrename() that is used for acquiring locking, it will now always use link(2)/unlink(2). The rename(2) is nice with NFS, I think. One message instead of a slew of them... It looks like ZMailer can never be put on top of Andrew-FS, which allows only one link to any file... (Not that I have any reason to want to do that, for that matter.) > I submit the following diff (against v2.99.46) for your consideration. I > have placed the resultant router into production use on several IRIX 5.3 > systems. > > -Andy /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Feb 14 22:06:46 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <55460-1676>; Fri, 14 Feb 1997 22:05:07 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86744-20108>; Fri, 14 Feb 1997 15:04:44 -0500 Date: Fri, 14 Feb 1997 15:04:43 -0500 (EST) From: Andy Poling To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: Multiple Routers Processing A Single Message (patch enclosed) In-Reply-To: <19970214192454Z55609-7011+128@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 14 Feb 1997, Matti Aarnio wrote: > > I've implemented this change after testing the theory using file locks for > > several days. The duplicate messages ceased. > > There were dupes even with file-locks ? The file locking did, in fact, prevent duplicates. It's just harder to make portable... :-( > It looks like ZMailer can never be put on top of Andrew-FS, which > allows only one link to any file... (Not that I have any reason > to want to do that, for that matter.) That should be OK because AFS was (at least last time I looked) really only designed to handle user's home dierctories. -Andy From zmailer-owner@nic.funet.fi Mon Feb 17 12:06:04 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <73636-26456>; Mon, 17 Feb 1997 12:05:37 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109640-238>; Mon, 17 Feb 1997 11:05:22 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Mon, 17 Feb 1997 11:05:05 +0100 Date: Mon, 17 Feb 1997 11:04:57 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: mailbox no go with HP-UX Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hello, today i see why mailbox doesn't deliver mail on HP-UX 9.05. If mailbox wants to deliver mail it seems as if it first puts a lock (~) in the control file. Then it crashes ! Here is what gdb says: (gdb) run -8 Starting program: /usr/wiss/goldt/zmailer-2.99.46/transports/mailbox/mailbox -8 #hungry 26678-223 Program received signal SIGSYS, Bad system call. 0x7afb7e90 in _mktime () (gdb) where #0 0x7afb7e90 in _mktime () #1 0x7b01ee14 in realloc () So now where does this happen ? Still have to find the bug. Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Mon Feb 17 14:59:30 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <74458-26456>; Mon, 17 Feb 1997 14:58:05 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109558-240>; Mon, 17 Feb 1997 13:57:44 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Mon, 17 Feb 1997 13:57:30 +0100 Date: Mon, 17 Feb 1997 13:57:19 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: bug found Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Ok, mailbox crashes when it calls 'gethostid' in dotlock.c. Here is a gethostid for HP-UX to be inserted in dotlock.c: #ifdef hpux /* * From: dd@mv.us.adobe.com (David DiGiacomo) */ #define _INCLUDE_HPUX_SOURCE #include int gethostid() { struct utsname uts; if (uname(&uts) < 0) return 0; return atoi(uts.idnumber); } #endif Maybe syscall(SYS_GETHOSTID) should work aswell. Now the next problem which zmailer has since at least 2.99.25: If someone sends a postscript file of about 50K in length or longer then zmailer adds some bytes with HP-UX (not with Solaris!). This is a diff: goldt@hprbg1 [61]% (~): diff /tmp/la1 sfb12.dfg.ps 434c434 < FFFFFFFFFEFFFFFFFFE28317CB031>90 D<00FFF0000003FFFE00000F803F80000FC00F --- > FFFFFFFFFEFFFFFFFFFE28317CB031>90 D<00FFF0000003FFFE00000F803F80000FC00F /tmp/la1 is what zmailer delivered to me. In the router dir and in the queue the mail is still fine, so i think mailbox mangles the mail. Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Mon Feb 17 17:51:44 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <74355-5969>; Mon, 17 Feb 1997 17:50:22 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109540-240>; Mon, 17 Feb 1997 16:50:07 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Mon, 17 Feb 1997 16:49:51 +0100 Date: Mon, 17 Feb 1997 16:49:51 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: mailbox.c/writebuf Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi, could you please tell my why writebuf checks for 'From ' in a mail ? The header has already been written and if i want a >From in the beginning of a line i want a From in the beginning of the line and not a >From . I haven't read more yet but i think writebuf() needs a rewrite, my next try is to use write() instead of writebuf()...maybe then it doesn't fake mails. Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Mon Feb 17 18:24:54 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <73699-5969>; Mon, 17 Feb 1997 18:23:02 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109541-237>; Mon, 17 Feb 1997 17:22:49 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Mon, 17 Feb 1997 17:22:37 +0100 Date: Mon, 17 Feb 1997 17:22:28 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: fwrite works Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Ok, i'll use this patch until mea releases a better one... Around line 2067 in mailbox.c: /* if (writebuf(fp, buffer, i, &lastch) != i) { */ if (fwrite(buffer, i, 1, fp) < 1) { Now the mail arrives as it was sent. I think writebuf is too complicated for me (as the whole zmailer source) so maybe someone else detects what's wrong in writebug/writebuf :) Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Mon Feb 17 19:12:17 1997 Received: from swix.ch ([195.210.0.34]) by nic.funet.fi with SMTP id <56076-5969>; Mon, 17 Feb 1997 19:10:46 +0200 Received: from tnt.swix.ch by swix.ch (8.6.10/SMI-4.1/10/pb) id RAA21184; Mon, 17 Feb 1997 17:09:00 GMT Sender: root@Server2.swix.ch Message-ID: <33089152.1E2E9F79@swix.ch> Date: Mon, 17 Feb 1997 18:11:46 +0100 From: "Gregor B. Heini" Organization: - X-Mailer: Mozilla 3.01 (X11; I; Linux 2.0.28 i586) MIME-Version: 1.0 To: zmailer@nic.funet.fi Subject: Unsubscribe - how? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: Hi please tell me how to unsubscribe, I tried sending a mail to majordomo@nic.funet.fi with 'unsubscribe zmailer' in the body, but it didn't work. Thanks From zmailer-owner@nic.funet.fi Mon Feb 17 19:32:02 1997 Received: from jupiter.scs.Ryerson.CA ([141.117.18.31]) by nic.funet.fi with SMTP id <55417-5969>; Mon, 17 Feb 1997 19:30:49 +0200 Received: by jupiter.scs.Ryerson.CA id <10730(3)>; Mon, 17 Feb 1997 12:30:33 -0500 From: Dave Mason To: zmailer@nic.funet.fi Subject: Re: mailbox.c/writebuf In-Reply-To: References: X-Face: %Q_F^9R-:'3MM7eZ6@E.x@f\*bgatzGv-8d%I~L[p^.F)3QF{kq\UTsu|e#?)3FPwJNvPPB !s*He|-*M^p*~bh"Nywm5NLL\\Rl3r(hWHY*F:$/RdKV*bS";n&#\Ov@*=]mu\}6tP Date: Mon, 17 Feb 1997 12:30:31 -0500 Return-Path: Sven Goldt writes: > Hi, > > could you please tell my why writebuf checks for 'From ' in > a mail ? The header has already been written and if i want a > >From in the beginning of a line i want a From in the beginning > of the line and not a >From . > I haven't read more yet but i think writebuf() needs a rewrite, > my next try is to use write() instead of writebuf()...maybe > then it doesn't fake mails. ``\n\nFrom '' is how you find the start/end of a message in a Unix mailbox. Amazingly ugly, but an ingrained fact. Change it at your great peril (unless you change all your mail readers too). ../Dave From zmailer-owner@nic.funet.fi Mon Feb 17 19:32:13 1997 Received: by nic.funet.fi id <55290-5969>; Mon, 17 Feb 1997 19:31:12 +0200 Subject: Re: Unsubscribe - how? From: Matti Aarnio To: tnt@swix.ch (Gregor B. Heini) Date: Mon, 17 Feb 1997 19:31:07 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <33089152.1E2E9F79@swix.ch> from "Gregor B. Heini" at Feb 17, 97 06:11:46 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 257 Message-Id: <19970217173112Z55290-5969+205@nic.funet.fi> Return-Path: > Hi > > please tell me how to unsubscribe, I tried sending a mail > to majordomo@nic.funet.fi with 'unsubscribe zmailer' in the body, > but it didn't work. zmailer-request@nic.funet.fi It does to . > Thanks /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Feb 17 20:34:43 1997 Received: by nic.funet.fi id <55967-5969>; Mon, 17 Feb 1997 20:33:19 +0200 Subject: Re: fwrite works From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Mon, 17 Feb 1997 20:33:17 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Feb 17, 97 05:22:28 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1210 Message-Id: <19970217183319Z55967-5969+209@nic.funet.fi> Return-Path: > Ok, > > i'll use this patch until mea releases a better one... > > Around line 2067 in mailbox.c: > /* if (writebuf(fp, buffer, i, &lastch) != i) { */ > if (fwrite(buffer, i, 1, fp) < 1) { > > Now the mail arrives as it was sent. > I think writebuf is too complicated for me (as the whole zmailer source) > so maybe someone else detects what's wrong in writebug/writebuf :) Like Dave Mason noted, you may get "surprises" :-) Right now I am at phase of developing the beast at my home machine, and I am trying to tie in SOME of the code for the IPv6 to be usable with it -- at least the smtpserver can now (perhaps) accept IPv6 ... ( "perhaps" == "My home machine has Caldera CND 1.0 with Linux 1.2 kernel, and it definitely does not speak IPv6 ..." ) I will try to look at the mailbox too. Btw: I just noticed that at smtp timeout the smtpserver leaves unfinished input files into $POSTOFFICE/public/ directory -- because it does not do proper cleanup on them. More things to look at at home.. ( Sure, I have a 5 line quick-patch for it, but this is something "I Want To Make Right"(TM), and it needs more lines... ) > Sven > -- > The only limits are in your mind ! /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Feb 18 19:59:31 1997 Received: by nic.funet.fi id <74482-27144>; Tue, 18 Feb 1997 19:55:15 +0200 Subject: Re: bug found From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Tue, 18 Feb 1997 19:51:39 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Feb 17, 97 01:57:19 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1452 Message-Id: <19970218175515Z74482-27144+223@nic.funet.fi> Return-Path: (gethostid() source picked up for HP-UX) ... > Now the next problem which zmailer has since at least 2.99.25: > If someone sends a postscript file of about 50K in length or longer > then zmailer adds some bytes with HP-UX (not with Solaris!). > This is a diff: > goldt@hprbg1 [61]% (~): diff /tmp/la1 sfb12.dfg.ps > 434c434 > < FFFFFFFFFEFFFFFFFFE28317CB031>90 D<00FFF0000003FFFE00000F803F80000FC00F > --- > > FFFFFFFFFEFFFFFFFFFE28317CB031>90 D<00FFF0000003FFFE00000F803F80000FC00F > > /tmp/la1 is what zmailer delivered to me. In the router dir and in the > queue the mail is still fine, so i think mailbox mangles the mail. The HP-UX does not have HAVE_MMAP defined in the config.h ? It occurs of the message is larger than BUFSIZE (8192 bytes ?), and the first segment ends with an 'F' in the scan-buffer for finding a 'From ' from the begining of line.. I have now rewritten the code responsible for that scanning to have a properly passed processing state structure instead of magic static variables.. ( .. what on earth had possessed me to write such horrible statics in the first place, I don't know .. I just know that I did them, and I should not have done them in that way :-/ ) Hmm.. 2.99.47 test complile at my machine gronks, because of some IPv6 things :-( Oh bugger... (But then I wrote them blindly at my home machine, which does not have IPv6 stuff :) ) > Sven > -- /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Feb 18 20:27:27 1997 Received: from clgrps05.agt.net ([198.161.156.16]) by nic.funet.fi with ESMTP id <74023-17122>; Tue, 18 Feb 1997 20:21:41 +0200 Received: from CAL_PC.autopro.ca ([198.161.104.17]) by mail.telusplanet.net with SMTP id <153195-25982>; Tue, 18 Feb 1997 11:21:18 -0700 Received: by CAL_PC.autopro.ca with Microsoft Mail id <01BC1D8D.EDDE62E0@CAL_PC.autopro.ca>; Tue, 18 Feb 1997 11:21:47 -0700 Message-ID: <01BC1D8D.EDDE62E0@CAL_PC.autopro.ca> From: Cal Demaine To: "zmailer@nic.funet.fi" Subject: Temporary Account Required Date: Tue, 18 Feb 1997 11:21:46 -0700 Encoding: 48 TEXT Return-Path: I require a temporary mail account on a Zmailer 2.99 machine for testing purposes. I am experiencing problems receiving mail from a Zmailer 2.99 machine through any of the Microsoft Internet Mail tools (Internet Mail, Mail, Exchange, and Outlook). The problem manifests itself as the loss of the last line of any incoming mail, and the addition of an empty line at the beginning of the message. Thus, when I send (or anyone sends) me the following message: ------------- 1 First Line 2 Second Line 3 Last Line ------------- I retrieve back (with the Microsoft tools) -------------- 1 First Line 2 Second Line -------------- The error does not manifest itself with Netscape 2.0's mail facilities. I don't immediately suspect Zmailer or Microsoft, because, with exactly the same software, this use to work like a charm. My ISP points their finger at Microsoft. Microsoft wants to duplicate the error, but my ISP doesn't want to help. I'm suspicious of my ISP's configuration, because, as I said, it just suddenly stopped working correctly. So, is there anyone out there kind and curious enough to help me verify the problem? A temporary mail account is all I require. Please respond by e-mail. Thanks. - Cal Cal Demaine Autopro Automation Consultants Ltd. (403) 539-2456 - cjd_acl@telusplanet.net From zmailer-owner@nic.funet.fi Tue Feb 18 20:44:14 1997 Received: from realbeer.realbeer.com ([204.152.97.15]) by nic.funet.fi with ESMTP id <56057-27144>; Tue, 18 Feb 1997 20:38:19 +0200 Received: (from wyllie@localhost) by realbeer.realbeer.com (8.7.5/8.6.6) id LAA16853 for zmailer@nic.funet.fi; Tue, 18 Feb 1997 11:43:08 -0800 From: Andrew Wyllie Message-Id: <199702181943.LAA16853@realbeer.realbeer.com> Subject: Need advice on large mailing lists To: zmailer@nic.funet.fi Date: Tue, 18 Feb 1997 11:43:08 -0800 (PST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-Path: I am setting up a very large mailing list on a Linux(2.0.0) machine and I am looking for any advice on what the best way to do this might be. The list has approx 50,000 recipients and will be used once a month to send out a newsletter. The current solution uses majordomo and sendmail. When a mail is sent out, it kills the machine (to put it lightly) for a few days which is not accpetable. The solutions I am looking at include replacing sendmail with zmailer and useing ProcList (free versions) or SmartList. We want people to be able to subscribe and unsubscribe to the list (we will probably also develop some type of web interface for this). Any thoughts, bits of wisdom, good/bad experiences would be appreciated. thanks andrew ..................................................................... Andrew Wyllie wyllie@realbeer.com From zmailer-owner@nic.funet.fi Tue Feb 18 21:43:13 1997 Received: from Lisa.engelska.se ([193.14.46.3]) by nic.funet.fi with ESMTP id <55302-27144>; Tue, 18 Feb 1997 21:39:16 +0200 Received: by engelska.se id <28340-123>; Tue, 18 Feb 1997 20:40:19 +0100 From: Nobody To: cjd_acl@telusplanet.net, zmailer@nic.funet.fi Subject: Re: Temporary Account Required Message-Id: <97Feb18.204019+0100_met.28340-123+1@engelska.se> Date: Tue, 18 Feb 1997 20:40:15 +0100 Return-Path: Welp, to ease your worries, I have 450 email accounts on a Zmailer box. all 450 people use Microsoft Internet Mail with no problem (though it does sometimes complain about the LEAVE ON SERVER option. No bugs here. (2.99.43 - yes a bit outdated, but why fix something that aint broken) thomas stromberg From zmailer-owner@nic.funet.fi Tue Feb 18 21:51:05 1997 Received: by nic.funet.fi id <55968-27144>; Tue, 18 Feb 1997 21:47:36 +0200 Subject: Re: Need advice on large mailing lists From: Matti Aarnio To: wyllie@realbeer.com (Andrew Wyllie) Date: Tue, 18 Feb 1997 21:47:24 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199702181943.LAA16853@realbeer.realbeer.com> from "Andrew Wyllie" at Feb 18, 97 11:43:08 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3415 Message-Id: <19970218194736Z55968-27144+236@nic.funet.fi> Return-Path: > I am setting up a very large mailing list on a Linux(2.0.0) machine > and I am looking for any advice on what the best way to do this might > be. The list has approx 50,000 recipients and will be used once > a month to send out a newsletter. The current solution uses > majordomo and sendmail. When a mail is sent out, it kills the machine > (to put it lightly) for a few days which is not accpetable. For that rare posting I suggest running some batch-sending script which picks circa 100 addresses a time from the list, and sends them in one message, then next 100, and so on. Sending 50 000 individual letters is also an option, but I don't suggest it -- though it would not kill ZMailer, while sendmail (and system) would die a horrible death.. Right now the ZMailer is not very good at expanding lists with many thousands of addresses. The router processes producing route analysis for such a bulks will easily grow into humongous sizes (tens of megabytes. It used to grow in N^2 manner, but I was fortunately able to contain that a bit..) A datapoint on that process growth: VGER.RUTGERS.EDU has its router processes grown to about 10 MB, which happens when it expands largest list it has: 1830 addresses. The growth is fairly linear, and the zero size is about 4 MB. With circa 3.3kB/recipient it gives circa 170 MB router process size, if it has to expand the list of 50 000 addresses in core... (The router process does not shrink, because the malloc()'s free() does not deallocate memory pages..) > The solutions I am looking at include replacing sendmail with zmailer > and useing ProcList (free versions) or SmartList. We want people to be > able to subscribe and unsubscribe to the list (we will probably also > develop some type of web interface for this). Yes, handling subscriptions, and doing postings can be done entirely separately from postings, and I suggest it warmly. Usually postings are done in a sort of list-expand operations, however that too easily kills your mailers... However, nothing prevents you from running the posting script on a pipe thru your mailer -- you have some alias, which expands to a pipe running proper scripts, and takeing the incoming message into some temporary storage, and then sending copies of that. > Any thoughts, bits of wisdom, good/bad experiences would be > appreciated. In the ZMailer source package there is an example of a posting script written in PERL (utils/lightpost.pl), which posts INDIVIDUAL messages. You can follow its lead, and write a posting program that posts the message in bundles of a few-dozen-to-few-hundred recipient addresses a time. The purpose of that sample utility is to show HOW to do it in your own code, and while it is a "mostly functional" replacement for the /usr/lib/sendmail, it is not intended to be run for each recipient (It being PERL should make that obvious :) ) If possible, have your posting program to sort the recipient email domains so that (if possible) you pool recipients to some target system to be submitted at the same time. In the SMTP transport all those recipients will be delivered on one session with SINGLE message block associated with all of the addresses. > thanks > andrew > ..................................................................... > Andrew Wyllie > wyllie@realbeer.com /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Feb 18 23:40:54 1997 Received: from netcreations.com ([206.243.201.4]) by nic.funet.fi with ESMTP id <73707-13295>; Tue, 18 Feb 1997 23:34:54 +0200 Received: from rscott.nc ([206.243.201.10]) by NetCreations.com with SMTP id <4545832-308>; Tue, 18 Feb 1997 16:21:42 -0500 Comments: Authenticated sender is From: "Ryan Scott" Organization: NetCreations, Inc To: wyllie@realbeer.com (Andrew Wyllie) Date: Tue, 18 Feb 1997 16:33:41 +0000 Subject: Re: Need advice on large mailing lists Reply-to: rscott@netcreations.com CC: zmailer@nic.funet.fi Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-Id: <97Feb18.162142-0500_est.4545832-308+175@NetCreations.com> Return-Path: > > I am setting up a very large mailing list on a Linux(2.0.0) > > machine and I am looking for any advice on what the best way to do > > this might be. The list has approx 50,000 recipients and will be > > used once a month to send out a newsletter. The current solution > > uses majordomo and sendmail. When a mail is sent out, it kills > > the machine (to put it lightly) for a few days which is not > > accpetable. > I mail quantities of > 60,000 daily all the time using zmailer, and some software I wrote to make sure 1) you don't run out of disk space 2) server load stays reasonable. Huge amts of experiementation brought me to this point: as you've discovered its easy to kill a machine just by sending mail. Ryan ______________________________________________________________________ Ryan Scott - rscott@netcreations.com - 718 522 1531 - Reinventing Direct Marketing on the Net - NetCreations, Inc - http://www.netcreations.com - Targeted Voluntary Email - PostMaster Direct Response - http://www.netcreations.com/postdirect/ *be sure to quote me in your reply* From zmailer-owner@nic.funet.fi Wed Feb 19 11:22:40 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <56115-15647>; Wed, 19 Feb 1997 11:21:26 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109630-241>; Wed, 19 Feb 1997 10:21:13 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Wed, 19 Feb 1997 10:20:56 +0100 Date: Wed, 19 Feb 1997 10:20:46 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: bug found In-Reply-To: <19970218175515Z74482-27144+223@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 18 Feb 1997, Matti Aarnio wrote: > The HP-UX does not have HAVE_MMAP defined in the config.h ? > Right ! So that's why it works with Solaris. Solaris has MMAP and uses therefore another function to append the mail. > It occurs of the message is larger than BUFSIZE (8192 bytes ?), I think BUFSIZ is 1024 bytes with HP-UX...not sure though. > finding a 'From ' from the begining of line.. I have now rewritten > the code responsible for that scanning to have a properly passed > processing state structure instead of magic static variables.. Could you please send me that code ? > Hmm.. 2.99.47 test complile at my machine gronks, because of > some IPv6 things :-( Oh bugger... (But then I wrote them > blindly at my home machine, which does not have IPv6 stuff :) ) Why don't you get a new kernel ? Just look at ftp.funet.fi :) Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Wed Feb 19 13:56:15 1997 Received: by nic.funet.fi id <74223-5704>; Wed, 19 Feb 1997 13:55:26 +0200 Subject: Re: bug found From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Wed, 19 Feb 1997 13:55:16 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Feb 19, 97 10:20:46 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 736 Message-Id: <19970219115526Z74223-5704+116@nic.funet.fi> Return-Path: ... > > Hmm.. 2.99.47 test complile at my machine gronks, because of > > some IPv6 things :-( Oh bugger... (But then I wrote them > > blindly at my home machine, which does not have IPv6 stuff :) ) > > Why don't you get a new kernel ? Just look at ftp.funet.fi :) At home I use CDROM distribution packages, and none of them are enough up to date to help me here :-( Soon we should get SyQuest removable cartridge drives for office machines, and then I can carry 100+ MB easily. (Working at a telecom company development group oddly does not mean I have a phone- line to home, all of us have a cellular, of course... I may sooner get an ADSL link to home, than ordinary phone-line :-)~ ) > Sven > -- /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Feb 19 19:07:29 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <73924-5704>; Wed, 19 Feb 1997 19:06:07 +0200 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id dh22589; Wed, 19 Feb 1997 18:03:51 +0100 Date: Wed, 19 Feb 1997 18:03:51 +0100 From: Gabor Kiss Message-Id: <199702191703.dh22589@zeronull.aszi.sztaki.hu> To: mea@nic.funet.fi, zmailer@nic.funet.fi Subject: Policy based filtering X-Charset: US X-Char-Esc: 0 Return-Path: Dear ZMailers, I've implemented the policy based filtering in smtpserver. The modified C sources, context diffs, and configuration files are in ftp://ftp.sztaki.hu/pub/private/kissg/zmailer/ The patched ZMaler version is 2.99.45. It is running in a Solaris 2.5 environment without any problems for almost 5 hours. :-) Regards Gabor From zmailer-owner@nic.funet.fi Wed Feb 19 19:58:20 1997 Received: by nic.funet.fi id <74359-31729>; Wed, 19 Feb 1997 19:57:15 +0200 Subject: Re: Policy based filtering From: Matti Aarnio To: kissg@zeronull.aszi.sztaki.hu (Gabor Kiss) Date: Wed, 19 Feb 1997 19:57:13 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199702191703.dh22589@zeronull.aszi.sztaki.hu> from "Gabor Kiss" at Feb 19, 97 06:03:51 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 709 Message-Id: <19970219175715Z74359-31729+147@nic.funet.fi> Return-Path: > Dear ZMailers, > > I've implemented the policy based filtering in smtpserver. > The modified C sources, context diffs, and configuration files > are in ftp://ftp.sztaki.hu/pub/private/kissg/zmailer/ > > The patched ZMaler version is 2.99.45. > It is running in a Solaris 2.5 environment without any problems > for almost 5 hours. :-) Interesting bits. You don't say it directly, but I get an implication from the diffs that you run router (albeit in 'server' mode) for each smtp- session now, and that each incoming address is run through that combined server... Which way ever, I am going for home at least once this week before midnight. Back tomorrow. > Regards > > Gabor /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Feb 19 20:09:09 1997 Received: from gpu.utcc.utoronto.ca ([128.100.102.1]) by nic.funet.fi with SMTP id <74245-15647>; Wed, 19 Feb 1997 20:08:02 +0200 Received: from localhost by gpu.utcc.utoronto.ca with SMTP id <336550>; Wed, 19 Feb 1997 13:07:49 -0500 Date: Wed, 19 Feb 1997 13:07:48 -0500 From: Alex Nishri X-Sender: nishri@gpu.utcc Reply-To: alex.nishri@utoronto.ca To: Andrew Wyllie cc: zmailer@nic.funet.fi Subject: Re: Need advice on large mailing lists In-Reply-To: <19970218194736Z55968-27144+236@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: > > I am setting up a very large mailing list on a Linux(2.0.0) machine > > and I am looking for any advice on what the best way to do this might > > be. The list has approx 50,000 recipients and will be used once > > a month to send out a newsletter. The current solution uses > > majordomo and sendmail. When a mail is sent out, it kills the machine > > (to put it lightly) for a few days which is not accpetable. > > For that rare posting I suggest running some batch-sending > script which picks circa 100 addresses a time from the list, > and sends them in one message, then next 100, and so on. We use bulk_mailer (reference ftp://ftp.greatcircle.com/pub/majordomo/FAQ). Bulk_mailer sorts the addresses belonging to the list by reversed domain and sorts them into groups containing no more than N domains; we trivially modified it to also limit the number of recipients. So with 50,000 recipients we actually hand Zmailer 1,000 separate messages with 50 recipients each. (This is called chunking.) Our /local/share/mail/db/aliases contains a line like: list-outgoing: "|bulk_mailer owner-address /local/file.with.list.of.addresses" You may also want to keep mailing list explosions in a separate /var/spool/postoffice to prevent delaying your non-mailing list email. Then there is L-Soft's LSMTP (http://www.lsoft.com), a commercial SMTP product especially designed for sending to large mailing lists. L-Soft claims 100,000 deliveries an hour running on a PC. I don't have experience with it, but I understand that one key difference between Zmailer/sendmail and LSMTP is that the latter will accept a single message with 50,000 recipients (ie no chunking) and work on the addresses concurrently. At bottlenecks like DNS resolution (which takes a long time for a signficant number of addresses on a large international mailing list) LSMTP can have many more concurrent DNS lookups outstanding. To gain that sort of concurrency with Zmailer you would have to break up the 50,000 recipients into 50,000 messages and run a huge number of routers and smtp delivery channels--but then the overhead of dealing with 50,000 messages would be a factor. Alex Nishri University of Toronto From zmailer-owner@nic.funet.fi Wed Feb 19 20:40:51 1997 Received: by nic.funet.fi id <55855-23671>; Wed, 19 Feb 1997 20:39:21 +0200 Subject: Re: Need advice on large mailing lists From: Matti Aarnio To: alex.nishri@utoronto.ca Date: Wed, 19 Feb 1997 20:39:15 +0200 (GMT) Cc: wyllie@realbeer.com, zmailer@nic.funet.fi In-Reply-To: from "Alex Nishri" at Feb 19, 97 01:07:48 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1751 Message-Id: <19970219183921Z55855-23671+180@nic.funet.fi> Return-Path: ... > You may also want to keep mailing list explosions in a separate > /var/spool/postoffice to prevent delaying your non-mailing list email. With present system that is not so easy, however you can define alternate router directories that are picked up when the main router directory is empty. (ROUTERDIRS definition, with "sendmail" program there is a -p -option.). > Then there is L-Soft's LSMTP (http://www.lsoft.com), a commercial SMTP > product especially designed for sending to large mailing lists. L-Soft claims > 100,000 deliveries an hour running on a PC. I don't have experience with it, > but I understand that one key difference between Zmailer/sendmail and LSMTP > is that the latter will accept a single message with 50,000 recipients (ie no > chunking) and work on the addresses concurrently. At bottlenecks like DNS > resolution (which takes a long time for a signficant number of addresses on a > large international mailing list) LSMTP can have many more concurrent DNS > lookups outstanding. To gain that sort of concurrency with Zmailer you would > have to break up the 50,000 recipients into 50,000 messages and run a huge > number of routers and smtp delivery channels--but then the overhead of > dealing with 50,000 messages would be a factor. I have used here an approach which you can see from TELE-FI.cf sample for router.cf -file (proto/cf/ -directory). It uses a list of "known toplevels", and will not do DNS lookups (to canonicalize names) if the last component of the domain name is in the list. That way the ROUTER does not spend time on querying the DNS, but the smtp channel programs can take their time -- they do it in parallel anyway. > Alex Nishri > University of Toronto /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Feb 19 22:11:43 1997 Received: from netcreations.com ([206.243.201.4]) by nic.funet.fi with ESMTP id <55254-5704>; Wed, 19 Feb 1997 22:10:08 +0200 Received: from rscott.nc ([206.243.201.10]) by NetCreations.com with SMTP id <4618228-307>; Wed, 19 Feb 1997 14:57:06 -0500 Comments: Authenticated sender is From: "Ryan Scott" Organization: NetCreations, Inc To: alex.nishri@utoronto.ca Date: Wed, 19 Feb 1997 15:09:05 +0000 Subject: Re: Need advice on large mailing lists Reply-to: rscott@netcreations.com CC: zmailer@nic.funet.fi Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-Id: <97Feb19.145706-0500_est.4618228-307+8017@NetCreations.com> Return-Path: > > > I am setting up a very large mailing list on a Linux(2.0.0) > > > machine and I am looking for any advice on what the best way to > > > do this might be. The list has approx 50,000 recipients and > > > will be used once a month to send out a newsletter. The current > > > solution uses majordomo and sendmail. When a mail is sent out, > > > it kills the machine (to put it lightly) for a few days which is > > > not accpetable. > > > > For that rare posting I suggest running some batch-sending > > script which picks circa 100 addresses a time from the list, and > > sends them in one message, then next 100, and so on. > > We use bulk_mailer (reference > ftp://ftp.greatcircle.com/pub/majordomo/FAQ). Bulk_mailer sorts the > addresses belonging to the list by reversed domain and sorts them > into groups containing no more than N domains; we trivially modified > it to also limit the number of recipients. So with 50,000 recipients > we actually hand Zmailer 1,000 separate messages with 50 recipients > each. (This is called chunking.) does chunking affect the to: or from: lines or the header in any way? Does the router do one DNS lookup once per message in that case? In the case of 50 messages with one recipient each, does it do 50 lookups, even if they are to one domain? I guess I'm just asking what kind of buffering of DNS its doing. ______________________________________________________________________ Ryan Scott - rscott@netcreations.com - 718 522 1531 - Reinventing Direct Marketing on the Net - NetCreations, Inc - http://www.netcreations.com - Targeted Voluntary Email - PostMaster Direct Response - http://www.netcreations.com/postdirect/ *be sure to quote me in your reply* From zmailer-owner@nic.funet.fi Wed Feb 19 23:48:40 1997 Received: from gpu.utcc.utoronto.ca ([128.100.102.1]) by nic.funet.fi with SMTP id <55703-15647>; Wed, 19 Feb 1997 23:47:25 +0200 Received: from localhost by gpu.utcc.utoronto.ca with SMTP id <336719>; Wed, 19 Feb 1997 16:46:23 -0500 Date: Wed, 19 Feb 1997 16:46:20 -0500 From: Alex Nishri X-Sender: nishri@gpu.utcc Reply-To: alex.nishri@utoronto.ca To: Ryan Scott cc: zmailer@nic.funet.fi Subject: Re: Need advice on large mailing lists In-Reply-To: <97Feb19.145706-0500_est.4618228-307+8017@NetCreations.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: > does chunking affect the to: or from: lines or the header in any way? The rfc822 headers are unchanged. Chunking simply groups the SMTP recipients in one message into several SMTP messages. If you use a mailing list manager such as Listserv[tm] or Listproc then it already chunks. See the Mailing List Manager (MLM) FAQ at ftp://ftp.uu.net/usenet/news.answers/mail/list-admin/software-faq We use Listserv[tm] with Zmailer on another machine and had to turn down the number of addresses it grouped to because it blew up the router. > Does the router do one DNS lookup once per message in that case? We turn off router DNS lookups. On international mailing lists we find that many times about 10% of the lookups time out or nearly time out. (With "ordinary" email we consistently see less than 1% of DNS lookups time out.) Even if you turn down the length of time to timeout, on international mailing lists this can be significant. Alex Nishri University of Toronto From zmailer-owner@nic.funet.fi Thu Feb 20 16:04:41 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <56852-10044>; Thu, 20 Feb 1997 16:03:11 +0200 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id LH27821; Thu, 20 Feb 1997 15:02:15 +0100 Date: Thu, 20 Feb 1997 15:02:15 +0100 From: Gabor Kiss Message-Id: <199702201402.LH27821@zeronull.aszi.sztaki.hu> To: mea@nic.funet.fi Subject: Re: Policy based filtering Cc: zmailer@nic.funet.fi X-Charset: US X-Char-Esc: 0 Return-Path: > > > I've implemented the policy based filtering in smtpserver. > > The modified C sources, context diffs, and configuration files > > are in ftp://ftp.sztaki.hu/pub/private/kissg/zmailer/ Ooops. A half sentence is missing from README file: smtpserver must run with at least -sfth options. (-sh is a new option to force 'HELO' checking mode.) > > Interesting bits. You don't say it directly, but > I get an implication from the diffs that you run > router (albeit in 'server' mode) for each smtp- > session now, and that each incoming address is run > through that combined server... Yes, over the simple syntax check, server() subroutine looks up addresses in policy database. Gabor From zmailer-owner@nic.funet.fi Thu Feb 20 16:04:44 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <56260-13929>; Thu, 20 Feb 1997 16:03:26 +0200 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109800-241>; Thu, 20 Feb 1997 15:03:15 +0100 Received: by sunrbg6.informatik.tu-muenchen.de id <7854-229>; Thu, 20 Feb 1997 15:03:04 +0100 Date: Thu, 20 Feb 1997 15:03:01 +0100 (MET) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: mailbox still buggy Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Well, today i test the newest zmailer on HP-UX 9.05 (no MMAP) and i have to reuse mailbox from 2.99.25 because mailbox 2.99.46/47 can't handle pipes ( |procmail ... ). It tries to write to it as a file as it seems... further investigation later. Sven -- The only limits are in your mind ! From zmailer-owner@nic.funet.fi Fri Feb 28 16:46:01 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <74426-20885>; Fri, 28 Feb 1997 16:44:35 +0200 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id kG18786; Fri, 28 Feb 1997 15:44:22 +0100 Date: Fri, 28 Feb 1997 15:44:22 +0100 From: Gabor Kiss Message-Id: <199702281444.kG18786@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: alias.cf inconsistency Cc: laszlo.kovacs@sztaki.hu X-Charset: US X-Char-Esc: 0 Return-Path: Dear ZMailers, I think there is some minor inconsistency in alias.cf (from ZMailer 2.99.44). When routeuser() looks for a .forward file of a local user, it its privilege is nobody, because default_attributes contains this. ------ # only allow .forward file reading if privs allow it case "$(get $attr type)" in expandsender) a=$(homedirectory "$user")/.forward && [ -f $a ] && \ return (($quad)) ;; esac priv=$(get $attr privilege) ###### THIS IS THE MATTER [ $PUNTHOST ] || [ $didexpand ] || a="$(homedirectory "$user")/.forward" && [ -f "$a" ] && db add expansions "$key" user && priv=$(getpriv "644" $priv "$a" .forward) && return $(listaddresses -e "$user" -c "$a .forward expansion" < "$a" | \ maprrouter $(newattribute $attr privilege $priv) \ "$a" "$user") ------- If user's home dir is unreadable, the mail won't be forwarded. However if the addresse is an alias, the privilege are set to equal to owner of alias database, usually root: ------- # alias expansion [ $didexpand ] || a="$(aliases "$user")" && db add expansions "$key" alias && priv=$(filepriv -M 644 $MAILVAR/db/aliases \ $(db owner aliases)) && #### LOOK THIS nattr=$(newattribute $attr privilege $priv) && return $(echo "$a" | listexpand -e root \ -c 'alias expansion' \ -N - \ "$nattr" "$a" "$user") ------- Then the .forward file can be read even in an unreadable homedir. Question: Why restricton of privilege is necessary when looking .forward? If user doesn't want to forward mails [s]he should remove the .forward file instead of protecting homedir. I think reading .forward by root force doesn't hurt her/his privacy. In other words, why peoples with protected homedir can't enjoy the benefits of .forward? Any comment? Gabor From zmailer-owner@nic.funet.fi Fri Feb 28 17:52:53 1997 Received: from dash.widearea.co.uk ([194.70.234.6]) by nic.funet.fi with ESMTP id <74381-15252>; Fri, 28 Feb 1997 17:51:24 +0200 Received: by dash.widearea.co.uk id <36949-12256>; Fri, 28 Feb 1997 15:55:04 +0000 From: Simon Brock To: zmailer@nic.funet.fi Subject: ta_logs_diagnostics and Linux Content-Length: 771 Message-Id: <97Feb28.155504+0000_gmt.36949-12256+48@dash.widearea.co.uk> Date: Fri, 28 Feb 1997 15:55:01 +0000 Return-Path: I have just installed 2.99.46 with patches on a Slackware Linux system and had a problem relating to the ta_logs_diagnostics in the transport agent library. The installation was fine and everything worked until there was a problem delivering mail. If the transport agents could not deliver mail then no error mail messages were being generated. This led to me the ta_logs_diagnostics variable in transports/libta/diagnostic.c. I think I see what this is meant to do and I have read the comments in README.UPGRADING (though this was a clean install). However I cannot error report to work unless I set the variable to 0. I don't think this is a huge problem to me but am wondering if I have got something wrong -- 2.99.43 on my local Sun works fine (:-). Simon. From zmailer-owner@nic.funet.fi Mon Mar 3 16:36:27 1997 Received: from ccduck.ke.sanet.sk ([193.87.100.12]) by nic.funet.fi with ESMTP id <73647-1332>; Mon, 3 Mar 1997 16:32:52 +0200 Received: by ccduck.ke.sanet.sk id <16249-3>; Mon, 3 Mar 1997 15:31:52 +0100 Subject: mail cycles around From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Mon, 3 Mar 1997 15:31:44 +0100 (MET ) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 354 Message-Id: <97Mar3.153152+0100met_dst.16249-3@ccduck.ke.sanet.sk> Return-Path: Hello, I am running zmailer 2.97 (no newer version worked...) and sometimes some mail loops and is being delivered againa and again... what's this ? any workarounds ? -- E-mail: Matus.Uhlar@tuke.sk WWW: http://ccsun.tuke.sk/users/uhlar IRC: fantomas, TALK: uhlar@ccnews.ke.sanet.sk ...and if you think I'm talking for my employer, you're wrong... From zmailer-owner@nic.funet.fi Tue Mar 4 15:52:12 1997 Received: from Lisa.engelska.se ([193.14.46.3]) by nic.funet.fi with ESMTP id <73703-10553>; Tue, 4 Mar 1997 15:50:48 +0200 Received: from Lisa.engelska.se ([193.14.46.3]) by engelska.se with SMTP id <28339-125>; Tue, 4 Mar 1997 14:51:57 +0100 Date: Tue, 4 Mar 1997 14:51:47 +0100 (MET) From: "Thomas Stromberg (Nobody)" To: zmailer@nic.funet.fi Subject: Shadow support for POP3? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I'm just curious if it has shadow password support for POP3.. Probably do but I never saw any mention of it in makefiles/documentation. thomas From zmailer-owner@nic.funet.fi Wed Mar 5 11:49:48 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <62074-6016>; Wed, 5 Mar 1997 09:13:46 +0200 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86877-10698>; Wed, 5 Mar 1997 02:13:13 -0500 Date: Wed, 5 Mar 1997 02:13:01 -0500 (EST) From: Andy Poling To: "Thomas Stromberg (Nobody)" cc: zmailer@nic.funet.fi Subject: Re: Shadow support for POP3? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 4 Mar 1997, Thomas Stromberg (Nobody) wrote: > I'm just curious if it has shadow password support for POP3.. > Probably do but I never saw any mention of it in makefiles/documentation. Unfortunately Thomas, the question doesn't apply to ZMAILER. ZMAILER is an MTA (Mail Transport Agent) only, while POP and IMAP servers are MUA's (Mail User Agents). An MTA is basically responsible for sending msgs to other machines, receiving msgs from other machines, and putting msgs into user mailboxes/maildrops. An MUA is basically responsible for providing the user with access to the msgs in their mailbox/maildrop. The best POP servers I've used are those included in the IMAP distribution available from ftp.cac.washington.edu (that's from memory). They are more protocol compliant and less bug-prone than the many others I've run into - especially those with special (read proprietary) extensions for proprietary POP clients (yech!). -Andy Global Auctions From zmailer-owner@nic.funet.fi Wed Mar 5 11:49:50 1997 Received: from Lisa.engelska.se ([193.14.46.3]) by nic.funet.fi with ESMTP id <62177-11365>; Wed, 5 Mar 1997 09:16:42 +0200 Received: from Lisa.engelska.se ([193.14.46.3]) by engelska.se with SMTP id <28342-123>; Wed, 5 Mar 1997 08:18:11 +0100 Date: Wed, 5 Mar 1997 01:18:05 -0600 (CST) From: "Thomas Stromberg (Nobody)" To: Andy Poling cc: zmailer@nic.funet.fi Subject: Re: Shadow support for POP3? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 5 Mar 1997, Andy Poling wrote: > > On Tue, 4 Mar 1997, Thomas Stromberg (Nobody) wrote: > > I'm just curious if it has shadow password support for POP3.. > > Probably do but I never saw any mention of it in makefiles/documentation. > > Unfortunately Thomas, the question doesn't apply to ZMAILER. ZMAILER is an > MTA (Mail Transport Agent) only, while POP and IMAP servers are MUA's (Mail > User Agents). An MTA is basically responsible for sending msgs to other > machines, receiving msgs from other machines, and putting msgs into user > mailboxes/maildrops. An MUA is basically responsible for providing the user > with access to the msgs in their mailbox/maildrop. > > The best POP servers I've used are those included in the IMAP distribution > available from ftp.cac.washington.edu (that's from memory). They are more > protocol compliant and less bug-prone than the many others I've run into - > especially those with special (read proprietary) extensions for proprietary > POP clients (yech!). > > -Andy > > Global Auctions > Oh, I had thought I was running a zmailer pop server the whole time :) In any case, I'm using qpopper right now, no problems yet. A bit of a misunderstanding I spose. thomas From zmailer-owner@nic.funet.fi Wed Mar 5 23:09:55 1997 Received: from gateway ([129.128.129.130]) by nic.funet.fi with ESMTP id <56154-29478>; Wed, 5 Mar 1997 23:07:45 +0200 Received: from hulkster ([192.139.18.252]) by gateway.edm.trlabs.ca with ESMTP id <75282>; Wed, 5 Mar 1997 14:07:38 -0700 Received: from mork by mailhost.edm.trlabs.ca id <12135>; Wed, 5 Mar 1997 14:07:34 -0700 Subject: Pine3.95 & Zmailer2.99.21 From: Luke Chong To: zmailer@nic.funet.fi Date: Wed, 5 Mar 1997 14:07:30 -0700 (MST) X-Mailer: ELM [version 2.4 PL25] Content-Type: text Content-Length: 798 Message-Id: <97Mar5.140734-0700_mst.12135+1367@mailhost.edm.trlabs.ca> Return-Path: Hi: We have Zmailer2.99.21 (I know it is really old) installed on a DEC/OSF machine running UNIX. Now, I've just started to use PINE3.95. However, I always get some error message regarding to disk space of the spool directory. I have checked the disk space and I still have a lot. SO, anyone out there has seen this problem with PINE & Zmailer? Thanks! _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/ Mr. Luke L. S. Chong Telecommunications Research Laboratories _/ _/ Computer Systems Administrator 800 Park Plaza, 10611 - 98 Ave _/ _/ Tel: (403) 441-3818 Edmonton, Alberta _/ _/ Fax: (403) 441-3600 Canada T5K 2P7 _/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ From zmailer-owner@nic.funet.fi Wed Mar 5 23:30:39 1997 Received: from aecgate.aec.ca ([142.56.1.1]) by nic.funet.fi with ESMTP id <55682-32727>; Wed, 5 Mar 1997 23:29:29 +0200 Received: from sol.aec.ca ([142.56.17.16]) by aecgate.aec.ca with ESMTP id <46778-3>; Wed, 5 Mar 1997 14:29:13 -0700 Received: from cygnus ([10.1.60.100]) by aec.ca with SMTP id <137380-2>; Wed, 5 Mar 1997 14:29:01 -0700 Received: by cygnus (SMI-8.6/aec_sub.1.03) id OAA05023; Wed, 5 Mar 1997 14:28:58 -0700 From: "Trevor Paquette" Message-Id: <9703051428.ZM5021@cygnus.aec.ca> Date: Wed, 5 Mar 1997 14:28:57 -0700 X-Mailer: Z-Mail (3.2.1 10apr95) To: zmailer@nic.funet.fi Subject: problem with zmailer 2.99.46, No MX, no address Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-Path: Here is a good one.. Mail stopped flowing this morning around 10:00am for ten minutes.. Our name server went down.. Mail going through at that time was getting the error: 500 (nameserver data inconsistency. No MX, no address I killed zmailer with zmailer kill, and restarted the nameserver. All mail now goes though fine ..EXCEPT for those messages that failed during that outage! They are still getting that error. All other mail flows around those stuck messages. I've tried stopping and restarting zmailer many times.. but those messages STILL get that error.. I've even restart the nameserver many times.. Any ideas? -- Trevor Paquette | MetroNet Solutions |Work:(403)543-2355 TrevorPaquette@mcc.net | 2700, 421 7th Ave SW | Fax:(403)543-2854 http://www.mcc.net | Calgary, Ab, Canada |ICBM:51'05"N/114'01"W Senior Unix Network Architect| T2P 4K9 |Mind:In the Rockies From zmailer-owner@nic.funet.fi Thu Mar 6 00:36:15 1997 Received: from aecgate.aec.ca ([142.56.1.1]) by nic.funet.fi with ESMTP id <55870-20035>; Thu, 6 Mar 1997 00:34:15 +0200 Received: from sol.aec.ca ([142.56.17.16]) by aecgate.aec.ca with ESMTP id <46852-2>; Wed, 5 Mar 1997 15:33:53 -0700 Received: from cygnus ([10.1.60.100]) by aec.ca with SMTP id <137427-1>; Wed, 5 Mar 1997 15:33:44 -0700 Received: by cygnus (SMI-8.6/aec_sub.1.03) id PAA05053; Wed, 5 Mar 1997 15:33:34 -0700 From: "Trevor Paquette" Message-Id: <9703051533.ZM5051@cygnus.aec.ca> Date: Wed, 5 Mar 1997 15:33:32 -0700 In-Reply-To: James MacKinnon "Re: problem with zmailer 2.99.46, No MX, no address" (Mar 5, 14:52) References: X-Mailer: Z-Mail (3.2.1 10apr95) To: James MacKinnon Subject: Re: problem with zmailer 2.99.46, No MX, no address Cc: zmailer@nic.funet.fi Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-Path: That did it!! Thanx! Matti? any observations? On Mar 5, 14:52, James MacKinnon wrote: > Subject: Re: problem with zmailer 2.99.46, No MX, no address > Hi Trevor, > > Try shutting down zmailer. > > Then move the problem children from postoffice/queue to postoffice/router > (so they'll go through the router again), and delete their counterpart > files in postoffice/transport and postoffice/scheduler/{hold|smtp}. > > Then do a zmailer bootclean and start up zmailer. > > Cheers, > -- > James S. MacKinnon Office: P-139 Avahd-Bhatia Physics Lab > Team Physics Voice : (403) 492-8226 > University of Alberta email : Jim.MacKinnon@Phys.UAlberta.CA > Edmonton, Canada T6G 2N5 > WWW: http://www.phys.ualberta.ca/~jmack/jmack.html > >-- End of excerpt from James MacKinnon -- Trevor Paquette | MetroNet Solutions |Work:(403)543-2355 TrevorPaquette@mcc.net | 2700, 421 7th Ave SW | Fax:(403)543-2854 http://www.mcc.net | Calgary, Ab, Canada |ICBM:51'05"N/114'01"W Senior Unix Network Architect| T2P 4K9 |Mind:In the Rockies From zmailer-owner@nic.funet.fi Thu Mar 6 21:39:18 1997 Received: from pobox3.funet.fi ([128.214.46.64]) by nic.funet.fi with ESMTP id <3179-479>; Thu, 6 Mar 1997 21:37:17 +0200 Received: from brimstone.netspace.org (brimstone.netspace.org [128.148.157.143]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with ESMTP id JAA25016 for ; Thu, 6 Mar 1997 09:24:06 +0200 (EET) Received: from netspace.org ([128.148.157.6]) by brimstone.netspace.org with ESMTP id <32816-22132>; Thu, 6 Mar 1997 02:23:28 -0500 Received: from localhost (bfisk@localhost) by netspace.org (8.8.5/8.8.2) with SMTP id CAA19657 for ; Thu, 6 Mar 1997 02:25:39 -0500 Date: Thu, 6 Mar 1997 02:25:39 -0500 (EST) From: Brian Fisk Reply-To: Brian Fisk To: zmailer@nic.funet.fi Subject: A couple quick questions In-Reply-To: <96Nov26.150311+0200_eet.61273-10503+130@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: A few quick questions, the rest of the list might like to know the answers too: 1) You can configure listserv to send out messages with as many recipients per message as you want. Is zmailer more efficient with large numbers of recipients (i.e. 200) or smaller (i.e. 20 or 50)? Or neither? 2) You can also sort the outgoing recipients by hostname, so that those with the same hostname are likely to appear in the same message. Does this matter to zmailer at all? Also, maybe a bug report: On my linux 1.2.13 system (486DX4/100, 40MB), whenever I send a SIGUSR1 to the scheduler to re-read the config file, it just dies. Anyone else had similar experiences? I'll send more info (Matti) if you need it. Thanks! -Brian -- Brian Fisk * bfisk@netspace.org * http://www.netspace.org/users/bfisk -- You're not going crazy, you're going sane in a crazy world! From zmailer-owner@nic.funet.fi Fri Mar 7 16:09:11 1997 Received: by nic.funet.fi id <4198-12162>; Fri, 7 Mar 1997 16:00:05 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: I have been traveling... Message-Id: <19970307140005Z4198-12162+32@nic.funet.fi> Date: Fri, 7 Mar 1997 15:59:59 +0200 Return-Path: Hello, I just got back from a week of meetings in USA. ... and I am already learning to hate airplanes, and jet-lag... I have already noticed some questions presented at ZMailer list (and privately), but it will take yet some days before I can process them all.. (That includes 2 days of solid sleep..) /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Mar 8 18:21:34 1997 Received: from monet.mingpaoxpress.com ([205.150.120.3]) by nic.funet.fi with ESMTP id <4604-11814>; Sat, 8 Mar 1997 18:20:58 +0200 Received: by www.mingpaoxpress.com id <1928-185>; Sat, 8 Mar 1997 11:20:50 -0500 Newsgroups: local.mailgate.zmailer Path: acli Sender: Ambrose Li From: acli@www.mingpaoxpress.com (Ambrose Li) Subject: Re: Pine3.95 & Zmailer2.99.21 Distribution: local Organization: Ming Pao Daily News (Canada) Message-ID: References: <97Mar5.140734-0700_mst.12135+1367@mailhost.edm.trlabs.ca> Date: Sat, 8 Mar 1997 16:20:44 GMT To: unlisted-recipients:; (no To-header on input) Return-Path: [cross-posted to the pine-info list] In article <97Mar5.140734-0700_mst.12135+1367@mailhost.edm.trlabs.ca>, Luke Chong wrote: >We have Zmailer2.99.21 (I know it is really old) installed on a DEC/OSF >machine running UNIX. Now, I've just started to use PINE3.95. However, >I always get some error message regarding to disk space of the spool >directory. I have checked the disk space and I still have a lot. SO, >anyone out there has seen this problem with PINE & Zmailer? Something must have changed between Pine 3.91 and 3.95 (for the worse). I run zmailer 2.99.43 on my Linux system (kernel 2.0.29, libc 5.2.18, ncurses 3.0), and have tried both pine 3.95 and 3.96, and they both hang (stay at 0%) whenever I *send* mail. (Reading mail works fine.) I have yet to take a real look in the sources at what it really does when sending mail. Has anyone successfully used Pine 3.95+ on a system running zmailer? -- Ambrose Li. acli@mingpaoxpress.com. Ming Pao Newspapers (Canada) Ltd., EDP department. 1355 Huntingwood Drive, Scarborough, Ontario, M1S 3J1, Canada. Voice +1 416 321 0088 x272 Fax +1 416 321 9663. My favourite OS has yet no typographic identity; but I would rather use Garamond than Franklin Gothic From zmailer-owner@nic.funet.fi Sat Mar 8 19:13:25 1997 Received: from Zone.pspt.fi ([193.166.55.160]) by nic.funet.fi with ESMTP id <4328-10715>; Sat, 8 Mar 1997 19:12:07 +0200 Received: from localhost (oh7lzb@localhost) by Zone.pspt.fi (8.8.5/8.8.5) with SMTP id TAA21279; Sat, 8 Mar 1997 19:11:51 +0200 Date: Sat, 8 Mar 1997 19:11:51 +0200 (EET) From: Heikki Hannikainen X-Sender: oh7lzb@Zone.pspt.fi Reply-To: Heikki Hannikainen To: Ambrose Li cc: zmailer@nic.funet.fi Subject: Re: Pine3.95 & Zmailer2.99.21 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sat, 8 Mar 1997, Ambrose Li wrote: > I run zmailer 2.99.43 on my Linux system (kernel 2.0.29, libc 5.2.18, > ncurses 3.0), and have tried both pine 3.95 and 3.96, and they both > hang (stay at 0%) whenever I *send* mail. (Reading mail works fine.) > Has anyone successfully used Pine 3.95+ on a system running zmailer? I use it here. Had the same problem, until i made pine send mail using smtp, ie. set smtp-server in the setup to localhost. --- < Heikki Hannikainen <> ax.25: oh7lzb@oh7rba.#kuo.fin.eu > < Internet: hessu@pspt.fi <> Amprnet: oh7lzb@gw.oh7rba.ampr.org > From zmailer-owner@nic.funet.fi Sat Mar 8 19:53:48 1997 Received: from pc164103.oulu.fi ([130.231.164.103]) by nic.funet.fi with ESMTP id <4813-9815>; Sat, 8 Mar 1997 19:52:51 +0200 Received: by pc164103.oulu.fi id <147790-198>; Sat, 8 Mar 1997 19:52:22 +0200 Date: Sat, 8 Mar 1997 19:52:15 +0200 (EET) Sender: Timo Karjalainen From: Timo Karjalainen To: zmailer@nic.funet.fi Subject: Re: Pine3.95 & Zmailer2.99.21 In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sat, 8 Mar 1997, Heikki Hannikainen wrote: > On Sat, 8 Mar 1997, Ambrose Li wrote: > > > I run zmailer 2.99.43 on my Linux system (kernel 2.0.29, libc 5.2.18, > > ncurses 3.0), and have tried both pine 3.95 and 3.96, and they both > > hang (stay at 0%) whenever I *send* mail. (Reading mail works fine.) > > > Has anyone successfully used Pine 3.95+ on a system running zmailer? > > I use it here. Had the same problem, until i made pine send mail using > smtp, ie. set smtp-server in the setup to localhost. That's one solution, another is to use pine's sendmail-path -setting. I have the following line in my pine.conf: sendmail-path=/usr/sbin/sendmail -t -Timo- timok@iki.fi From zmailer-owner@nic.funet.fi Sun Mar 9 15:33:00 1997 Received: from dub-gw-2.compuserve.com ([149.174.207.12]) by nic.funet.fi with SMTP id <4490-7677>; Sun, 9 Mar 1997 15:32:00 +0200 Received: from worldnet.worldnet.att.net (usr8-dialup6.mix1.Sacramento.mci.net [166.55.4.198]) by dub-gw-2.compuserve.com (8.6.10/5.941228sam) id GAA11898; Sun, 9 Mar 1997 06:34:39 -0500 Date: Sun, 9 Mar 1997 06:34:39 -0500 From: Message-Id: <199703091134.GAA11898@dub-gw-2.compuserve.com> Subject: To: unlisted-recipients:; (no To-header on input) Return-Path: Hi Guys, Look at this ad I saw last week. I thought you might be interested. ****************** IT's HOT! ********************* ************* 4 in 1 LIVE DATELINE *************** ******** LOCAL in your area or NATIONWIDE ******** YOU'VE READ ABOUT IT, NOW WE'VE GOT IT! THE NATION'S MOST POPULAR DATELINE. 1. Listen to Personal Messages! 2. Record Personal Messages! 3. Leave Replies! 4. "PRESS A KEY" AND TALK *LIVE**ONE-ON-ONE**! 1-900-860-7444 Ext.# 467 see ya soon! Sue Russo From zmailer-owner@nic.funet.fi Mon Mar 10 07:54:55 1997 Received: by nic.funet.fi id <4734-25588>; Mon, 10 Mar 1997 07:53:57 +0200 Subject: SPAM to ZMailer -list ... From: Matti Aarnio To: zmailer@nic.funet.fi Date: Mon, 10 Mar 1997 07:53:43 +0200 (GMT) In-Reply-To: <199703091134.GAA11898@dub-gw-2.compuserve.com> from "susan140@juno.com" at Mar 9, 97 06:34:39 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 808 Message-Id: <19970310055357Z4734-25588+55@nic.funet.fi> Return-Path: > Hi Guys, > > Look at this ad I saw last week. I thought you might > be interested. Wow -- SPAM to ZMailer list... Ok, I must speed up the incoming smtp-filtering introduction into the ZMailer ... /Matti Aarnio 20670# connection from IDENT-FD2-FAULT-1-38@dub-gw-2.compuserve.com (port 31756) 20670w 220 nic.funet.fi ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Sun, 9 Mar 1997 15:31:44 +0200 20670# remote from [149.174.207.12] 20670r HELO dub-gw-2.compuserve.com 20670w 250 nic.funet.fi Hello dub-gw-2.compuserve.com 20670r MAIL From: 20670w 250 2.1.0 Sender syntax Ok 20670r RCPT To: 20670w 250 2.1.5 Recipient address syntax Ok 20670r RCPT To: 20670w 250 2.1.5 Recipient address syntax Ok 20670r DATA From zmailer-owner@nic.funet.fi Mon Mar 10 09:17:10 1997 Received: by nic.funet.fi id <4734-9566>; Mon, 10 Mar 1997 07:53:57 +0200 Subject: SPAM to ZMailer -list ... From: Matti Aarnio To: zmailer@nic.funet.fi Date: Mon, 10 Mar 1997 07:53:43 +0200 (GMT) In-Reply-To: <199703091134.GAA11898@dub-gw-2.compuserve.com> from "susan140@juno.com" at Mar 9, 97 06:34:39 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 808 Message-Id: <19970310055357Z4734-9566+57@nic.funet.fi> Return-Path: > Hi Guys, > > Look at this ad I saw last week. I thought you might > be interested. Wow -- SPAM to ZMailer list... Ok, I must speed up the incoming smtp-filtering introduction into the ZMailer ... /Matti Aarnio 20670# connection from IDENT-FD2-FAULT-1-38@dub-gw-2.compuserve.com (port 31756) 20670w 220 nic.funet.fi ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Sun, 9 Mar 1997 15:31:44 +0200 20670# remote from [149.174.207.12] 20670r HELO dub-gw-2.compuserve.com 20670w 250 nic.funet.fi Hello dub-gw-2.compuserve.com 20670r MAIL From: 20670w 250 2.1.0 Sender syntax Ok 20670r RCPT To: 20670w 250 2.1.5 Recipient address syntax Ok 20670r RCPT To: 20670w 250 2.1.5 Recipient address syntax Ok 20670r DATA From zmailer-owner@nic.funet.fi Mon Mar 10 15:10:07 1997 Received: by nic.funet.fi id <4957-22778>; Mon, 10 Mar 1997 15:08:39 +0200 Subject: Re: Pine3.95 & Zmailer2.99.21 From: Matti Aarnio To: wizzo@pc164103.oulu.fi (Timo Karjalainen) Date: Mon, 10 Mar 1997 15:08:29 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Timo Karjalainen" at Mar 8, 97 07:52:15 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1205 Message-Id: <19970310130839Z4957-22778+126@nic.funet.fi> Return-Path: > On Sat, 8 Mar 1997, Heikki Hannikainen wrote: > > On Sat, 8 Mar 1997, Ambrose Li wrote: > > > > > I run zmailer 2.99.43 on my Linux system (kernel 2.0.29, libc 5.2.18, > > > ncurses 3.0), and have tried both pine 3.95 and 3.96, and they both > > > hang (stay at 0%) whenever I *send* mail. (Reading mail works fine.) > > > > > Has anyone successfully used Pine 3.95+ on a system running zmailer? > > > > I use it here. Had the same problem, until i made pine send mail using > > smtp, ie. set smtp-server in the setup to localhost. > > That's one solution, another is to use pine's sendmail-path -setting. I > have the following line in my pine.conf: > > sendmail-path=/usr/sbin/sendmail -t The default value appears to be: sendmail-path=/usr/lib/sendmail -bs -odb -oem It just happens that the ZMailer's 'sendmail' ``complains'' about ignored options, and it looks like the complaint goes to stdin of the PINE -- which expects to get the SMTP greeting: 220 hostname ZMailer .... Once I recognized that, fix was easy.. ( The complaint goes to the stderr, but PINE apparently binds child process stdout, and stderr together :-( ) > -Timo- > timok@iki.fi /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 12 13:34:55 1997 Received: from pobox.csc.fi ([128.214.46.62]) by nic.funet.fi with ESMTP id <4156-20818>; Wed, 12 Mar 1997 13:33:03 +0200 Received: from chronos (chronos.sovam.com [194.67.3.135]) by pobox.csc.fi (8.7.5/8.6.9+CSC-2.0) with ESMTP id LAA25356 for ; Wed, 12 Mar 1997 11:32:54 GMT Content-Transfer-Encoding: 8bit Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-3253>; Wed, 12 Mar 1997 14:06:45 +0300 Date: Wed, 12 Mar 1997 14:06:41 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Where are nondelivery reports? To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Am I missing something or notaryreports are broken in .46? It seems that since I installed .46 (plus two patches) almost all non-delivery reports stopped. Messages are delivered successfully when it can be, but when it cannot, there are no reports. This is definitely true for "No such user" condition, and maybe for some other, but *some* reports (e.g. DNS inconsistency) are generated. Maybe I've broken something obvious... Solaris 2.5.1 on Sparc, Zmailer 2.99.46+p1+p2 compiled with gcc. Eugene From zmailer-owner@nic.funet.fi Wed Mar 12 15:21:40 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <3773-20328>; Wed, 12 Mar 1997 15:20:17 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-7290>; Wed, 12 Mar 1997 16:20:03 +0300 Date: Wed, 12 Mar 1997 16:19:57 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: non-delivery reports - again To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Folloup to my previous message: it looks like setting int ta_logs_diagnostics = 1; in transports/libta/diagnostic.c line 45 was a bit early: as far as I understand, transports do not file non-delivery *yet*, and the scheduler does not do it *already* with this setting. Setting this to zero makes things work as usual. Matti, am I missing something? Eugene From zmailer-owner@nic.funet.fi Wed Mar 12 21:54:24 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <3872-20818>; Wed, 12 Mar 1997 21:53:13 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17409-13389>; Wed, 12 Mar 1997 12:52:17 -0700 Date: Wed, 12 Mar 1997 12:52:30 -0700 () From: Daryle Tilroe To: zmailer@nic.funet.fi Subject: Re: Where are nondelivery reports? In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 12 Mar 1997, Eugene Crosser wrote: > Am I missing something or notaryreports are broken in .46? > > It seems that since I installed .46 (plus two patches) almost > all non-delivery reports stopped. Messages are delivered I too am noticing this effect. As well once in a while I am now getting 2 copies of email messages delivered. So far my brief troubleshooting has not turned up any pattern. Although the extra copy seems not to be created internally but rater due to the sending smtp connection sending 2 copies. It is as if the other machine thinks zmailer did not recieve the message properly the first time. SunOS 4.1.3 U1 on Sparc2, Zmailer 2.99.46+p1+p2 compiled with gcc. Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Thu Mar 13 18:48:48 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <4287-2168>; Thu, 13 Mar 1997 18:47:27 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-7292>; Thu, 13 Mar 1997 19:47:15 +0300 Date: Thu, 13 Mar 1997 19:47:13 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: is the mailing list working? To: zmailer@nic.funet.fi cc: mea@utu.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Is the list dead or did Majordomo forget about me (again!)? Yesterday I posted two messages about a bug in .46 to the list but did not receive them back. Eugene From zmailer-owner@nic.funet.fi Thu Mar 13 19:00:25 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <5035-8764>; Thu, 13 Mar 1997 18:59:17 +0200 Received: by utu.fi id <30984-2543>; Thu, 13 Mar 1997 18:58:31 +0200 Subject: Re: is the mailing list working? In-Reply-To: from Eugene Crosser at "Mar 13, 97 07:47:13 pm" From: Matti Aarnio To: crosser@online.ru Date: Thu, 13 Mar 1997 18:58:20 +0200 (EET) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 542 Message-Id: <19970313165831Z30984-2543+1582@utu.fi> Return-Path: > Is the list dead or did Majordomo forget about me (again!)? > Yesterday I posted two messages about a bug in .46 to the list > but did not receive them back. You haven't ? Hmm.. No Eugene (nor crosser) on the list ? The subscription server is my very old writeup, perhaps it needs debugging ? Anybody want to spend a moment ? ftp.funet.fi:/pub/local/src/mailserver.tar.gz I did receive the questions, though, but I am positively swamped... (if there is anything positive in that situation, that is...) > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Mar 14 13:18:47 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <4127-17007>; Fri, 14 Mar 1997 13:17:59 +0200 Received: by utu.fi id <31157-2543>; Fri, 14 Mar 1997 13:17:07 +0200 Subject: Re: is the mailing list working? In-Reply-To: from Eugene Crosser at "Mar 13, 97 08:46:49 pm" From: Matti Aarnio To: crosser@online.ru Date: Fri, 14 Mar 1997 13:16:58 +0200 (EET) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1528 Message-Id: <19970314111707Z31157-2543+2101@utu.fi> Return-Path: > > > Is the list dead or did Majordomo forget about me (again!)? > > > Yesterday I posted two messages about a bug in .46 to the list > > > but did not receive them back. > > > > You haven't ? Hmm.. No Eugene (nor crosser) on the list ? > > This already happend a year ago. I was subscribed and then stopped > receiving postings... Will you put me (crosser@online.ru) back or > shall I subscribe myself again? Please "help yourself" -- I try to avoid touching on the file manually -- it may cause surprises too.. > > The subscription server is my very old writeup, perhaps > > it needs debugging ? Anybody want to spend a moment ? > > ftp.funet.fi:/pub/local/src/mailserver.tar.gz > > Why not use SmartList? It is working in production here and I like it. If I have time to learn it. However that server does other things too, it can send files out from that servers archive in email, and that you don't get in any listserver... > > I did receive the questions, though, but I am positively > > swamped... (if there is anything positive in that situation, > > that is...) > > I know :) > BTW, nic.funet.fi does not generage non-deliveries too. I tried mailing > to some-unlikely-to-exist-name@nic.funet.fi and did not receive non-delivery > report from it. Yes, I spent couple early morning hours yesterday (at home, instead of sleeping) on it, and now have fixed ZMailer src at my office machine... Perhaps I can release it already today. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Mar 15 17:04:55 1997 Received: from mailrelay.tiac.net ([199.0.65.237]) by nic.funet.fi with ESMTP id <3681-20265>; Sat, 15 Mar 1997 17:04:42 +0200 Received: from tiac.net (p6.ts1.sprin.MA.tiac.com [206.119.231.167]) by mailrelay.tiac.net (8.8.5/) with SMTP id JAA21688; Sat, 15 Mar 1997 09:49:50 -0500 (EST) From: alanmako@tiac.net Message-Id: <199703151449.JAA21688@mailrelay.tiac.net> To: alanmako@tiac.net Date: Sat, 15 Mar 97 09:53:39 -0660 Sent: Sat, 15 Mar 97 09:53:39 -0660 Subject: Free Income Opportunity Info! X-Sender: X-Mailer: Extractor Pro v5.0 X-Sender: Extractor Pro v5.0 Return-Path: We believe that the following information may be of interest to you. If this is not the case, then we apologize for your inconvenience and ask that you simply delete this message. We do not keep a "remove" list because our bulk email software automatically keeps track of the email addresses we have already sent this message to. We have no inten- tion of sending another bulk email to this email address again. Thank- you and have a nice day! :-) Hello! May I have your permission to email you some more information? I'd like to email you more information about a fantastic income opportunity which pays you a multi-level commission for referring people to high-quality nutritional and skincare products. Simply respond to this email with the word YES in the subject heading, and We'll email you this information without cost or any further obligation. If you'd like to learn about dozens of outstanding nutritional and skincare products such as ones that have helped THOUSANDS of people to: * Lose or Gain Weight. (MILLIONS of people have a weight problem!) * Improve memory and concentration. (Great for students!) * Feel more alert and energetic. (Products that give you a natural boost which lasts hours longer than coffee!) * Reduce the risks of diabetes, heart disease and high blood pressure. * Reduce problems associated with arthritis, asthma and allergies. * Take minerals in collodial form (10 times more absorbable by your body than minerals taken in tablet form!) Then you want to respond "Yes" to this email! If you'd like to learn about an incredible income opportunity that can allow you to: * Start your own home-based business which you can work on a full or part-time basis that requires no "sign-up" fee or other hidden costs. * Earn a monthly royalty check of a few extra HUNDRED or THOUSANDS of dollars based on your efforts and the efforts of others. * Purchase dozens of different products at wholesale prices, (without any inventory or "warehousing" requirements.) * Earn an MLM commission of up to 25% (or more!) up to 6 levels (or more!) deep for referring other people to these fine products. * Build an INTERNATIONAL organization of people who use these products and refer other people to these products. (You can earn a MULTI-LEVEL commision on their efforts!) * EARN WHAT YOU'RE WORTH! Then you want to respond "Yes" to this email! Opportunity often only knocks once in a lifetime. Will you answer the door? WHAT HAVE YOU GOT TO LOSE BY RESPONDING "YES" TO THIS EMAIL???!!! Thank-you for taking the time to read this message! From zmailer-owner@nic.funet.fi Tue Mar 18 16:09:01 1997 Received: from pobox3.funet.fi ([128.214.46.64]) by nic.funet.fi with ESMTP id <4829-11737>; Tue, 18 Mar 1997 16:08:07 +0200 Received: from outbound.Princeton.EDU (outbound.Princeton.EDU [128.112.128.84]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with ESMTP id QAA00691 for ; Tue, 18 Mar 1997 16:05:33 +0200 (EET) Received: from yo.Princeton.EDU ([128.112.233.37]) by outbound.Princeton.EDU with ESMTP id <541121-20619>; Tue, 18 Mar 1997 09:08:00 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id JAA01793 for ; Tue, 18 Mar 1997 09:07:55 -0500 Message-Id: <199703181407.JAA01793@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: Issues for Solaris 2.5.1 and 2.99.45 Organization: Princeton University, Network Systems, Class. 56 X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 18 Mar 1997 09:07:55 EST From: "Michael R. Gettes" Return-Path: Solaris 2.5.1 Zmailer 2.99.45 A few problems. I tried pushing about 100,000 messages through it as my first acid test (real traffic). (1) Logging scheduler output I got 53MB of the following message type: 19970317192004 DBGdiag: # smtpclient:15563: lockaddr: file '541305-13635' host '?host?' expected '~' found ' ' (2) I tried setting USE_FCNTLLOCK in transports/libta/lockaddr.c. After futzing with the header files to get fcntl.h incorporated, this code did not build. Looking at the fcntl code in lockaddr.c, I don't think this code works. (3) I have 6 smtp processes looping, apparently. I am unable to truss or debug them. I sent them a SIGABRT and then gdb the core file and got: #0 0x148b0 in process (SS=0xeffff3c8, dp=0x3fb7c, smtpstatus=0, host=Cannot access memory at address 0xeffff184. ) at smtp.c:701 701 for (rp = rphead = dp->recipients; rp != NULL; rp = rp->next) { (4) HUPing the scheduler process does not cause new processes to write a new scheduler.perflog. Killing the scheduler process does not kill the children, as documented. HUPing the router process only causes that one router process to write to the new log file, the others continue writing to the old (it would seem they never get HUPed). (5) process.cf has a return 0 just prior to the log info: line that prevents the info: line from being generated. This, of course, breaks any stats gathering you might want. (6) I can't find any documentation about the output of mailq -Q. Can someone describe the following please?: smtp/mx1.gzic.gd.cn/0 R=9 A=3 P=22688 HA=1482s FA=1484s OF=9 QA=14h18m56s (7) What is considered to be the last 'stable' release of zmailer? Is anyone moving 200,000 messages per day on Solaris 2.5.1??? Thanks!! /mrg From zmailer-owner@nic.funet.fi Tue Mar 18 19:37:56 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <3272-12037>; Tue, 18 Mar 1997 19:36:51 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17412-19443>; Tue, 18 Mar 1997 10:35:47 -0700 Date: Tue, 18 Mar 1997 10:35:57 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: zmailer@nic.funet.fi Subject: Of error mail and incoming messages getting resent. Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I had brought up this problem in a previous message of messages getting sent more than once. However I had noticed no repeatability. I now have a message that the remote site continues to sent to me and I believe it is that same problem with Zmailer. I have also not seen any confirmed solutions to the lack of proper error mail creation. Is Mr Eugene Crosser's solution the correct one? Note that I had previously been using 99.11 for two years without any such problems. In any event this is the SMTP conversation of the message that keeps getting resent: 15084# connection from IDENT-FD2-FAULT-1-38@cc09ss.unity.ncsu.edu (port 3688) 15084w 220 amc-edm.amc.ab.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Tue, 18 Mar 1997 01:07:44 -0700 15084# remote from [152.1.1.224] 15084r EHLO cc09ss.unity.ncsu.edu 15084w 250-amc-edm.amc.ab.ca Hello cc09ss.unity.ncsu.edu 15084w 250-SIZE 0 15084w 250-8BITMIME 15084w 250-PIPELINING 15084w 250-CHUNKING 15084w 250-ENHANCEDSTATUSCODES 15084w 250-EXPN 15084w 250-VRFY 15084w 250-DSN 15084w 250-X-RCPTLIMIT 10000 15084w 250-X-TURNME 15084w 250-ETRN 15084w 250 HELP 15084r MAIL From: SIZE=3828 15084w 250 2.1.0 Sender syntax Ok 15084r RCPT To: 15084w 250 2.1.5 Ok; can accomodate 3828 byte message 15084r DATA 15084w 354 Start mail input; end with . 15084w 250 2.6.0 Roger 15084# 17420: 1233 bytes 15084r QUIT 15084w 221 2.0.0 amc-edm.amc.ab.ca Out Any suggestions/confirmations on this or the error mail problem are welcome. SunOS 4.1.3 U1 on Sparc2, Zmailer 2.99.46+p1+p2 compiled with gcc. Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Tue Mar 18 20:14:29 1997 Received: by nic.funet.fi id <3576-960>; Tue, 18 Mar 1997 20:13:41 +0200 Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Tue, 18 Mar 1997 20:13:37 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199703181407.JAA01793@yo.Princeton.EDU> from "Michael R. Gettes" at Mar 18, 97 09:07:55 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3806 Message-Id: <19970318181341Z3576-960+128@nic.funet.fi> Return-Path: > Solaris 2.5.1 > Zmailer 2.99.45 > > A few problems. I tried pushing about 100,000 messages through it > as my first acid test (real traffic). > > (1) Logging scheduler output I got 53MB of the following message type: > > 19970317192004 DBGdiag: # smtpclient:15563: lockaddr: file '541305-13635' host > '?host?' expected '~' found ' ' It means that two smtp-clients have been processing same message in parallel. It is a no-no situation actually. The presense of '?host?' propably means that you are running scheduler without sending host-selectors down to the TA programs. That is, you have the smtp/* defined in the ancient manner of: command="smtp $host" Umm.. No, it doesn't look like it from things you describe below. Hmm... > (2) I tried setting USE_FCNTLLOCK in transports/libta/lockaddr.c. After > futzing with the header files to get fcntl.h incorporated, this code > did not build. Looking at the fcntl code in lockaddr.c, I don't think > this code works. No, it does not work. I did never finish off the code. (And I am not certain it would be cheaper to the system, than the current method -- having HUNDREDS of small log regions in dozens of files...) > (3) I have 6 smtp processes looping, apparently. I am unable to truss or > debug them. I sent them a SIGABRT and then gdb the core file and got: > > #0 0x148b0 in process (SS=0xeffff3c8, dp=0x3fb7c, smtpstatus=0, host=Cannot > access memory at address 0xeffff184. > ) at smtp.c:701 > 701 for (rp = rphead = dp->recipients; rp != NULL; rp = rp->next) { ??? Perhaps a faulty transport-spec file ? > (4) HUPing the scheduler process does not cause new processes to write a new > scheduler.perflog. Killing the scheduler process does not kill the children, > as documented. HUPing the router process only causes that one router process > to write to the new log file, the others continue writing to the old (it would > seem they never get HUPed). - Scheduler HUP -- The logfn processing (of log file) is handled by the lib/loginit.c routine, which does not know about the perflog. - Router HUP -- You must HUP the GROUP, not single ("top parent") process that forked off some siblings to do paral- lelized routing - The children of the scheduler will die some day, once they are done with all the job-specs they have received. However smtp connection timeouts (for example) are not always very short... - For the 2.99.47 I have done some checking on SMTP timeouts, and how sure they are to happen. Perhaps I have now been able to improve the sureness a bit now. > (5) process.cf has a return 0 just prior to the log info: line that prevents > the info: line from being generated. This, of course, breaks any stats > gathering you might want. > > (6) I can't find any documentation about the output of mailq -Q. Can someone > describe the following please?: > smtp/mx1.gzic.gd.cn/0 R=9 A=3 P=22688 HA=1482s FA=1484s OF=9 QA=14h18m56s R = Recipient addresses in the thread A = Attempts done on this channel/host selector P = Process PID of the Transport Agent HA = Hunger Age -- when it last reported a need for "food" FA = Feed Age -- when a job was sent to the TA OF = OverFeed -- how many jobs are still unprocessed by the TA from those that were sent to it ? QA = QueueAge -- Age of the oldest message in the thread > (7) What is considered to be the last 'stable' release of zmailer? Is anyone > moving 200,000 messages per day on Solaris 2.5.1??? I have 2.99.46p3 at mailhost.utu.fi, though I don't know how many mails it processes daily -- fairly many.. .. and no, I would not call 46p3 STABLE, for that matter all of 46 are somewhat buggy. > Thanks!! > > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Mar 18 21:13:32 1997 Received: from outbound.Princeton.EDU ([128.112.128.84]) by nic.funet.fi with ESMTP id <4746-12037>; Tue, 18 Mar 1997 21:13:00 +0200 Received: from yo.Princeton.EDU ([128.112.233.37]) by outbound.Princeton.EDU with ESMTP id <541121-20622>; Tue, 18 Mar 1997 14:12:50 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id OAA28966 for ; Tue, 18 Mar 1997 14:12:37 -0500 Message-Id: <199703181912.OAA28966@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes In-reply-to: Your message of Tue, 18 Mar 1997 20:13:37 +0200. <19970318181341Z3576-960+128@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 18 Mar 1997 14:12:36 EST From: "Michael R. Gettes" Return-Path: On Tue, 18 Mar 1997 20:13:37 +0200 mea said: > > Solaris 2.5.1 > > Zmailer 2.99.45 > > > > A few problems. I tried pushing about 100,000 messages through it > > as my first acid test (real traffic). > > > > (1) Logging scheduler output I got 53MB of the following message type: > > > > 19970317192004 DBGdiag: # smtpclient:15563: lockaddr: file '541305-13635' host > > '?host?' expected '~' found ' ' > > It means that two smtp-clients have been processing same message > in parallel. It is a no-no situation actually. > > The presense of '?host?' propably means that you are running > scheduler without sending host-selectors down to the TA programs. > That is, you have the smtp/* defined in the ancient manner of: > command="smtp $host" > > Umm.. No, it doesn't look like it from things you describe below. > Hmm... > As you figured out... I am not specifying "smtp $host". I am doing "smtp -e -s". > > (3) I have 6 smtp processes looping, apparently. I am unable to truss or > > debug them. I sent them a SIGABRT and then gdb the core file and got: > > > > #0 0x148b0 in process (SS=0xeffff3c8, dp=0x3fb7c, smtpstatus=0, host=Cannot > > access memory at address 0xeffff184. > > ) at smtp.c:701 > > 701 for (rp = rphead = dp->recipients; rp != NULL; rp = rp->next) { > > ??? Perhaps a faulty transport-spec file ? Ok. Not sure where I should look for this... Any help is appreciated. > > > (4) HUPing the scheduler process does not cause new processes to write a new > > scheduler.perflog. Killing the scheduler process does not kill the children, > > as documented. HUPing the router process only causes that one router process > > to write to the new log file, the others continue writing to the old (it would > > seem they never get HUPed). > > - Scheduler HUP -- The logfn processing (of log file) is handled by > the lib/loginit.c routine, which does not know > about the perflog. I presume I should disable the perflog then... since it would seem that there aren't any programs to grok this information for useful reporting, yes? I guess I would point the perflog and /dev/null so I can get the nice -Q output, yes? > > - Router HUP -- You must HUP the GROUP, not single ("top parent") > process that forked off some siblings to do paral- > lelized routing How is this done? Sorry for such a stupid question. > > - The children of the scheduler will die some day, once they are done > with all the job-specs they have received. However smtp connection > timeouts (for example) are not always very short... Can there be an option to the scheduler which says that it should send a SIGTERM or SIGKILL to all children??? Then I can reliably roll over the scheduler log file. > > - For the 2.99.47 I have done some checking on SMTP timeouts, and > how sure they are to happen. Perhaps I have now been able to > improve the sureness a bit now. good. > > (7) What is considered to be the last 'stable' release of zmailer? Is anyone > > moving 200,000 messages per day on Solaris 2.5.1??? > > I have 2.99.46p3 at mailhost.utu.fi, though I don't know > how many mails it processes daily -- fairly many.. > .. and no, I would not call 46p3 STABLE, for that matter > all of 46 are somewhat buggy. Is there a previous version I should go to instead?? /mrg From zmailer-owner@nic.funet.fi Tue Mar 18 22:42:10 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <4898-960>; Tue, 18 Mar 1997 22:40:55 +0200 Received: by utu.fi id <31476-2543>; Tue, 18 Mar 1997 22:40:04 +0200 Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 In-Reply-To: <199703181912.OAA28966@yo.Princeton.EDU> from "Michael R. Gettes" at "Mar 18, 97 02:12:36 pm" From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Tue, 18 Mar 1997 22:40:03 +0200 (EET) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 3900 Message-Id: <19970318204004Z31476-2543+4436@utu.fi> Return-Path: > As you figured out... I am not specifying "smtp $host". I am doing "smtp -e > -s". Ah! Leave away the '-e' -- it causes the smtp transporter to process also messages it is not supposed to touch. (Well, '-e' did make sense before I rewrote the scheduler.) > > > (3) I have 6 smtp processes looping, apparently. I am unable to truss or > > > debug them. I sent them a SIGABRT and then gdb the core file and got: > > > > > > #0 0x148b0 in process (SS=0xeffff3c8, dp=0x3fb7c, smtpstatus=0, host=Cannot > > > access memory at address 0xeffff184. > > > ) at smtp.c:701 > > > 701 for (rp = rphead = dp->recipients; rp != NULL; rp = rp->next) { > > > > ??? Perhaps a faulty transport-spec file ? > > Ok. Not sure where I should look for this... Any help is appreciated. I suspect the "-e" is the reason for this too. If it recurs, then at the main() level of the smtp transporter there are variables "channel", "host", and "file", which may contain the clue.. > > > (4) HUPing the scheduler process does not cause new processes to write a new > > > scheduler.perflog. Killing the scheduler process does not kill the children, > > > as documented. HUPing the router process only causes that one router process > > > to write to the new log file, the others continue writing to the old (it would > > > seem they never get HUPed). > > > > - Scheduler HUP -- The logfn processing (of log file) is handled by > > the lib/loginit.c routine, which does not know > > about the perflog. > > I presume I should disable the perflog then... since it would seem > that there aren't any programs to grok this information for useful reporting, > yes? I guess I would point the perflog and /dev/null so I can get the > nice -Q output, yes? Just leave away the '-l scheduler.perflog' -option from the SCEDULEROPTIONS. The 2.99.47 will do 'SIGHUP' properly on them both. > > > > - Router HUP -- You must HUP the GROUP, not single ("top parent") > > process that forked off some siblings to do paral- > > lelized routing > > How is this done? Sorry for such a stupid question. I am tempted to say: "man 2 kill", but... routerpid=`head -1 $POSTOFFICE/.pid.router` kill -HUP -$routerpid That is, the listed .pid.router is the pid of the process-group leader. > > > > - The children of the scheduler will die some day, once they are done > > with all the job-specs they have received. However smtp connection > > timeouts (for example) are not always very short... > > Can there be an option to the scheduler which says that it should send > a SIGTERM or SIGKILL to all children??? Then I can reliably roll over > the scheduler log file. The logfiles should be opened with close-on-exec flags.. ... not that it should matter. Child processes have only FDs 0,1/2 open for them, and they are pipes from the scheduler. > > - For the 2.99.47 I have done some checking on SMTP timeouts, and > > how sure they are to happen. Perhaps I have now been able to > > improve the sureness a bit now. > > good. I have tried also to do some additional HighAvailablility things there -- we want to use ZMailer on a H-A system on which the system components backup each other, but it becomes rather hairy when it comes to the MX processing, and detection of which IP address in system is for what kind of use... > > > (7) What is considered to be the last 'stable' release of zmailer? Is anyone > > > moving 200,000 messages per day on Solaris 2.5.1??? > > > > I have 2.99.46p3 at mailhost.utu.fi, though I don't know > > how many mails it processes daily -- fairly many.. > > .. and no, I would not call 46p3 STABLE, for that matter > > all of 46 are somewhat buggy. > > Is there a previous version I should go to instead?? I would say 2.99.45 -- except that you have problems with it.. > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Mar 18 23:32:04 1997 Received: by nic.funet.fi id <5062-12404>; Tue, 18 Mar 1997 23:30:35 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: Partial IPv6 support coming up on 2.99.47 ZMailer Cc: netdev@roxanne.nuclecu.unam.mx Message-Id: <19970318213035Z5062-12404+156@nic.funet.fi> Date: Tue, 18 Mar 1997 23:30:29 +0200 Return-Path: Hello, ( Cc: netdev@roxanne.nuclecu.unam.mx -- Linux network code wizarding with interests also on IPv6 ) I started my local ZMailer smtpserver with options: -a6s ve -l smtp.log and this is what it reported, when I telneted to the local port: 00000# started server pid 5599 at Tue, 18 Mar 1997 20:42:12 +0200 5602# connection from IDENT-NONSENSE@[ipv6.::ffff:194.251.252.33] (port 40736) 5602w 220 mea.tmt.tele.fi ZMailer Server 2.99.47 #1 ESMTP+IDENT ready at Tue, 18 Mar 1997 20:42:22 +0200 5602# remote from [ipv6.::ffff:194.251.252.33] 5602r quit 5602w 221 2.0.0 mea.tmt.tele.fi Out Interestingly the server does listen on IPv6 port, but it will transparently map back to IPv4 services at the same machine. Now to figure out when to query for IPv6 AAAA -records from the DNS, and when to use plain ordinary A-records so that the SMTP client can do its tricks properly. (Oh yes, using IP-address-literal trick I can already send to an "IPv6" address, but it is kludgeish: nobody@[ipv6.::ffff:1.2.3.4] ) ... and somebody needs to define IPv6 version of the ident (aka "auth") protocol. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 19 00:10:07 1997 Received: by nic.funet.fi id <4159-23773>; Wed, 19 Mar 1997 00:09:10 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: version 2.99.47 released... Message-Id: <19970318220910Z4159-23773+1@nic.funet.fi> Date: Wed, 19 Mar 1997 00:09:03 +0200 Return-Path: ... after long delay... I have now made available new release which I use at my workstation, and which does compile at DEC UNIX 4.0 (nic.funet.fi). It has initial IPv6 support in, and can be coerced to receive IPv6 (and IPv4-tunneled) SMTP sessions by adding option "-6" to the startup parameters of the smtpserver. Hmm.. Apparently I still have not addressed ( = debugged ) the problem that HP-UX machines have with the lattest ZMailer mailboxes. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 19 00:24:08 1997 Received: by nic.funet.fi id <4551-14871>; Wed, 19 Mar 1997 00:23:24 +0200 Subject: Re: Of error mail and incoming messages getting resent. From: Matti Aarnio To: data@amc.ab.ca Date: Wed, 19 Mar 1997 00:23:20 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Daryle Tilroe" at Mar 18, 97 10:35:57 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2114 Message-Id: <19970318222324Z4551-14871+8@nic.funet.fi> Return-Path: > I had brought up this problem in a previous message of messages getting > sent more than once. However I had noticed no repeatability. I now have > a message that the remote site continues to sent to me and I believe it is > that same problem with Zmailer. Remote site continues to send ?? Propably there is some SMTP transaction sync lock loss ? > I have also not seen any confirmed solutions to the lack of proper error > mail creation. Is Mr Eugene Crosser's solution the > correct one? 2.99.47 fixes the missing error report messages > Note that I had previously been using 99.11 for two years without any > such problems. > > In any event this is the SMTP conversation of the message that keeps > getting resent: > > 15084# connection from IDENT-FD2-FAULT-1-38@cc09ss.unity.ncsu.edu (port > 3688) > 15084w 220 amc-edm.amc.ab.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready > at Tue, 18 Mar 1997 01:07:44 -0700 > 15084# remote from [152.1.1.224] It is a machine with sendmail-8.8.4. Are you sure the message in question is received repeatedly from that remote machine ? If so, then there definitely is problem in between the machines. (Or inside the remote one.) > 15084r EHLO cc09ss.unity.ncsu.edu > 15084w 250-amc-edm.amc.ab.ca Hello cc09ss.unity.ncsu.edu > 15084w 250 HELP > 15084r MAIL From: SIZE=3828 > 15084w 250 2.1.0 Sender syntax Ok > 15084r RCPT To: > 15084w 250 2.1.5 Ok; can accomodate 3828 byte message > 15084r DATA > 15084w 354 Start mail input; end with . > 15084w 250 2.6.0 Roger > 15084# 17420: 1233 bytes > 15084r QUIT > 15084w 221 2.0.0 amc-edm.amc.ab.ca Out > > Any suggestions/confirmations on this or the error mail problem are > welcome. > > SunOS 4.1.3 U1 on Sparc2, Zmailer 2.99.46+p1+p2 compiled with gcc. Hmm.. apparently I never published my patch3 for that.. > Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca > Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 19 00:44:26 1997 Received: from outbound.Princeton.EDU ([128.112.128.84]) by nic.funet.fi with ESMTP id <4690-23773>; Wed, 19 Mar 1997 00:43:52 +0200 Received: from yo.Princeton.EDU ([128.112.233.37]) by outbound.Princeton.EDU with ESMTP id <541121-20620>; Tue, 18 Mar 1997 17:43:42 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id RAA18975 for ; Tue, 18 Mar 1997 17:43:41 -0500 Message-Id: <199703182243.RAA18975@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes In-reply-to: Your message of Tue, 18 Mar 1997 22:40:03 +0200. <19970318204004Z31476-2543+4436@utu.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 18 Mar 1997 17:43:40 EST From: "Michael R. Gettes" Return-Path: On Tue, 18 Mar 1997 22:40:03 +0200 mea said: > > > > > > - Router HUP -- You must HUP the GROUP, not single ("top parent") > > > process that forked off some siblings to do paral- > > > lelized routing > > > > How is this done? Sorry for such a stupid question. > > I am tempted to say: "man 2 kill", but... > > routerpid=`head -1 $POSTOFFICE/.pid.router` > kill -HUP -$routerpid > > That is, the listed .pid.router is the pid of the process-group leader. > Well, I guess I did not make myself clear... I did this and it only the group leader began logging to the new file. And thanks for the quick replies... much appreciated. /mrg From zmailer-owner@nic.funet.fi Wed Mar 19 00:49:23 1997 Received: by nic.funet.fi id <3914-23813>; Wed, 19 Mar 1997 00:48:27 +0200 Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Wed, 19 Mar 1997 00:48:22 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199703182243.RAA18975@yo.Princeton.EDU> from "Michael R. Gettes" at Mar 18, 97 05:43:40 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 356 Message-Id: <19970318224827Z3914-23813+9@nic.funet.fi> Return-Path: > Well, I guess I did not make myself clear... I did this and it only the group > leader began logging to the new file. Uh.. Then perhaps the detach() routine does not work properly, and does not aquire real process-group/session leadership for the leader process ? > And thanks for the quick replies... much appreciated. > > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 19 13:54:30 1997 Received: from atol.icm.edu.pl ([148.81.209.1]) by nic.funet.fi with SMTP id <4730-23773>; Wed, 19 Mar 1997 13:53:40 +0200 Received: from palma.icm.edu.pl (palma.icm.edu.pl [148.81.208.142]) by atol.icm.edu.pl (8.6.12/8.6.9) with ESMTP id MAA10974 for ; Wed, 19 Mar 1997 12:50:00 +0100 Received: from krzak.icm.edu.pl (krzak.icm.edu.pl [148.81.208.169]) by palma.icm.edu.pl (8.6.12/8.6.12) with ESMTP id MAA10949 for ; Wed, 19 Mar 1997 12:49:59 +0100 Received: (from jwr@localhost) by krzak.icm.edu.pl (8.7.5/8.7.3) id MAA00197; Wed, 19 Mar 1997 12:49:59 +0100 (MET) Sender: jwr@icm.edu.pl To: zmailer@nic.funet.fi Subject: How do fqdnaliases work ? X-Operating-System: SunOS 5.5 Generic Mime-Version: 1.0 (generated by tm-edit 7.85) Content-Type: text/plain; charset=US-ASCII From: Jan Rychter Date: 19 Mar 1997 12:49:58 +0100 Message-ID: Lines: 12 X-Mailer: Red Gnus v0.44/XEmacs 19.14 Return-Path: I'm using 2.99.46. How do I use the fqdnaliases ? Say, I want to create an alias: someuser@thisdomain.com: otheruser@otherdomain.com Do I add thisdomain.com to localnames ? Simply adding the alias as above to fqdnaliases and running newfqdnaliases doesn't seem to do anything on my machine. Or do I need to flag something at compile-time, too? thanks, --J. From zmailer-owner@nic.funet.fi Wed Mar 19 20:13:40 1997 Received: from lupinella.troll.no ([195.0.254.19]) by nic.funet.fi with ESMTP id <4850-23813>; Wed, 19 Mar 1997 20:13:06 +0200 Received: by troll.no id <45686-21930>; Wed, 19 Mar 1997 19:13:01 +0100 Sender: agulbra@troll.no To: zmailer@nic.funet.fi Subject: 2.99.47? From: Arnt Gulbrandsen Date: 19 Mar 1997 19:12:59 +0100 Message-ID: Lines: 36 X-Mailer: Gnus v5.2.25/XEmacs 19.14 Return-Path: I'm setting up another zmailer box, and wouldn't like to do it _just_ before 2.99.47 is released. Matti? Btw, an anecdote for some of you. I just sent NOTIFY=SUCCESS on a mailing-list message (to probe for address changes) and got this thing back from Microsoft. I looked at the uuencoded blob, and, yes: all the relevant information is there, in some proprietary, unreadable binary format. DSN is a standard, so of course Microsoft supports it. --Arnt From: System Administrator Subject: Delivered: To: owner-frc@troll.no Your message was delivered to the following recipient(s): Bruce Jones on Wed, 19 Mar 1997 09:42:37 -0800 MSEXCH:MSExchangeMTA:northamerica:RED-103-MSG begin 600 winmail.dat M>)\^(BH1`0:0"``$```````!``$``0>0!@`(````Y`0```````#H``$(@`<` M&````$E032Y-:6-R;W-O9G0@36%I;"Y.;W1E`#$(`0V`!``"`````@`"``$) M@`$`(0```#%",T0T-SA$-T1!,$0P,3$Y.34Y,#`X,#5&1#0P03(X``D'`0.0 M!@```0``"@````,`-@``````0``',+``F_",-+P!"P`I```````+`",````` M``,`!A`+#[;_`P`'$(4````#`!`0``````,`$1``````'@`($`$```!E```` M64]54DU%4U-!1T5705-$14Q)5D5214143U1(149/3$Q/5TE.1U)%0TE0245. M5"A3*3I"4E5#14I/3D533TY7140L,3E-05(Q.3DW,#DZ-#(Z,S; Wed, 19 Mar 1997 23:13:10 +0200 Received: from amc-edm.amc.ab.ca (amc-edm.amc.ualberta.ca [129.128.50.68]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with ESMTP id XAA17401 for ; Wed, 19 Mar 1997 23:10:33 +0200 (EET) Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17427-2765>; Wed, 19 Mar 1997 14:10:55 -0700 Date: Wed, 19 Mar 1997 14:11:08 -0700 () From: Daryle Tilroe To: zmailer@nic.funet.fi Subject: Trying to compile 2.99.47 under SunOS 4.1.3_U1 with gcc Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Changes have been made in this release to dnsgetrr.c and dnsgetrr.h mostly for IPv6 from what I can see. I am now unable to compile due to int32_t not being properly defined. I have hacked around a bit but would rather have the proper solution if it is available. Thanks. Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Thu Mar 20 00:13:59 1997 Received: from outbound.Princeton.EDU ([128.112.128.84]) by nic.funet.fi with ESMTP id <3246-12458>; Thu, 20 Mar 1997 00:13:15 +0200 Received: from yo.Princeton.EDU ([128.112.233.37]) by outbound.Princeton.EDU with ESMTP id <541415-29675>; Wed, 19 Mar 1997 17:12:38 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id RAA22031 for ; Wed, 19 Mar 1997 17:12:30 -0500 Message-Id: <199703192212.RAA22031@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: Re: Issues for Solaris 2.5.1 and 2.99.45 Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes In-reply-to: Your message of Wed, 19 Mar 1997 00:48:22 +0200. <19970318224827Z3914-23813+9@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 19 Mar 1997 17:12:29 EST From: "Michael R. Gettes" Return-Path: Just an update on the problems I posted. After following Matti's advice and removing the -e option to the smtp transport as well as removing the stats logging on the scheduler, all is well. Actually, things are better than well. Zmailer is moving traffic with great ease and is quite fast even on a small Sparc 5. We are already pushing about 150,000 recipients and my guesstimate for max thruput is around 800,000 recipients. Very nice!!!!! This is exceeding my expectations. Nice work! /mrg From zmailer-owner@nic.funet.fi Thu Mar 20 16:25:33 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <4860-28165>; Thu, 20 Mar 1997 16:24:20 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-322>; Thu, 20 Mar 1997 17:24:10 +0300 Date: Thu, 20 Mar 1997 17:24:03 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: smtpserver segfault To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Today I got a core from smtpserver. I happends that sometimes (?) reporterr is called with NULL first parameter (SmtpState *SS). Then syslog() call in line 939 segfaults. Fix is obvious. Matti? Eugene Core was generated by `/usr/zmailer/bin/smtpserver -asve -8X -l /var/log/zmailer/smtpserver'. Program terminated with signal 11, Segmentation fault. procfs (find_procinfo): Couldn't locate pid 0 #0 0x157a8 in reporterr (SS=0x0, tell=0, msg=0x33e70 "SMTP protocol timed out") at smtpserver.c:939 939 syslog(LOG_ERR, (gdb) bt #0 0x157a8 in reporterr (SS=0x0, tell=0, msg=0x33e70 "SMTP protocol timed out") at smtpserver.c:939 #1 0x156f8 in timedout (sig=14) at smtpserver.c:900 #2 0xef5f8c84 in _end () #3 0x15830 in s_getc (SS=0xefffb5a0) at smtpserver.c:963 #4 0x1aae4 in mvdata (SS=0xefffb5a0, msg=0x0) at smtpserver.c:3286 #5 0x178d4 in smtp_data (SS=0xefffb5a0, buf=0xefff93c0 "DATA", cp=0x0) at smtpserver.c:1802 #6 0x18dcc in smtpserver (SS=0xefffb5a0, insecure=1) at smtpserver.c:2343 #7 0x155f4 in main (argc=8, argv=0xefffb5a0) at smtpserver.c:812 (gdb) From zmailer-owner@nic.funet.fi Thu Mar 20 16:36:44 1997 Received: from lists.Princeton.EDU ([128.112.129.249]) by nic.funet.fi with ESMTP id <4408-17443>; Thu, 20 Mar 1997 16:36:09 +0200 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370183.s2-1>; Thu, 20 Mar 1997 09:35:59 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id JAA17331 for ; Thu, 20 Mar 1997 09:35:56 -0500 Message-Id: <199703201435.JAA17331@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: lookups in router not wanted Organization: Princeton University, Network Systems, Class. 56 X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 20 Mar 1997 09:35:55 EST From: "Michael R. Gettes" Return-Path: I have a configuration problem. I have set toplevels= all toplevel domains. From my testing, if I give an address that cannot be located via DNS (like yzhou@MECAD.HUST.EDU.CN) the router does not get hung in looking it up DNS. However, when in production, I am getting search_res: deferred: mecad.hust.edu.cn: any (host name lookup/try again) error How I tested is shown below... if my test is invalid, let me know my error. I have hundreds of messages going to this address so the router is getting blocked and now I have a MAJOR backlog of mail. I am not sure how to get out of this. Any help or pointers is appreciated. I am using 2.99.45 without modifications to the .cf files except as noted above. /mrg P.S. I tested with the following: [2198]su.outbound.Princeton.EDU--> router -i ZMailer router (2.99.45 #1: Tue Mar 11 21:10:19 EST 1997) gettes@outbound.Princeton.EDU:/var/local/c/zmailer-2.99.45/router Copyright 1992 Rayan S. Zachariassen Copyright 1992-1996 Matti Aarnio z# . router.cf relation: expansions is already a defined database! relation: hostexpansions is already a defined database! relation: recipients is already a defined database! relation: istoplevel is already a defined database! relation: thishost is already a defined database! relation: routesdb is already a defined database! relation: canon is already a defined database! relation: mxhosts is already a defined database! relation: ipaddress is already a defined database! relation: revaddress is already a defined database! relation: aliases is already a defined database! zz# router yzhou@MECAD.HUST.EDU.CN (((smtp MECAD.HUST.EDU.CN yzhou@MECAD.HUST.EDU.CN default_attributes))) zz# From zmailer-owner@nic.funet.fi Thu Mar 20 20:53:21 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <4182-12458>; Thu, 20 Mar 1997 20:49:37 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-24585>; Thu, 20 Mar 1997 21:49:22 +0300 Date: Thu, 20 Mar 1997 21:49:13 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Re: Trying to compile 2.99.47 under SunOS 4.1.3_U1 with gcc To: Daryle Tilroe cc: zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > Changes have been made in this release to dnsgetrr.c and dnsgetrr.h mostly > for IPv6 from what I can see. I am now unable to compile due to int32_t > not being properly defined. I have hacked around a bit but would rather Same here at Solaris/SPARC 2.5.1 On Soalris, there is no int32_t. There is uint32_t defined in some strange includes like pthread.h but not in normal. I changed it to long as a temporary measure. Then, there is gethostbyaddr_r() and it has different number of parameters than that in dnsgetrr.h -> they would conflict. Is this function used anywhere in Zmailer, anyway? I did not find any references. Also, you need to #include into dnsgetrr.c Also, there are both "makefile" and "Makefile" in lib/ subdirectory, and my make prefers "makefile" that is not updated during configure. When compiled it at last and made install-bin and started, it accepted incoming mail by SMTP, but then the router wrote to the syslog: routing deferred, and in its own log: router: rename(68041-24552,../deferred/68041.ctrlfile): No such file or directory As far as I can tell, these messages where lost... Eugene From zmailer-owner@nic.funet.fi Thu Mar 20 23:53:54 1997 Received: by nic.funet.fi id <3644-28165>; Thu, 20 Mar 1997 23:46:25 +0200 Subject: Re: smtpserver segfault From: Matti Aarnio To: crosser@online.ru Date: Thu, 20 Mar 1997 23:44:14 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Mar 20, 97 05:24:03 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 386 Message-Id: <19970320214625Z3644-28165+205@nic.funet.fi> Return-Path: > Today I got a core from smtpserver. I happends that sometimes (?) > reporterr is called with NULL first parameter (SmtpState *SS). > Then syslog() call in line 939 segfaults. Fix is obvious. Matti? Actually it wasn't obvious. You don't mention the version you are using, but I think it was 2.99.45 ? I did rewrite the timeout processing for 2.99.47. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Mar 21 00:39:50 1997 Received: by nic.funet.fi id <4583-25589>; Fri, 21 Mar 1997 00:33:59 +0200 Subject: Re: lookups in router not wanted From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Fri, 21 Mar 1997 00:29:43 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199703201435.JAA17331@yo.Princeton.EDU> from "Michael R. Gettes" at Mar 20, 97 09:35:55 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2456 Message-Id: <19970320223359Z4583-25589+3@nic.funet.fi> Return-Path: > I have a configuration problem. > > I have set toplevels= all toplevel domains. From my testing, if I give > an address that cannot be located via DNS (like yzhou@MECAD.HUST.EDU.CN) > the router does not get hung in looking it up DNS. However, when > in production, I am getting ... Odd.. The way I do my tests is following: # $MAILBIN/router -i ... z# trace all z# untrace regexp rfc822 z# router whatever@where.ever.CN You will see that there happens 'istoplevel' test, and that it does the test in lower case. I hope your list of toplevel domains is in lower-case too... (My samples have it in lowercase -- TELE-FI.cf, for example.) The output is rather lengthy, but most informative :-) Another test I do is to turn off the regular routers, send in a couple messages to the test address, and then run router with trace options turned on -- all except regexp.. $MAILBIN/router -i ... z# trace all z# untrace regexp z# cd $POSTOFFICE/router (now the difficulty: know which spool-id it is...) z# process 12345 and watch thru the reports.... (Your xterm is better to have HUGE scroll-back buffer ;) ) > search_res: deferred: mecad.hust.edu.cn: any (host name lookup/try again) error > > How I tested is shown below... if my test is invalid, let me know my error. > > I have hundreds of messages going to this address so the router is > getting blocked and now I have a MAJOR backlog of mail. I am not sure > how to get out of this. Any help or pointers is appreciated. You could define MAILVAR/db/routes entry: .cn smtp! > I am using 2.99.45 without modifications to the .cf files except as noted > above. > > /mrg > > P.S. I tested with the following: > > [2198]su.outbound.Princeton.EDU--> router -i > ZMailer router (2.99.45 #1: Tue Mar 11 21:10:19 EST 1997) > gettes@outbound.Princeton.EDU:/var/local/c/zmailer-2.99.45/router > Copyright 1992 Rayan S. Zachariassen > Copyright 1992-1996 Matti Aarnio > > z# . router.cf No need to re-include it. It will be loaded in at the start of the router process -- presuming this router.cf is the one that is on MAILSHARE/router.cf > relation: expansions is already a defined database! > relation: hostexpansions is already a defined database! > relation: recipients is already a defined database! ... > zz# router yzhou@MECAD.HUST.EDU.CN > (((smtp MECAD.HUST.EDU.CN yzhou@MECAD.HUST.EDU.CN default_attributes))) > zz# Double-z ? Hmm.. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Mar 21 14:31:47 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <3840-11798>; Fri, 21 Mar 1997 14:29:57 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-24584>; Fri, 21 Mar 1997 15:29:40 +0300 Date: Fri, 21 Mar 1997 15:29:34 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Re: smtpserver segfault To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970320214625Z3644-28165+205@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > Today I got a core from smtpserver. I happends that sometimes (?) > > reporterr is called with NULL first parameter (SmtpState *SS). > > Then syslog() call in line 939 segfaults. Fix is obvious. Matti? > > Actually it wasn't obvious. You don't mention the version > you are using, but I think it was 2.99.45 ? 2.99.46+p1+p2 + my temp.fix for non-delivery reports. I cannot switch to .47 yet as it would not work for me... Eugene From zmailer-owner@nic.funet.fi Fri Mar 21 18:18:04 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <4385-11798>; Fri, 21 Mar 1997 18:16:57 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17409-27733>; Fri, 21 Mar 1997 09:15:46 -0700 Date: Fri, 21 Mar 1997 09:15:59 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: zmailer@nic.funet.fi Subject: Stable version? In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I have been trying to upgrade our mailserver to the lastest version of zmailer (2.99.46+p1+p2 and now 2.99.47) and am running into fairly bothersome problems either compling or running both. Since I need to get back to normal and fight other fires fairly soon I would like to know what is a stable release that most people are running? Over the last two years I had been rather happily running 2.99.11 and I am wondering if I need to go back that far? Alternatively (and I guess this question is for Matti) when do you see coming out with a patch for the 2.99.47 problems? BTW I hope this does not appear petulant or ranting. I like zmailer quite a bit and am very happy with the development being done on it. It is just that I run a system for 100 heavy email users and I need to be running fairly reliable code. Thanks! Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Fri Mar 21 22:14:43 1997 Received: by nic.funet.fi id <3915-11798>; Fri, 21 Mar 1997 22:13:03 +0200 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: Patch for zmailer 2.99.47 -- solaris works :) Message-Id: <19970321201303Z3915-11798+164@nic.funet.fi> Date: Fri, 21 Mar 1997 22:13:01 +0200 Return-Path: Hello, This patch gets Solaris machines back on track -- and many others.. Frequent problem was an error in lib/esyslog.c: erename(), which lost a "return 0;" at some point -- and then spodarically started to cause problems... This patch also generates changes to the configure script, which may cause problems to you.. Thus pick ALSO file: zmailer-2.99.47.patch1.configure.gz and once you have expanded it AFTER you have applied these patches, touch it! (To make sure its date is latter than any of the other files..) Oh yes, also remember to remove "lib/makefile" which is spurious remnant from I don't know what... /Matti Aarnio diff -u -r zmailer-2.99.47/config.h.in zmailer-2.99.47b2/config.h.in --- zmailer-2.99.47/config.h.in Tue Mar 18 17:16:59 1997 +++ zmailer-2.99.47b2/config.h.in Fri Mar 21 03:43:19 1997 @@ -388,11 +388,14 @@ #undef HAVE_RENAME #undef HAVE_SETRLIMIT #undef HAVE_GETDTABLESIZE -#undef HAVE_NETINET_IN_H #undef HAVE_DUP2 #undef HAVE_SYSCONF #undef HAVE_FPATHCONF #undef HAVE_WRITEV +#undef HAVE_NETINET_IN_H +/* Sigh.... Linux 2.1.x series with IPv6 */ +#undef HAVE_NETINET6_IN6_H +#undef HAVE_LINUX_IN6_H /* or ? Pre or post ANSI-C ? */ #undef HAVE_VARARGS_H Only in zmailer-2.99.47: config.status Only in zmailer-2.99.47b2: configure diff -u -r zmailer-2.99.47/configure.in zmailer-2.99.47b2/configure.in --- zmailer-2.99.47/configure.in Tue Mar 18 23:19:38 1997 +++ zmailer-2.99.47b2/configure.in Fri Mar 21 03:42:59 1997 @@ -667,6 +667,7 @@ sys/param.h sys/statfs.h sys/fstyp.h mnttab.h mntent.h utime.h \ sys/statvfs.h sys/vfs.h sys/mount.h sys/filsys.h sys/fs_types.h) AC_CHECK_HEADERS(netinet/in.h sys/utsname.h maillock.h) +AC_CHECK_HEADERS(netinet6/in6.h linux/in6.h) AC_CHECK_HEADERS(varargs.h stdarg.h) AC_CHECK_FUNCS(flock lockf lstat fsync) AC_SUBST(LIBMAIL) diff -u -r zmailer-2.99.47/include/dnsgetrr.h zmailer-2.99.47b2/include/dnsgetrr.h --- zmailer-2.99.47/include/dnsgetrr.h Wed Mar 12 10:29:13 1997 +++ zmailer-2.99.47b2/include/dnsgetrr.h Thu Mar 20 23:12:57 1997 @@ -17,4 +17,3 @@ extern int getrrtype __((char *, int *, int, int, int, FILE *)); extern struct hostent *gethostbyname2_r __((const char *, int, struct dnsresult *)); extern struct hostent *gethostbyaddr_r __((const char *, int, int, struct dnsresult *)); - diff -u -r zmailer-2.99.47/lib/esyslib.c zmailer-2.99.47b2/lib/esyslib.c --- zmailer-2.99.47/lib/esyslib.c Mon Feb 17 02:57:40 1997 +++ zmailer-2.99.47b2/lib/esyslib.c Fri Mar 21 21:34:34 1997 @@ -232,6 +232,7 @@ return -1; } #endif /* !HAVE_RENAME */ + return 0; } int Only in zmailer-2.99.47b2/lib: esyslib.c~ Only in zmailer-2.99.47/lib: makefile diff -u -r zmailer-2.99.47/lib/selfaddrs.c zmailer-2.99.47b2/lib/selfaddrs.c --- zmailer-2.99.47/lib/selfaddrs.c Tue Mar 18 21:58:39 1997 +++ zmailer-2.99.47b2/lib/selfaddrs.c Fri Mar 21 03:48:13 1997 @@ -36,7 +36,16 @@ #include #endif #include +#ifndef EAI_AGAIN +# include "netdb6.h" /* IPv6 API stuff */ +#endif #include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #include #include diff -u -r zmailer-2.99.47/libc/getaddrinfo.c zmailer-2.99.47b2/libc/getaddrinfo.c --- zmailer-2.99.47/libc/getaddrinfo.c Tue Mar 18 23:10:06 1997 +++ zmailer-2.99.47b2/libc/getaddrinfo.c Fri Mar 21 03:55:18 1997 @@ -11,6 +11,7 @@ /* To do what POSIX says, even when it's broken: */ /* #define BROKEN_LIKE_POSIX 1 */ +#include "hostenv.h" #include #include #include @@ -26,6 +27,12 @@ #endif #include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #include #ifndef EAI_AGAIN #include "netdb6.h" @@ -642,7 +649,7 @@ #include #include #include -#if INET6 +#if AF_INET6 #include #endif /* INET6 */ #include @@ -683,7 +690,7 @@ case AF_INET: addrsize = sizeof(struct in_addr); break; -#if INET6 +#if AF_INET6 case AF_INET6: addrsize = sizeof(struct in6_addr); break; @@ -727,7 +734,7 @@ struct hostent *rval = (struct hostent *)NULL; switch (af) { -#if INET6 +#if AF_INET6 case AF_INET6: rval = _hostname2addr_hosts(name, AF_INET6); if (rval) @@ -832,7 +839,7 @@ #include #include #include -#if INET6 +#if AF_INET6 #include #endif /* INET6 */ #include @@ -849,11 +856,11 @@ /* If this isn't big enough to store any address you're going to parse using this function, you're gonna lose. */ -#if INET6 +#if AF_INET6 #define MAX_ADDR_SZ sizeof(struct in6_addr) -#else /* INET6 */ +#else /* AF_INET6 */ #define MAX_ADDR_SZ sizeof(struct in_addr) -#endif /* INET6 */ +#endif /* AF_INET6 */ /* These are the buffers that become the returned static buffer */ static char h_addr_buf[MAX_ADDR_SZ * MAX_ADDR]; diff -u -r zmailer-2.99.47/libc/getnameinfo.c zmailer-2.99.47b2/libc/getnameinfo.c --- zmailer-2.99.47/libc/getnameinfo.c Tue Mar 18 23:11:30 1997 +++ zmailer-2.99.47b2/libc/getnameinfo.c Fri Mar 21 13:13:19 1997 @@ -7,6 +7,7 @@ v0.02 */ +#include "hostenv.h" #include #include @@ -15,6 +16,12 @@ #endif #include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #include #include @@ -30,6 +37,9 @@ #ifndef AF_LOCAL #define AF_LOCAL AF_UNIX #endif /* AF_LOCAL */ +#ifndef MAXHOSTNAMELEN +# define MAXHOSTNAMELEN 256 +#endif #if HOSTTABLE struct hostent *_addr2hostname_hosts(const char *, int, int); Only in zmailer-2.99.47b2/libc: getnameinfo.c~ diff -u -r zmailer-2.99.47/libc/inet_ntop.c zmailer-2.99.47b2/libc/inet_ntop.c --- zmailer-2.99.47/libc/inet_ntop.c Fri Feb 21 00:06:10 1997 +++ zmailer-2.99.47b2/libc/inet_ntop.c Fri Mar 21 03:57:45 1997 @@ -22,10 +22,12 @@ #include #include -#include -#include +#include /* For AF_**** */ +#if 0 +#include /* Actually not needed.. */ #include #include +#endif #include #include #include diff -u -r zmailer-2.99.47/libc/inet_pton.c zmailer-2.99.47b2/libc/inet_pton.c --- zmailer-2.99.47/libc/inet_pton.c Fri Feb 21 00:06:20 1997 +++ zmailer-2.99.47b2/libc/inet_pton.c Fri Mar 21 03:57:40 1997 @@ -22,10 +22,12 @@ #include #include -#include +#include /* For AF_**** */ +#if 0 #include #include #include +#endif #include #include /* #include "portability.h" */ diff -u -r zmailer-2.99.47/libident/identuser.c zmailer-2.99.47b2/libident/identuser.c --- zmailer-2.99.47/libident/identuser.c Wed Feb 19 22:16:05 1997 +++ zmailer-2.99.47b2/libident/identuser.c Fri Mar 21 04:48:41 1997 @@ -22,6 +22,12 @@ #include /*XXX*/ #include #include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #include #include #ifdef ISC @@ -133,7 +139,7 @@ } -char *ident_sockuser2(s,local,remote,realbuf,realbuflen) +static char *ident_sockuser2(s,local,remote,realbuf,realbuflen) int s; unsigned short local; unsigned short remote; @@ -176,8 +182,10 @@ close(s); if (errno == ECONNREFUSED) return "NO-IDENT-SERVICE"; - else - return "SOCKWRITE"; + else { + sprintf(realbuf,"SOCKWRITE-%d", errno); + return realbuf; + } } else { buf += w; buflen -= w; @@ -195,7 +203,8 @@ CLORETS("SOCKREAD"); *buf = 0; - if (sscanf(realbuf,"%hd,%hd: USERID :%*[^:]:%s",&rremote,&rlocal,realbuf) < 3) { + if (sscanf(realbuf,"%hd,%hd: USERID :%*[^:]:%s", + &rremote,&rlocal,realbuf) < 3) { close(s); errno = EIO; /* makes sense, right? well, not when USERID failed to match ERROR */ @@ -214,11 +223,17 @@ } +static void sig_alrm (sig) + int sig; +{ + longjmp(jmpalarm,1); +} + char *ident_tcpuser9(af,len,inlocal,inremote,local,remote,timeout,buf,buflen) - int af, len; - int *inlocal; + int af, len; /* Address family, and address size */ + int *inlocal; /* Addresses */ int *inremote; - unsigned short local; + unsigned short local; /* Ports */ unsigned short remote; int timeout; char *buf; @@ -229,12 +244,24 @@ fd_set wfds; int saveerrno; void (*old_sig)__((int)); + void (*old_alrm)__((int)); + int oldival; char *retval; - old_sig = signal(SIGPIPE, SIG_IGN); + old_sig = signal(SIGPIPE, SIG_IGN); + oldival = alarm(10); /* 10 seconds until... */ + old_alrm = signal(SIGALRM, sig_alrm); s = -1; - r = ident_tcpsock5(af, len, inlocal, inremote, &s); + if (setjmp(jmpalarm) == 0) { + r = ident_tcpsock5(af, len, inlocal, inremote, &s); + } else { + /* We reach here in case of alarm timer chime... */ + if (s >= 0) close(s); + r = -1; + } + signal(SIGALRM, old_alrm); + alarm(oldival); if (r < 0) { signal(SIGPIPE, old_sig); return "SOCKFAULT1"; diff -u -r zmailer-2.99.47/libresolv/gethnamaddr.c zmailer-2.99.47b2/libresolv/gethnamaddr.c --- zmailer-2.99.47/libresolv/gethnamaddr.c Tue Mar 18 23:15:52 1997 +++ zmailer-2.99.47b2/libresolv/gethnamaddr.c Fri Mar 21 13:20:36 1997 @@ -58,6 +58,8 @@ static char rcsid[] = "$Id: gethnamaddr.c,v 8.16 1996/06/26 23:17:26 vixie Exp $"; #endif /* LIBC_SCCS and not lint */ +#include "hostenv.h" + #include #include #include Only in zmailer-2.99.47b2/libresolv: gethnamaddr.c~ diff -u -r zmailer-2.99.47/router/libdb/bind.c zmailer-2.99.47b2/router/libdb/bind.c --- zmailer-2.99.47/router/libdb/bind.c Wed Mar 12 10:13:46 1997 +++ zmailer-2.99.47b2/router/libdb/bind.c Fri Mar 21 03:46:49 1997 @@ -23,6 +23,12 @@ #endif /* BSDTYPES_H */ #include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #ifdef NOERROR #undef NOERROR /* On Solaris 2.3 the netinet/in.h includes diff -u -r zmailer-2.99.47/router/rfc822.c zmailer-2.99.47b2/router/rfc822.c --- zmailer-2.99.47/router/rfc822.c Fri Feb 7 22:00:23 1997 +++ zmailer-2.99.47b2/router/rfc822.c Fri Mar 21 21:29:55 1997 @@ -2051,7 +2051,16 @@ #ifdef HAVE_FSYNC fsync(FILENO(ofp)); #endif - if ((fclose(ofp) != 0) || (erename(file, qpath) != 0)) { + if (fclose(ofp) != 0) { + unlink(qpath); + unlink(ofpname); +#ifndef USE_ALLOCA + free(ofpname); + free(qpath); +#endif + return PERR_CTRLFILE; + } + if (erename(file, qpath) != 0) { unlink(qpath); unlink(ofpname); #ifndef USE_ALLOCA Only in zmailer-2.99.47b2/router: rfc822.c~ Only in zmailer-2.99.47/smtpserver: core Only in zmailer-2.99.47b2/smtpserver: kissg Only in zmailer-2.99.47/smtpserver: kissg.tgz diff -u -r zmailer-2.99.47/smtpserver/relaytest.c zmailer-2.99.47b2/smtpserver/relaytest.c --- zmailer-2.99.47/smtpserver/relaytest.c Mon Feb 17 00:36:00 1997 +++ zmailer-2.99.47b2/smtpserver/relaytest.c Fri Mar 21 03:59:56 1997 @@ -2,6 +2,12 @@ #ifdef HAVE_SYS_SOCKET_H # include # include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #endif #include diff -u -r zmailer-2.99.47/smtpserver/smtpserver.c zmailer-2.99.47b2/smtpserver/smtpserver.c --- zmailer-2.99.47/smtpserver/smtpserver.c Tue Mar 18 21:02:46 1997 +++ zmailer-2.99.47b2/smtpserver/smtpserver.c Fri Mar 21 03:59:46 1997 @@ -82,6 +82,12 @@ #ifdef HAVE_SYS_SOCKET_H # include # include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #endif #include "syslog.h" /* we use local -- we should anyway */ diff -u -r zmailer-2.99.47/transports/libta/dnsgetrr.c zmailer-2.99.47b2/transports/libta/dnsgetrr.c --- zmailer-2.99.47/transports/libta/dnsgetrr.c Wed Mar 12 10:18:55 1997 +++ zmailer-2.99.47b2/transports/libta/dnsgetrr.c Fri Mar 21 04:06:51 1997 @@ -17,6 +17,7 @@ #include #include #include +#include #include "zmsignal.h" #include "malloc.h" @@ -26,7 +27,9 @@ # include "netdb6.h" #endif #include +#if 0 #include +#endif #endif /* HAVE_RESOLVER */ #include "ta.h" @@ -129,15 +132,30 @@ #define MAXPACKET 1024 #endif +#ifndef INADDRSZ +# define INADDRSZ 4 +#endif +#ifndef IN6ADDRSZ +# define IN6ADDRSZ 16 +#endif +#ifndef AF_INET6 +# define AF_INET6 999 /* If the system does not define this, we use a value + that nobody has as AF_ value -- I hope.. */ +#endif +#ifndef INT16SZ +# define INT16SZ 2 +#endif +#ifndef INT32SZ +# define INT32SZ 4 +#endif + typedef union { HEADER hdr; u_char buf[MAXPACKET]; } querybuf; -typedef union { - int32_t al; - char ac; -} align; +#define ALIGN 4 /* IPv4 works ok with 32-bit alignment */ +#define ALIGN_AAAA 8 /* IPv6 can use 64-bit machines more efficiently.. */ extern int h_errno; extern FILE *verboselog; @@ -311,23 +329,6 @@ static int res_hnok(cp)const char *cp;{return 1;} static int res_dnok(cp)const char *cp;{return 1;} -#ifndef INADDRSZ -# define INADDRSZ 4 -#endif -#ifndef IN6ADDRSZ -# define IN6ADDRSZ 16 -#endif -#ifndef AF_INET6 -# define AF_INET6 999 /* If the system does not define this, we use a value - that nobody has as AF_ value -- I hope.. */ -#endif -#ifndef INT16SZ -# define INT16SZ 2 -#endif -#ifndef INT32SZ -# define INT32SZ 4 -#endif - static const char AskedForGot[] = "gethostby*.getanswer_r: asked for \"%s\", got \"%s\""; @@ -364,7 +365,7 @@ hp->h_addrtype = AF_INET6; hp->h_length = IN6ADDRSZ; for (ap = hp->h_addr_list; *ap; ap++) { - int i = sizeof(align) - ((u_long)*bpp % sizeof(align)); + int i = ALIGN_AAAA - ((u_long)*bpp % ALIGN_AAAA); if (*lenp < (i + IN6ADDRSZ)) { /* Out of memory. Truncate address list here. XXX */ @@ -598,7 +599,10 @@ buflen -= nn; } - bp += sizeof(align) - ((u_long)bp % sizeof(align)); + if (type == T_A) + bp += ALIGN - ((u_long)bp % ALIGN ); + else + bp += ALIGN_AAAA - ((u_long)bp % ALIGN_AAAA); if (bp + n >= &result->hostbuf[sizeof(result->hostbuf)]) { DPRINTF(("size (%d) too big\n", n)); @@ -765,7 +769,7 @@ char hname2[MAXDNAME+1]; #endif /*SUNSECURITY*/ extern struct hostent *_gethtbyaddr(); - + if ((_res.options & RES_INIT) == 0 && res_init() == -1) { h_errno = NETDB_INTERNAL; return (NULL); @@ -820,7 +824,7 @@ } n = res_query(qbuf, C_IN, T_PTR, (u_char *)buf.buf, sizeof buf.buf); if (n < 0) { - DPRINTF(("res_query failed (%d)\n", n)); + DPRINTF(("res_query failed (%d)\n", n)); #if 0 /* No fallback to file! */ if (errno == ECONNREFUSED) return (_gethtbyaddr(addr, len, af)); diff -u -r zmailer-2.99.47/transports/smtp/smtp.c zmailer-2.99.47b2/transports/smtp/smtp.c --- zmailer-2.99.47/transports/smtp/smtp.c Tue Mar 18 23:40:41 1997 +++ zmailer-2.99.47b2/transports/smtp/smtp.c Fri Mar 21 04:07:39 1997 @@ -47,12 +47,19 @@ #include #include +#include +#include +#ifdef HAVE_LINUX_IN6_H +# include +#endif +#ifdef HAVE_NETINET6_IN6_H +# include +#endif #include #ifndef EAI_AGAIN # include "netdb6.h" #endif -#include -#include + #include "mail.h" #include "syslog.h" #include "ta.h" Only in zmailer-2.99.47: zmailer-2.99.47.tar.gz From zmailer-owner@nic.funet.fi Fri Mar 21 22:28:27 1997 Received: by nic.funet.fi id <4157-8725>; Fri, 21 Mar 1997 22:27:03 +0200 Subject: Re: 2.99.47? From: Matti Aarnio To: agulbra@troll.no (Arnt Gulbrandsen) Date: Fri, 21 Mar 1997 22:26:48 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Arnt Gulbrandsen" at Mar 19, 97 07:12:59 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1018 Message-Id: <19970321202703Z4157-8725+236@nic.funet.fi> Return-Path: > I'm setting up another zmailer box, and wouldn't like to do it _just_ > before 2.99.47 is released. Matti? Are you some sort of foreteller ? (Soothsayer?) Of course you asked this just after I had made ..47 available, but it took me two days to figure out all the necessary bits for it to work :-/ > Btw, an anecdote for some of you. I just sent NOTIFY=SUCCESS on a > mailing-list message (to probe for address changes) and got this thing > back from Microsoft. I looked at the uuencoded blob, and, yes: all > the relevant information is there, in some proprietary, unreadable > binary format. > > DSN is a standard, so of course Microsoft supports it. Uhh.. And it was about the same information (that was in clear ASCII) but in binary form. I have also let myself to hear that that binary format is extremely version dependent, and it won't interoperate unless both parties have exactly same revision and build- level of the mail system clients... > --Arnt /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Mar 22 04:39:58 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <4125-2713>; Sat, 22 Mar 1997 04:39:04 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17411-27736>; Fri, 21 Mar 1997 19:38:03 -0700 Date: Fri, 21 Mar 1997 19:38:14 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: Eugene Crosser cc: zmailer@nic.funet.fi Subject: Any luck with 2.99.47+p1+new_configure_file? In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I am still having the make fail under SunOS 4.1.3_U1 with gcc while compiling dnsgetrr.c: gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o dnsgetrr.o In file included from dnsgetrr.c:45: /usr/include/resolv.h:29: field `nsaddr_list' has incomplete type /usr/include/resolv.h:35: field `sort_list' has incomplete type dnsgetrr.c: In function `getrrtype': dnsgetrr.c:296: `IPPROTO_TCP' undeclared (first use this function) dnsgetrr.c:296: (Each undeclared identifier is reported only once dnsgetrr.c:296: for each function it appears in.) dnsgetrr.c: In function `getanswer_r': dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) dnsgetrr.c:575: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyname2_r': dnsgetrr.c:674: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:701: warning: assignment makes pointer from integer without a cast dnsgetrr.c:734: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyaddr_r': dnsgetrr.c:774: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:883: `NETDB_SUCCESS' undeclared (first use this function) make[3]: *** [dnsgetrr.o] Error 1 make[3]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/transports/ libta' make[2]: *** [libta.a] Error 1 make[2]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/libs' make[1]: *** [../libs/libtag] Error 1 make[1]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/smtpserver' make: *** [all] Error 1 Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Sat Mar 22 04:44:53 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <4125-30164>; Sat, 22 Mar 1997 04:39:04 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17411-27736>; Fri, 21 Mar 1997 19:38:03 -0700 Date: Fri, 21 Mar 1997 19:38:14 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: Eugene Crosser cc: zmailer@nic.funet.fi Subject: Any luck with 2.99.47+p1+new_configure_file? In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I am still having the make fail under SunOS 4.1.3_U1 with gcc while compiling dnsgetrr.c: gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o dnsgetrr.o In file included from dnsgetrr.c:45: /usr/include/resolv.h:29: field `nsaddr_list' has incomplete type /usr/include/resolv.h:35: field `sort_list' has incomplete type dnsgetrr.c: In function `getrrtype': dnsgetrr.c:296: `IPPROTO_TCP' undeclared (first use this function) dnsgetrr.c:296: (Each undeclared identifier is reported only once dnsgetrr.c:296: for each function it appears in.) dnsgetrr.c: In function `getanswer_r': dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) dnsgetrr.c:575: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyname2_r': dnsgetrr.c:674: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:701: warning: assignment makes pointer from integer without a cast dnsgetrr.c:734: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyaddr_r': dnsgetrr.c:774: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:883: `NETDB_SUCCESS' undeclared (first use this function) make[3]: *** [dnsgetrr.o] Error 1 make[3]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/transports/ libta' make[2]: *** [libta.a] Error 1 make[2]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/libs' make[1]: *** [../libs/libtag] Error 1 make[1]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/smtpserver' make: *** [all] Error 1 Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Sat Mar 22 06:28:29 1997 Received: from uxmail.ust.hk ([143.89.14.30]) by nic.funet.fi with ESMTP id <4739-10260>; Sat, 22 Mar 1997 06:27:16 +0200 Received: from ccsu51.ust.hk ([143.89.103.20]) by uxmail.ust.hk with SMTP id <102279-20006>; Sat, 22 Mar 1997 12:25:43 +0800 Received: by ccsu51.ust.hk (SMI-8.6/SMI-SVR4) id MAA08459; Sat, 22 Mar 1997 12:25:37 +0800 From: ccyflai@uxmail.ust.hk (Lai Yiu Fai) Message-Id: <199703220425.MAA08459@ccsu51.ust.hk> Subject: Re: Any luck with 2.99.47+p1+new_configure_file? To: zmailer@nic.funet.fi Date: Sat, 22 Mar 1997 12:25:36 +0800 (HKT) In-Reply-To: from "Daryle Tilroe" at Mar 21, 97 07:38:14 pm X-Mailer: ELM [version 2.4 PL24alpha3] Content-Type: text Return-Path: Also the same problem compiling with Solaris 2.5 with gcc. > > > I am still having the make fail under SunOS 4.1.3_U1 with gcc while > compiling dnsgetrr.c: > > gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o > dnsgetrr.o > In file included from dnsgetrr.c:45: > /usr/include/resolv.h:29: field `nsaddr_list' has incomplete type > /usr/include/resolv.h:35: field `sort_list' has incomplete type > dnsgetrr.c: In function `getrrtype': > dnsgetrr.c:296: `IPPROTO_TCP' undeclared (first use this function) > dnsgetrr.c:296: (Each undeclared identifier is reported only once > dnsgetrr.c:296: for each function it appears in.) > dnsgetrr.c: In function `getanswer_r': > dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) > dnsgetrr.c:575: `NETDB_SUCCESS' undeclared (first use this function) > dnsgetrr.c: In function `gethostbyname2_r': > dnsgetrr.c:674: `NETDB_INTERNAL' undeclared (first use this function) > dnsgetrr.c:701: warning: assignment makes pointer from integer without a > cast > dnsgetrr.c:734: `NETDB_SUCCESS' undeclared (first use this function) > dnsgetrr.c: In function `gethostbyaddr_r': > dnsgetrr.c:774: `NETDB_INTERNAL' undeclared (first use this function) > dnsgetrr.c:883: `NETDB_SUCCESS' undeclared (first use this function) > make[3]: *** [dnsgetrr.o] Error 1 > make[3]: Leaving directory > `/usr/mail/zmailer/source/zmailer-2.99.47/transports/ > libta' > make[2]: *** [libta.a] Error 1 > make[2]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/libs' > make[1]: *** [../libs/libtag] Error 1 > make[1]: Leaving directory > `/usr/mail/zmailer/source/zmailer-2.99.47/smtpserver' > make: *** [all] Error 1 > > > > Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca > Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca > #318, 11315-87 Avenue |----------------------------------------------------| > Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | > Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | > T6G 2T9 /__\ |----------------------------------------------------| > > From zmailer-owner@nic.funet.fi Sat Mar 22 17:56:11 1997 Received: from gost6.indirect.com ([165.247.198.6]) by nic.funet.fi with ESMTP id <4886-2713>; Sat, 22 Mar 1997 17:55:41 +0200 Received: from direct.net (ip97.santa-ana3.ca.pub-ip.psi.net [38.29.18.97]) by gost6.indirect.com (8.8.5/8.6.9) with SMTP id IAA25997; Sat, 22 Mar 1997 08:54:39 -0700 (MST) Date: Sat, 22 Mar 1997 08:54:39 -0700 (MST) From: g4y4t7y5@direct.net Message-Id: <199703221554.IAA25997@gost6.indirect.com> Subject: bank 14 times @~!2=lkl peaceful check i mailed good try to reach 8522 To: unlisted-recipients:; (no To-header on input) Return-Path: Your letters & faxes can have your signature without you having to sign them! And because we create your signature as a True Type font, it will look excellent! If you are currently using a scanned image of your signature to solve this problem, let us turn your signature into a True Type font and say good-bye to cutting and pasting or importing from outside files forever. You will be able to access your signature right from the font menu. It's easy and it's fast. In the past you had to import a previously drawn or scanned graphic that used not only a tremendous amount of space and time to display - but it also took forever to print. Well, not anymore. With the MASTER GRAPHICS "Signature" font, you just click on your font menu for your signature, and presto! Your selection appears in a snap as a True Type font in your document, fully scaleable to any size. Here is what David from San Jose had to say about the MASTER GRAPHICS "Signature" font... "Thank you for the quality job that you have done. Also, I found the installation and use of the font to be explained in such a way that even a novice, such as myself, was able to accomplish it successfully." ...Paulina faxed us..."Congratulations, you did a great job! Thanks a lot, it's wonderful to be able "to sign". John from West Haven, CT says... "I just wanted to let you know that I received the signature font...and it works just fine. Thank you very much for your very kind attention to the needs of this customer. Gregg the "Shirtsleeves Accountant" says:..."I installed the new font... and it works like a charm. Thank you. S I G N A T U R E F O N T S are priced at only $28.95 per full signature. Any questions? Call us at (619) 723-8724 Monday through Friday, 10:00 am to 5:00 pm Pacific Standard Time. ====>W E C A N A L S O D O Y O U R L O G O ! !<==== Most Logo's cost between $60.00 and $100.00 Please send us a sample and we will call, fax or E-mail you back with a quick quote. ----------------------------------------------------------- YES: Please send me my signature as a True Type font. I understand that if my signature doesn't work to my satisfaction, I may return everything for a full refund. Name...:_____________________________________________________ Company:_____________________________________________________ Address:_____________________________________________________ City...:_____________________________________________________ State..:________________________ Zip:_______________________ Phone..:_____________________________________________________ Fax....:_____________________________________________________ Email..:_____________________________________________________ SIGNATURE INSTRUCTIONS: 1. Use a BLACK FINE FELT TIP pen that has plenty of ink or a black "UNI-BALL" pen. These pens produce the best results. At least avoid using a blue ink pen or a Bic Type pen, these pens produce poor results. 2. Practice writing your signature on a plain white sheet of paper. 3. When finished practicing, please indicate which signature you like the most by pointing to it with an arrow. Be carefull not to touch your signature with this arrow or any other marks. 4. Read the following two descriptions and pick one. Then fill out the rest of this form. | E | R | U | T | A | N | G | I | S |________________________________________ Some people like to sign on a slant. If you want us to maintain the slant that is on your sheet of paper, then CHECK HERE [ ] | | | | | S I G N A T U R E | | | | |___________________________________________ Some people like to keep their signature very straight. If you want us to make your signature as straight as possible, then CHECK HERE [ ] I WILL BE USING MY FONT ON A: [ ]PC [ ]MAC [ ] Please send me my full signature as a True Type font $28.95 [ ] For an additional $9.95 we will create a 2nd image of your first name only for times that you might want a more personal signature. [ ] Get both PC & MAC for an additional $14.95 Please allow three weeks to create and ship your font. [ ] RUSH! We can create your font within 2 days and ship it to you via 3 day letter for an additional $25.00 Total amount of sale: (If you are in CA please include 7.25% sales tax) $_______________________ Method of payment: [ ]My check is enclosed. (USA customers only) Please charge my: [ ]VISA [ ]MasterCard [ ]American Express EXP. DATE:_____________ CARD # _________________________________________ SIGNATURE:_______________________________________________________________ (This signature is for credit card approval only, we don't use this one to make your font.) SEND TO: MASTER GRAPHICS 1119 S. MISSION RD. FALLBROOK, CA 92028 USA (Did you remember to include your signature?) Any questions? Call us at (619) 723-8724 Monday through Friday, 10:00 am to 5:00 pm Pacific Standard Time. From zmailer-owner@nic.funet.fi Sat Mar 22 22:25:12 1997 Received: by nic.funet.fi id <4368-5977>; Sat, 22 Mar 1997 22:23:47 +0200 Subject: Re: Any luck with 2.99.47+p1+new_configure_file? From: Matti Aarnio To: data@amc.ab.ca Date: Sat, 22 Mar 1997 22:23:45 +0200 (GMT) Cc: crosser@online.ru, zmailer@nic.funet.fi In-Reply-To: from "Daryle Tilroe" at Mar 21, 97 07:38:14 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2725 Message-Id: <19970322202347Z4368-5977+149@nic.funet.fi> Return-Path: > I am still having the make fail under SunOS 4.1.3_U1 with gcc while > compiling dnsgetrr.c: Propably that is because of a "#if 0":ed include of in the transports/libta/dnsgetrr.c I did forget in. I did fix it at nic.funet.fi test compilation, but didn't put it back to the base source.. Right now I am sitting on a back seat of a car driving across southern Finland -- I participated on a radio amateur meeting 200 km from my home... Nice to have a well working cellular data network as a functionality of the cell-phone infrastructure :) ( And its use does not cost all that much :-) Not that I will ever see the bill -- benefits of working at a telco company.. ) /Matti Aarnio > gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o > dnsgetrr.o > In file included from dnsgetrr.c:45: > /usr/include/resolv.h:29: field `nsaddr_list' has incomplete type > /usr/include/resolv.h:35: field `sort_list' has incomplete type > dnsgetrr.c: In function `getrrtype': > dnsgetrr.c:296: `IPPROTO_TCP' undeclared (first use this function) > dnsgetrr.c:296: (Each undeclared identifier is reported only once > dnsgetrr.c:296: for each function it appears in.) > dnsgetrr.c: In function `getanswer_r': > dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) > dnsgetrr.c:575: `NETDB_SUCCESS' undeclared (first use this function) > dnsgetrr.c: In function `gethostbyname2_r': > dnsgetrr.c:674: `NETDB_INTERNAL' undeclared (first use this function) > dnsgetrr.c:701: warning: assignment makes pointer from integer without a > cast > dnsgetrr.c:734: `NETDB_SUCCESS' undeclared (first use this function) > dnsgetrr.c: In function `gethostbyaddr_r': > dnsgetrr.c:774: `NETDB_INTERNAL' undeclared (first use this function) > dnsgetrr.c:883: `NETDB_SUCCESS' undeclared (first use this function) > make[3]: *** [dnsgetrr.o] Error 1 > make[3]: Leaving directory > `/usr/mail/zmailer/source/zmailer-2.99.47/transports/ > libta' > make[2]: *** [libta.a] Error 1 > make[2]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/libs' > make[1]: *** [../libs/libtag] Error 1 > make[1]: Leaving directory > `/usr/mail/zmailer/source/zmailer-2.99.47/smtpserver' > make: *** [all] Error 1 > > > > Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca > Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca > #318, 11315-87 Avenue |----------------------------------------------------| > Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | > Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | > T6G 2T9 /__\ |----------------------------------------------------| > > From zmailer-owner@nic.funet.fi Mon Mar 24 16:11:01 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <4433-15488>; Mon, 24 Mar 1997 16:09:43 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-24584>; Mon, 24 Mar 1997 17:09:17 +0300 Date: Mon, 24 Mar 1997 17:09:12 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Re: Patch for zmailer 2.99.47 -- solaris works :) To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970321201303Z3915-11798+164@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: It still would not compile: In file included from bind.c:46: ../../include/dnsgetrr.h:19: conflicting types for `gethostbyaddr_r' /usr/include/netdb.h:142: previous declaration of `gethostbyaddr_r' Matti, why do you need gethostbyaddr_r() at all? I do not see any reference in the source? SPARC/Solaris 2.5.1, gcc. Eugene From zmailer-owner@nic.funet.fi Mon Mar 24 16:39:52 1997 Received: by nic.funet.fi id <4871-15488>; Mon, 24 Mar 1997 16:38:55 +0200 Subject: Re: Patch for zmailer 2.99.47 -- solaris works :) From: Matti Aarnio To: crosser@online.ru Date: Mon, 24 Mar 1997 16:38:54 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Mar 24, 97 05:09:12 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 550 Message-Id: <19970324143855Z4871-15488+235@nic.funet.fi> Return-Path: > It still would not compile: > > In file included from bind.c:46: > ../../include/dnsgetrr.h:19: conflicting types for `gethostbyaddr_r' > /usr/include/netdb.h:142: previous declaration of `gethostbyaddr_r' > > Matti, why do you need gethostbyaddr_r() at all? I do not see any > reference in the source? That's right. They are FUTURE HOOKS for something I plan to do sometime, when I have a spare moment. I have now decided to rename them to be *_rz to avoid this naming conflict. > SPARC/Solaris 2.5.1, gcc. > > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Mar 24 17:00:18 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <5005-16844>; Mon, 24 Mar 1997 16:59:35 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-26226>; Mon, 24 Mar 1997 17:59:16 +0300 Date: Mon, 24 Mar 1997 17:59:10 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Re: Patch for zmailer 2.99.47 -- solaris works :) To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970324143855Z4871-15488+235@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > > In file included from bind.c:46: > > ../../include/dnsgetrr.h:19: conflicting types for `gethostbyaddr_r' > > /usr/include/netdb.h:142: previous declaration of `gethostbyaddr_r' > > > > Matti, why do you need gethostbyaddr_r() at all? I do not see any > > reference in the source? > > That's right. They are FUTURE HOOKS for something I plan > to do sometime, when I have a spare moment. I see. BTW, after renaming the function and un-commenting netinet/in.h include, it seems to work. Eugene From zmailer-owner@nic.funet.fi Mon Mar 24 18:28:00 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <5041-15488>; Mon, 24 Mar 1997 18:26:49 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68041-26226>; Mon, 24 Mar 1997 19:23:58 +0300 Date: Mon, 24 Mar 1997 19:23:50 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Delivery reports vs non-delivery reports To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: I'd like to point a "feature" of recent Zmailers that is not a big deal but still a bit annoying: delivery reports are going the same way as non-delivery reports (which is good), and as a concequence, they are sent to the local postmaster as well as to the remote user. Which is bad, beacuse it overloads my mailbox with unimportant junk. Any idea how to prevent this (aside from installing procmail)? Eugene From zmailer-owner@nic.funet.fi Mon Mar 24 20:09:56 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <3180-25379>; Mon, 24 Mar 1997 20:08:27 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17415-27808>; Mon, 24 Mar 1997 11:07:16 -0700 Date: Mon, 24 Mar 1997 11:07:20 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: Matti Aarnio cc: crosser@online.ru, zmailer@nic.funet.fi Subject: Re: Any luck with 2.99.47+p1+new_configure_file? In-Reply-To: <19970322202347Z4368-5977+149@nic.funet.fi> Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sat, 22 Mar 1997, Matti Aarnio wrote: > Propably that is because of a "#if 0":ed include of > in the transports/libta/dnsgetrr.c Commenting it back in helps a bit but I still get this: ------------- gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o dnsgetrr.o dnsgetrr.c: In function `getanswer_r': dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) dnsgetrr.c:429: (Each undeclared identifier is reported only once dnsgetrr.c:429: for each function it appears in.) dnsgetrr.c:575: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyname2_r': dnsgetrr.c:674: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:701: warning: assignment makes pointer from integer without a cast dnsgetrr.c:734: `NETDB_SUCCESS' undeclared (first use this function) dnsgetrr.c: In function `gethostbyaddr_r': dnsgetrr.c:774: `NETDB_INTERNAL' undeclared (first use this function) dnsgetrr.c:883: `NETDB_SUCCESS' undeclared (first use this function) make: *** [dnsgetrr.o] Error 1 -------------- I have tried playing around with the include of but to no avail so far. 2.99.47+p1+new_configure_file under SunOS 4.1.3_U1 with gcc Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Mon Mar 24 20:38:05 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <4093-16844>; Mon, 24 Mar 1997 20:36:41 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68040-26228>; Mon, 24 Mar 1997 21:36:17 +0300 Date: Mon, 24 Mar 1997 21:36:12 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel Reply-To: Eugene Crosser Subject: Re: Any luck with 2.99.47+p1+new_configure_file? To: Daryle Tilroe cc: Matti Aarnio , zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > > Propably that is because of a "#if 0":ed include of > > in the transports/libta/dnsgetrr.c > > Commenting it back in helps a bit but I still get this: > > ------------- > > gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o > dnsgetrr.o > dnsgetrr.c: In function `getanswer_r': > dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) If you are using gcc, look if there is a `replacement' nameser.h in /usr/local/lib/gcc-lib/whatever/whatever/include/arpa. It may conflict with `real' .h files and cause problems. I had to rename it away. Eugene From zmailer-owner@nic.funet.fi Mon Mar 24 21:45:24 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <4274-15488>; Mon, 24 Mar 1997 21:43:47 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17409-27807>; Mon, 24 Mar 1997 12:42:30 -0700 Date: Mon, 24 Mar 1997 12:42:37 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: Eugene Crosser cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: Any luck with 2.99.47+p1+new_configure_file? In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 24 Mar 1997, Eugene Crosser wrote: > > > Propably that is because of a "#if 0":ed include of > > > in the transports/libta/dnsgetrr.c > > > > Commenting it back in helps a bit but I still get this: > > > > ------------- > > > > gcc -g -O2 -DHAVE_CONFIG_H -I../../include -I../.. -c dnsgetrr.c -o > > dnsgetrr.o > > dnsgetrr.c: In function `getanswer_r': > > dnsgetrr.c:429: `HFIXEDSZ' undeclared (first use this function) > > If you are using gcc, look if there is a `replacement' nameser.h in > /usr/local/lib/gcc-lib/whatever/whatever/include/arpa. > It may conflict with `real' .h files and cause problems. > I had to rename it away. Thanks for the help unfortunatly it didn't work. It did however lead me to the likely cause. I believe the compile is now broken for BSD based machines. It looks like in configure BSD should not use the provided bind (libresolv directory: nameser.h, netdb.h, etc). It is however in those that all the above undefined *_r stuff (HFIXEDSZ,NETDB_INTERNAL,etc) is defined. I am going to try defining them by hand. Perhaps there should be some ifndef's in the code. Is this the correct diagnosis? If so what are we BSD folks supposed to do? Besides upgrade to Solaris 2.5. Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Wed Mar 26 19:21:12 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2519-3044>; Wed, 26 Mar 1997 19:20:06 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68117-32>; Wed, 26 Mar 1997 20:19:40 +0300 Date: Wed, 26 Mar 1997 20:19:27 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: address recognition change? bug? To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: Matti, Immediately after I installed .47 on our main mail relay, I noticed strange behavior. I am running a watchdog here from cron. It runs on a machine named chronos.sovam.com, and sends a message to the address "@online.ru:mailping@chronos.sovam.com". So, the message is routed to the main "online.ru" relay (deimos), which cuts off "@online.ru:" and sends the message back to chronos where it is processed by a script. This worked before (incl .46) but when I installed .47, it writes: Mar 26 19:39:09 deimos smtp[17216]: .nCIpl228018: to=@online.ru:mailping@chronos.sovam.com, delay=00:15:56, xdelay=00:00:01, mailer=smtp, stat=retryat +60 smtp; 500 (nameserver data inconsistency. All MXes rejected (we are the best?), no address: 'online.ru', errno=Error 0) and would not route the message back. If I change the address to mailping%chronos.sovam.com@online.ru things work as before. Although, I think that inability to process correct rfc822 host-route is a bug? Eugene From zmailer-owner@nic.funet.fi Wed Mar 26 19:59:13 1997 Received: from outbound.Princeton.EDU ([128.112.128.84]) by nic.funet.fi with ESMTP id <3204-3212>; Wed, 26 Mar 1997 19:57:59 +0200 Received: from lists.Princeton.EDU ([128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <541337-24827>; Wed, 26 Mar 1997 12:57:11 -0500 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370055.s1-2>; Wed, 26 Mar 1997 12:57:06 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id MAA29741 for ; Wed, 26 Mar 1997 12:56:55 -0500 Message-Id: <199703261756.MAA29741@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: 822 header rewrite Organization: Princeton University, Network Systems, Class. 56 X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 26 Mar 1997 12:56:54 EST From: "Michael R. Gettes" Return-Path: I would like to rewrite rfc822 headers with zmailer 2.99.45. It is not clear to me how to do this. Do I modify the HEADERS db? Does anyone have sample code? Thanks. /mrg From zmailer-owner@nic.funet.fi Wed Mar 26 20:06:55 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <5023-30206>; Wed, 26 Mar 1997 20:05:57 +0200 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17416-19838>; Wed, 26 Mar 1997 11:04:42 -0700 Date: Wed, 26 Mar 1997 11:04:49 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: zmailer@nic.funet.fi cc: Matti Aarnio Subject: Still in need: Best stable version or 2.99.47 fix... In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I have fooled around with 2.99.47+p1+new_configure_file for a while now but without much sucesss. Manually defining in dnsgetrr.c: #define HFIXEDSZ 12 #define NETDB_SUCCESS 0 #define NETDB_INTERNAL -1 helps but now I am up against: make[1]: Leaving directory `/usr/mail/zmailer/source/zmailer-2.99.47/transports/hold' gcc -g -O2 -DHAVE_CONFIG_H -I../.. -I../../include -o hold hold.o version.o -L../../libs -lta -lz -lzc -lmalloc -lresolv ld: Undefined symbol ___hostalias collect2: ld returned 2 exit status and I am running out of time as I am running a slightly buggy 2.99.46+p1+p2 (although it compiled quite nicely) without proper error messaging and with the propensity to "hang-up" just a wee bit to early for some incoming smtp connections. This seems to comfuse them and they try again even though the message has been delivered and leads to multiple copies of the message or in the worst case (has happened three times now) deliver a copy every few minutes from the sending machine until the three day time-out is reached. These might be fixed in .47 but as I said I am still unable to get it to compile and even if I were to I could not be certain that the hacks would not cause any unexpected problems. Thus I really need to know either what the previouly most stable release was (especially for SunOS 4.1.3_U1, I had previously been using .11) or when a fix might be forthcoming. Thanks, Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Wed Mar 26 20:33:24 1997 Received: by nic.funet.fi id <2190-3212>; Wed, 26 Mar 1997 20:32:40 +0200 Subject: Re: address recognition change? bug? From: Matti Aarnio To: crosser@online.ru Date: Wed, 26 Mar 1997 20:32:31 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Mar 26, 97 08:19:27 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1384 Message-Id: <19970326183240Z2190-3212+632@nic.funet.fi> Return-Path: > Matti, > > Immediately after I installed .47 on our main mail relay, I noticed > strange behavior. I am running a watchdog here from cron. It > runs on a machine named chronos.sovam.com, and sends a message to > the address "@online.ru:mailping@chronos.sovam.com". So, the message > is routed to the main "online.ru" relay (deimos), which cuts off > "@online.ru:" and sends the message back to chronos where it is processed > by a script. This worked before (incl .46) but when I installed .47, it > writes: Whoops.. That is due to some change in between .46 and .47 in the proto/cf/*.cf -files, which you undoubtly have installed ? The rfc-source-route should be recognized, and be stripped, but some other of my changes might have affected it. > Mar 26 19:39:09 deimos smtp[17216]: .nCIpl228018: > to=@online.ru:mailping@chronos.sovam.com, delay=00:15:56, xdelay=00:00:01, > mailer=smtp, stat=retryat +60 smtp; 500 (nameserver data inconsistency. All > MXes rejected (we are the best?), no address: 'online.ru', errno=Error 0) > > and would not route the message back. If I change the address to > > mailping%chronos.sovam.com@online.ru > > things work as before. Although, I think that inability to process > correct rfc822 host-route is a bug? It is in process of being depreciated, but it should be processable, that is true. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 26 20:38:21 1997 Received: by nic.funet.fi id <2185-3212>; Wed, 26 Mar 1997 20:37:03 +0200 Subject: Re: 822 header rewrite From: Matti Aarnio To: gettes@princeton.edu (Michael R. Gettes) Date: Wed, 26 Mar 1997 20:37:00 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <199703261756.MAA29741@yo.Princeton.EDU> from "Michael R. Gettes" at Mar 26, 97 12:56:54 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 572 Message-Id: <19970326183703Z2185-3212+633@nic.funet.fi> Return-Path: > I would like to rewrite rfc822 headers with zmailer 2.99.45. > It is not clear to me how to do this. Do I modify the HEADERS > db? Does anyone have sample code? The addresses in the headers are rewriteable -- presuming the router knows about those headers.. (Like it knows about addresses present in "From:", and "To:" headers, etc.) That is controlled by the function called crossbar in MAILSHARE/cf/crossbar.cf The outlook of the headers and/or addition/deletion of some is not controllable -- no rewrite in that sense. > Thanks. > > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Mar 26 21:51:52 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2928-3212>; Wed, 26 Mar 1997 21:51:03 +0200 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68117-33>; Wed, 26 Mar 1997 22:50:32 +0300 Date: Wed, 26 Mar 1997 22:50:21 +0300 (MSK) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: address recognition change? bug? To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970326183240Z2190-3212+632@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > > Matti, > > > > Immediately after I installed .47 on our main mail relay, I noticed > > strange behavior. I am running a watchdog here from cron. It > > runs on a machine named chronos.sovam.com, and sends a message to > > the address "@online.ru:mailping@chronos.sovam.com". So, the message > > is routed to the main "online.ru" relay (deimos), which cuts off > > "@online.ru:" and sends the message back to chronos where it is processed > > by a script. This worked before (incl .46) but when I installed .47, it > > writes: > > Whoops.. That is due to some change in between .46 and .47 > in the proto/cf/*.cf -files, which you undoubtly have installed ? Yes. And I am *almost* sure that I installed .46 together with its .cf files (although sometimes I only do "make install-bin"). In case of .47, I installed Zmailer on a completely fresh machine. Eugene From zmailer-owner@nic.funet.fi Wed Mar 26 22:55:56 1997 Received: from press-gopher.uchicago.edu ([128.135.234.36]) by nic.funet.fi with ESMTP id <2809-3044>; Wed, 26 Mar 1997 22:54:58 +0200 Received: from localhost by press-gopher.uchicago.edu with SMTP id <26488-16294>; Wed, 26 Mar 1997 14:54:46 -0600 Date: Wed, 26 Mar 1997 14:54:39 -0600 (CST) Sender: Roy C Bixler From: Roy C Bixler To: zmailer@nic.funet.fi Subject: Z-mailer 2.99.47+ on Linux Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: It seems to run fine for the most part, but there is a problem I notice sometimes. On two separate occasions, outgoing messages got stuck on the queue after 3 or 6 attempts. Here is example 'mailq' optput from the latest one: p0 14:21 press[442] mailq -s -Q ~ 0 entries in router queue: idle 1 message in transport queue: working smtp/*.edu/0 smtp/artic.edu/0 R=1 A=6 P=9301 HA=26973s FA=26971s OF=1 QA=22h33m7s Threads: 1 Msgs: 1 Procs: 1 Idle: 0 Plim: 20 Flim: 150 Kids: 1 Idle: 0 Msgs: 1 Thrds: 1 Rcpnts: 1 Uptime: 7h36m36s If I re-start Z-mailer, the message is then delivered properly, but I see that the old process with PID 9301 is still running. This is on a Linux 2.0.29 system with GCC 2.7.2.1 compiler and libc 5.4.20. I did the following trickery to get the thing to compile: since 'AF_INET6' is #define'd in the 'linux/socket.h' header file but having this definition causes the compilation to fail, I copied this file into the Z-mailer includes under the 'linux' directory I created there and edited out the 'AF_INET6' definition. Also, if it matters, I did a fresh install of Z-mailer. Also, has anyone had any luck with the 'identd' stuff? At best (i.e. if I 'telnet' direct to the SMTP port from a machine running 'identd'), I get something like: Mar 26 14:34:20 press smtpserver[15470]: connection from IDENT-NONSENSE@host TIA, Roy rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Fri Mar 28 22:44:19 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <4645-969>; Fri, 28 Mar 1997 22:43:21 +0200 Received: by redhotmomma.ssr.com via suspension id <36109-17668>; Fri, 28 Mar 1997 15:57:19 -0500 Received: by redhotmomma.ssr.com id <36108-17649>; Fri, 28 Mar 1997 15:56:33 -0500 Sender: Scott Ballantyne Fake-Sender: Scott Ballantyne From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: Bug in smtpserver Message-Id: <97Mar28.155633-0500_est.36108-17649+2@redhotmomma.ssr.com> Date: Fri, 28 Mar 1997 15:56:29 -0500 Return-Path: smtpserver 2.99.38 core dumps on lines like smtpserver -i ; Sat, 29 Mar 1997 00:54:35 +0200 Subject: Re: Bug in smtpserver From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Sat, 29 Mar 1997 00:54:26 +0200 (GMT) Cc: zmailer@nic.funet.fi In-Reply-To: <97Mar28.155633-0500_est.36108-17649+2@redhotmomma.ssr.com> from "Scott Ballantyne" at Mar 28, 97 03:56:29 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1120 Message-Id: <19970328225435Z1617-388+349@nic.funet.fi> Return-Path: > smtpserver 2.99.38 core dumps on lines like > smtpserver -i > The reason is that logfp is null, and the fprintf statements don't > check for that. A workaround is of course, > smtpserver -i > Better would be to clean up the logging code. > > sdb I recall detecting it a while back too -- and fixing it at 2.99.46. The problem was in openlogfp() routine. Whoops, there are still unprotected logfp references. Although these should not matter in case of "-i" option. I just got finished with one high-priority project at the office, and now I could relax a bit -- and pick some code for hacking... just that I have to try to squeeze it thru this d*n 9600 bps cellular link... (2.1 MB of ZMailer source ... shudder...) In Finland we have a proveb "The showmakers children don't have shoes" - nor telecom workers have wire-line phones, only these cell-things... Though we are looking for ways to get leased line service for our production, and development people sometime this year :) (The development team where I am in was formed last summer.) /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Mar 29 01:33:40 1997 Received: from rainier.nds.com ([199.74.149.12]) by nic.funet.fi with ESMTP id <6321-10636>; Sat, 29 Mar 1997 01:32:45 +0200 Received: by rainier.nds.com id <171332-237>; Fri, 28 Mar 1997 15:32:37 -0800 In-Reply-To: Daryle Tilroe "Still in need: Best stable version or 2.99.47 fix..." (Mar 26, 11:04am) X-Mailer: Mail User's Shell (7.2.4 2/2/92) From: Brian Dockter To: zmailer@nic.funet.fi Subject: Re: Still in need: Best stable version or 2.99.47 fix... Message-Id: <97Mar28.153237pst.171332-237+76@rainier.nds.com> Date: Fri, 28 Mar 1997 15:32:35 -0800 Return-Path: On Mar 26, 11:04am, Daryle Tilroe wrote: > Subject: Still in need: Best stable version or 2.99.47 fix... > > These might be fixed in .47 but as I said I am still unable to get it to > compile and even if I were to I could not be certain that the hacks would > not cause any unexpected problems. Thus I really need to know either what > the previouly most stable release was (especially for SunOS 4.1.3_U1, I > had previously been using .11) or when a fix might be forthcoming. I too would like to get this information. I am getting ready to set up a Solaris 2.x system for a company which doesn't have any full time administrators. I would like to install Zmailer, but I don't want to give them an administration burden. Thanks, Brian -- Brian Dockter (KC7JZL) | Email: brian@nds.com Sr. Software Engineer | Voice: 206-524-0014 Northwest Digital Systems | FAX: 206-524-3440 From zmailer-owner@nic.funet.fi Sat Mar 29 17:56:20 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <8088-6434>; Sat, 29 Mar 1997 17:55:05 +0200 Received: by redhotmomma.ssr.com id <36096-17669>; Sat, 29 Mar 1997 11:12:45 -0500 Sender: Scott Ballantyne From: Scott Ballantyne To: mea@nic.funet.fi CC: zmailer@nic.funet.fi In-reply-to: <19970328225435Z1617-388+349@nic.funet.fi> (message from Matti Aarnio on Sat, 29 Mar 1997 00:54:26 +0200 (GMT)) Subject: Re: Bug in smtpserver Message-Id: <97Mar29.111245-0500_est.36096-17669+9@redhotmomma.ssr.com> Date: Sat, 29 Mar 1997 11:12:40 -0500 Return-Path: > > I recall detecting it a while back too -- and fixing it at > 2.99.46. The problem was in openlogfp() routine. > Well, looks like I need to upgrade... Thanks. sdb From zmailer-owner@nic.funet.fi Mon Mar 31 17:45:36 1997 Received: from outbound.Princeton.EDU ([128.112.128.84]) by nic.funet.fi with ESMTP id <3336-14450>; Mon, 31 Mar 1997 17:44:48 +0300 Received: from lists.Princeton.EDU ([128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <541247-21549>; Mon, 31 Mar 1997 09:43:52 -0500 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370048.s1-1>; Mon, 31 Mar 1997 09:43:42 -0500 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id JAA17161 for ; Mon, 31 Mar 1997 09:43:38 -0500 Message-Id: <199703311443.JAA17161@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: scheduler/hold loop Organization: Princeton University, Network Systems, Class. 56 X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 31 Mar 1997 09:43:37 EST From: "Michael R. Gettes" Return-Path: I have a loop in the "hold" transport. It is slowly eating up memory. This is zmailer 2.99.45. The hold process is: write(1, " # h u n g r y\n", 8) = 8 read(0, " # i d l e\n W / J / 5 4".., 8192) = 100 write(1, " # h u n g r y\n", 8) = 8 open("W/J/541381-24826", O_RDWR) = 3 fstat(3, 0xEFFFF230) = 0 mmap(0x00000000, 3322, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0xEF780000 read(3, " i 5 4 1 3 8 1 - 2 4 8".., 3322) = 3322 open("../queue/W/J/541381-24826", O_RDONLY) = 4 fstat(4, 0xEFFFF230) = 0 mmap(0x00000000, 3744, PROT_READ, MAP_SHARED, 4, 0) = 0xEF690000 write(1, " 5 4 1 3 8 1 / 8 8 4\t t".., 138) = 138 write(1, " 5 4 1 3 8 1 / 7 6 0\t t".., 138) = 138 munmap(0xEF690000, 3744) = 0 munmap(0xEF780000, 3322) = 0 close(3) = 0 close(4) = 0 The scheduler process is: read(33, " # h u n g r y\n 5 4 1 3".., 2048) = 292 write(33, " # i d l e\n", 6) = 6 write(2, " s c h e d u l e r : m".., 73) = 73 poll(0xEFFFD478, 1, 0) = 0 poll(0xEFFFD598, 1, 0) = 0 write(33, " W / J / 5 4 1 3 8 1 - 2".., 44) = 44 poll(0xEFFFD598, 1, 0) = 0 poll(0xEFFFD588, 201, 0) = 1 /mrg From zmailer-owner@nic.funet.fi Mon Mar 31 21:53:14 1997 Received: by nic.funet.fi id <4483-17221>; Mon, 31 Mar 1997 21:50:55 +0300 Subject: Re: scheduler/hold loop From: Matti Aarnio To: gettes@princeton.edu (Michael R. Gettes) Date: Mon, 31 Mar 1997 21:50:43 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199703311443.JAA17161@yo.Princeton.EDU> from "Michael R. Gettes" at Mar 31, 97 09:43:37 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2832 Message-Id: <19970331185055Z4483-17221+7345@nic.funet.fi> Return-Path: > > I have a loop in the "hold" transport. > > It is slowly eating up memory. This is zmailer 2.99.45. Nope, at least the "hold" program does not grow. Now if you could get the strace/truss(whatever) to show longer strings, we might learn something :) > The hold process is: > > write(1, " # h u n g r y\n", 8) = 8 > read(0, " # i d l e\n W / J / 5 4".., 8192) = 100 The TA gets at first "#idle\n", then (in same PIPE read()) it gets a TWO-LEVEL HASHED subdirectory pointer to "W/J/54..." > write(1, " # h u n g r y\n", 8) = 8 This is responce to "#idle" -- "Thanks, I got it, but still I am free for a job"... It starts with opening the files, and reading (mmap()ing) them in. > open("W/J/541381-24826", O_RDWR) = 3 > fstat(3, 0xEFFFF230) = 0 > mmap(0x00000000, 3322, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0xEF780000 > read(3, " i 5 4 1 3 8 1 - 2 4 8".., 3322) = 3322 > open("../queue/W/J/541381-24826", O_RDONLY) = 4 > fstat(4, 0xEFFFF230) = 0 > mmap(0x00000000, 3744, PROT_READ, MAP_SHARED, 4, 0) = 0xEF690000 > write(1, " 5 4 1 3 8 1 / 8 8 4\t t".., 138) = 138 > write(1, " 5 4 1 3 8 1 / 7 6 0\t t".., 138) = 138 Hmm.. These are the status report lines, though they are way longer than visible here... > munmap(0xEF690000, 3744) = 0 > munmap(0xEF780000, 3322) = 0 > close(3) = 0 > close(4) = 0 Tail of the processing -- once the responces have been sent, it is time to release the locks, and return the resources. > The scheduler process is: > > read(33, " # h u n g r y\n 5 4 1 3".., 2048) = 292 > write(33, " # i d l e\n", 6) = 6 > write(2, " s c h e d u l e r : m".., 73) = 73 It gets '#hungry', and couple status reports, then it wrote something to the stderr propably relating to this case. > poll(0xEFFFD478, 1, 0) = 0 > poll(0xEFFFD598, 1, 0) = 0 > write(33, " W / J / 5 4 1 3 8 1 - 2".., 44) = 44 > poll(0xEFFFD598, 1, 0) = 0 > poll(0xEFFFD588, 201, 0) = 1 > > /mrg I am rewriting parts of the scheduler-ta interaction protocols, and during it I came to think, that perhaps the "#idle" message should be counted same way as any other job-spec message. Also it looks like the 'hold' channel needs to have deeper understanding of the 'host' concept. It didn't pay attention to the 'host' selector, and perhaps thus became rejected as it received two job-specifiers on same file, and the first one processed them both. (I am not convinced of this explanation, I must do a test...) /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 1 05:10:51 1997 Received: from mercury.ausnetinfo.com.au ([202.14.210.87]) by nic.funet.fi with ESMTP id <14012-7333>; Tue, 1 Apr 1997 05:09:34 +0300 Received: (from mark@localhost) by mercury.ausnetinfo.com.au (8.8.3/8.7.3) id MAA16148 for zmailer@nic.funet.fi; Tue, 1 Apr 1997 12:12:09 +1000 (EST) From: Mark Message-Id: <199704010212.MAA16148@mercury.ausnetinfo.com.au> Subject: Multi domain server not wanting acceptance for user@all.domains To: zmailer@nic.funet.fi Date: Tue, 1 Apr 1997 12:12:09 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Return-Path: Hi, Have searched the archives and install info, nothing really relevant found. One of our clients has zmailer 2_2e10 running on a Solaris 2.5.1 x86 machine. They have it accepting multiple domains properly but have an issue with the mailer accepting email for anne@domain1.com.au, anne@domain2.com.au and anne@domain3.com.au. Anne should only be accepted for anne@domain1.com.au. Anything else shouldn't be allowed and should be bounced back. What is the best thing to do here? A) Alter the scripts so anne@domain1.com.au maps to a login name and then refuse all other anne@*.com.au's? If so, how is this done? B) Have aliases for anne@all.domains and have anne@domain1.com.au map to the login name, and the rest map to some bounce alias? If so, how? The chosen scheme has to be maintainable as more and more domains are added to the machine, so the act of adding a mail alias or full name alias etc should be a once only task that doesn't have to be done again in the future. Although I can live with a perl script that's run after a new alias is added to enforce the desired behaviour. The idea of setting a array of anne@*.com.au to bounce and then plugging in valid users to deliver properly seems a little inelegant. I'd rather have a bounce all rule after the valid aliases have been parsed. It's been a long time since I installed this zmailer on the machine so please be as verbose as you can in any replies. Thanks, Mark mark@ausnetinfo.com.au From zmailer-owner@nic.funet.fi Tue Apr 1 06:18:02 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <14042-18077>; Tue, 1 Apr 1997 06:06:32 +0300 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86856-13321>; Mon, 31 Mar 1997 22:06:17 -0500 Date: Mon, 31 Mar 1997 22:06:13 -0500 (EST) From: Andy Poling To: Mark cc: zmailer@nic.funet.fi Subject: Re: Multi domain server not wanting acceptance for user@all.domains In-Reply-To: <199704010212.MAA16148@mercury.ausnetinfo.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 31 Mar 1997, Mark wrote: > One of our clients has zmailer 2_2e10 running on a Solaris 2.5.1 x86 machine. > They have it accepting multiple domains properly but have an issue with the > mailer accepting email for anne@domain1.com.au, anne@domain2.com.au and > anne@domain3.com.au. Anne should only be accepted for anne@domain1.com.au. > Anything else shouldn't be allowed and should be bounced back. > > What is the best thing to do here? > > A) Alter the scripts so anne@domain1.com.au maps to a login name and then > refuse all other anne@*.com.au's? If so, how is this done? > > B) Have aliases for anne@all.domains and have anne@domain1.com.au map to the > login name, and the rest map to some bounce alias? If so, how? > > The chosen scheme has to be maintainable as more and more domains are added > to the machine, so the act of adding a mail alias or full name alias etc should > be a once only task that doesn't have to be done again in the future. Although > I can live with a perl script that's run after a new alias is added to enforce > the desired behaviour. The idea of setting a array of anne@*.com.au to bounce > and then plugging in valid users to deliver properly seems a little inelegant. > I'd rather have a bounce all rule after the valid aliases have been parsed. I've done this for a client, and while I can't give you the software (it's what they do), I can tell you the strategy that we used. Putting everything inside the MTA is what (IMHO) sunk sendmail. Compartmentalization of function can be a good thing. Don't try to do it all in zmailer. Use the routes database to route mail for your local domains to a seperate delivery agent that you write. Then that agent (similar to "mailbox") has all of your local ugliness in it and zmailer remains basically stock, easily maintainable and update-able. Define your own database of some sort to map all of the usernames in the virtual domains to the proper mailboxes. You're done. Sounds so easy, eh? :-) -Andy PS - Just wait until they want anne@blah and anne@foo to go to two different people... then the fun really begins... PPS - Because of the aforementioned compartmentalization we were able to migrate our "solution" from sendmail to zmailer with little effort when the time came. From zmailer-owner@nic.funet.fi Tue Apr 1 10:59:45 1997 Received: from takthq.lipetsk.su ([193.232.237.17]) by nic.funet.fi with SMTP id <14136-18303>; Tue, 1 Apr 1997 10:58:52 +0300 Received: by takthq.lipetsk.su id IAA16110 (8.6.10 for zmailer@nic.funet.fi); Tue, 1 Apr 1997 08:27:28 +0300 From: Victor Gamov Message-Id: <199704010527.IAA16110@takthq.lipetsk.su> Subject: Problem with aliases to program To: zmailer@nic.funet.fi Date: Tue, 1 Apr 1997 09:27:27 +0400 (MSD) X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Content-Length: 255 Return-Path: Hi! I have problem when I try send mail to user with next record in aliases file: user: "|/usr/local/bin/myprogram" I've got error message: "Mail to program disallowed w/o proper privileges". Any suggestions? -- CU, Victor Gamov From zmailer-owner@nic.funet.fi Tue Apr 1 18:21:04 1997 Received: by nic.funet.fi id <12530-18077>; Tue, 1 Apr 1997 18:20:06 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: April-1st snapshot available... Message-Id: <19970401152006Z12530-18077+2684@nic.funet.fi> Date: Tue, 1 Apr 1997 18:19:57 +0300 Return-Path: ... however not at the usual place. For reasons of my own, I use anonymous-ftp-server at my workstation to keep various transitory files. Right now there are my current snapshow + patch1 to get it to compile at Solaris 2.5.1. Apparently it also runs :) Remember, it is "Work in Progress", and may change without warning. Next test - DEC UNIX 4.0 compilability and runability. Then to continue processing "BUGS"-file. Comments gladly accepted. I will try to stabilize this one before making the "real" release. Oh yes, the location: ftp mea.tmt.tele.fi /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 1 20:26:59 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <6056-7333>; Tue, 1 Apr 1997 20:24:53 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-27144>; Tue, 1 Apr 1997 21:24:00 +0400 Date: Tue, 1 Apr 1997 21:23:52 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: April-1st snapshot available... To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970401152006Z12530-18077+2684@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > For reasons of my own, I use anonymous-ftp-server at my workstation > to keep various transitory files. Right now there are my current > snapshow + patch1 to get it to compile at Solaris 2.5.1. > Apparently it also runs :) I honetstly took it for the fools' day joke. But it compiles and even (seemingly) runs! My question is: WHY?! :-)))) Eugene P.S. It still does not understand the source route "@somewhere:smb@elswhere" From zmailer-owner@nic.funet.fi Tue Apr 1 22:27:54 1997 Received: by nic.funet.fi id <10465-15000>; Tue, 1 Apr 1997 22:24:46 +0300 Subject: Re: April-1st snapshot available... From: Matti Aarnio To: crosser@online.ru Date: Tue, 1 Apr 1997 22:24:34 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 1, 97 09:23:52 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1028 Message-Id: <19970401192446Z10465-15000+3882@nic.funet.fi> Return-Path: > > For reasons of my own, I use anonymous-ftp-server at my workstation > > to keep various transitory files. Right now there are my current > > snapshow + patch1 to get it to compile at Solaris 2.5.1. > > Apparently it also runs :) > > I honetstly took it for the fools' day joke. > > But it compiles and even (seemingly) runs! > > My question is: WHY?! :-)))) Now did I claim it to be a joke ? :-) I have spent the last two weeks at the office on creating infrastructure for a new type of dialup service billing. Until now we have had only phonecall associated time charge, which has been charged on callers phone-bill. Now we introduced a "fixed rate" service, which is billed separately. And no, it isn't "unlimited usage" for USD 20/month, instead it is n hours per month for m FIM, and the extra usage has a time-based charge. > Eugene > > P.S. It still does not understand the source route "@somewhere:smb@elswhere" Logged into "BUGS", so that I hopefully remember to look at it. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 2 04:30:14 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <5189-19430>; Wed, 2 Apr 1997 04:28:42 +0300 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17421-18989>; Tue, 1 Apr 1997 18:27:45 -0700 Date: Tue, 1 Apr 1997 18:27:59 -0700 () From: Daryle Tilroe To: Matti Aarnio cc: crosser@online.ru, zmailer@nic.funet.fi Subject: 2.99.48+p1 compile In-Reply-To: <19970401192446Z10465-15000+3882@nic.funet.fi> Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I finally had a chance to give compiling zmailer under my SunOS 4.1.3_U1 system another whirl. It seems to compile with the following two kludges in dnsgetrr.c: At the beginning: #define HFIXEDSZ 12 #define NETDB_SUCCESS 0 #define NETDB_INTERNAL -1 and in the gethostbyname2_rz function: /*if (!strchr(name, '.') && (cp = __hostalias(name)))*/ if (!strchr(name, '.') ) name = cp; Now my question is should I go ahead and install this (other than the standard risks associated with a pre-release version)? Are these changes going to impact anything other than future "hooks" which are not currently used? Further can anything be done about this in the future for systems that do not use the libresolv that is distributed with zmailer? And finally will these future hooks be compatable with the stock resolver stuff or will I have to go to another OS to run zmailer as a mailserver? Thanks! Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Wed Apr 2 19:11:51 1997 Received: by nic.funet.fi id <15076-6745>; Wed, 2 Apr 1997 19:10:20 +0300 Subject: Re: 2.99.48+p1 compile From: Matti Aarnio To: data@amc.ab.ca (Daryle Tilroe) Date: Wed, 2 Apr 1997 19:10:12 +0300 (EET DST) Cc: crosser@online.ru, zmailer@nic.funet.fi In-Reply-To: from "Daryle Tilroe" at Apr 1, 97 06:27:59 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 4216 Message-Id: <19970402161020Z15076-6745+5434@nic.funet.fi> Return-Path: > I finally had a chance to give compiling zmailer under my > SunOS 4.1.3_U1 system another whirl. It seems to compile with the > following two kludges in dnsgetrr.c: > > At the beginning: > > #define HFIXEDSZ 12 > #define NETDB_SUCCESS 0 > #define NETDB_INTERNAL -1 Solaris 2.5(.0) did need the first of these three, the Solaris 2.5.1 has them all. > and in the gethostbyname2_rz function: > > /*if (!strchr(name, '.') && (cp = __hostalias(name)))*/ > if (!strchr(name, '.') ) > name = cp; Ok, I have to look deeper into the code. Solaris 2.5.1 didn't complain about it, because it apparently has bind 4.9.x resolver library with __hostalias() routine.. > Now my question is should I go ahead and install this (other than the > standard risks associated with a pre-release version)? Are these changes > going to impact anything other than future "hooks" which are not > currently used? These routines should not be affected with anything I will introduce latter. And I am logging your notes for mergeing into my developement source. The use I have in mind: I will introduce an internal shared db-based DNS cache, which the scheduler, and the transport agents can use coherently. Also it will be able to retain (until timeout) the data, and not at all to query the DNS while it has valid data. Previously it was possible to run smtp with '-e' mode, alias 'expensive' which selected all target addresses that can be delivered to same MX host. However the current 'thread grouping' gets the scheduler, and the transporters rather confused on what is going on, if the smtp runs on '-e' mode. Therefore I am looking for ways to get MX information back to the scheduler without compromising scheduler's speed by querying DNS... Consider: *.inet.fi. IN MX 10 mta.inet.fi. IN MX 20 smtp.inet.fi. IN MX 30 smtp.tele.fi. (Ok, '*' in the DNS is rather serious offence on sanity, but it is a hack that the people can live with -- and are successfull!) Now all kinds of (mail-)domains don't need explicitely registered DNS entry can be routed to the mailbox service. Another example: foo.mail.somebody.com. IN MX 10 mailgw.somebody.com. bar.mail.somebody.com. IN MX 10 mailgw.somebody.com. bat.mail.somebody.com. IN MX 10 mailgw.somebody.com. Another example: foo.mail.somebody.com. IN A 1.2.3.4 bar.mail.somebody.com. IN A 1.2.3.4 bat.mail.somebody.com. IN A 1.2.3.4 (yes, I have seen even this!) And propably I saw also this: *.mail.somebody.com. IN A 1.2.3.4 (no MX, just wildcarded plain A..) In both cases the 'channel'+'host' selection that the scheduler uses at the moment causes it to run parallel smtp sessions that end up connecting to the same server. I don't like that type of parallellism, when this would work far better by sending all mail thru same (single) transport-agent. I even got a report from one of you that there are SMTP-servers out there that can do only ONE smtp-session at any time, and still they have several MXes pointing to them.. Problem spins around of questions on how to get the MX info (trivial), how to modify the queues per that info (not trivial?) and how to run the smtp-ta-programs with that in most efficient manner (with least SMTP reconnects/least parallel runners with connections to same remote server.) > Further can anything be done about this in the future for systems that do > not use the libresolv that is distributed with zmailer? And finally will > these future hooks be compatable with the stock resolver stuff or will I > have to go to another OS to run zmailer as a mailserver? Yes, SPARC Linux is most viable alternate :) You can run (propably) all SunOS binaries, most Solaris binaries, and of course have native SparcLinux binaries :) Seriously, this kind of problems surface just because I don't anymore have SunOS 4.1.x available for development. (Same applies to any other OS that I don't test before release...) > Thanks! > > Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca > Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 2 20:28:27 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <13075-28869>; Wed, 2 Apr 1997 20:24:17 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-27144>; Wed, 2 Apr 1997 21:23:53 +0400 Date: Wed, 2 Apr 1997 21:23:43 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: SMTP transport hangs To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: Matti, I'm experiencing this problem for a *long* time, but as this happens rarely, it did not bother me much until recently. Sometimes smtp transport process hangs and stays in the process list for a long time (days). This effectively stops the queue for the domain it is contacting to. The process uses zero CPU and leaves *no* traces in the smtp log. Netstat does not show any open connection on the receiving end (and I cannot check if there is a connection associated with this process on the originating end). Today, I caught such process and killed it with sig 11 (with -QUIT, it leaves no core file?). Zmailer 2.99.47 + all patches, SPARC Solaris 2.5.1. This is what gdb shows: Core was generated by `/usr/zmailer/bin/ta/smtp -s8H -l /var/log/zmailer/smtp'. Program terminated with signal 11, Segmentation fault. procfs (find_procinfo): Couldn't locate pid 0 #0 0xef636d58 in _end () (gdb) bt #0 0xef636d58 in _end () #1 0xef6732cc in _end () #2 0xef6475f8 in _end () #3 0xef6f2158 in _end () #4 0xef6d77c8 in _end () #5 0xef6f1bb8 in _end () #6 0x1fec0 in stachmyaddress (host=0x44fc7 "koi.smtp.online.ru") at selfaddrs.c:296 #7 0x2011c in stachmyaddresses (host=0x44fd9 "") at selfaddrs.c:420 #8 0x16604 in smtpconn (SS=0xeffff998, host=0x454b0 "office.sob.tulane.edu", noMX=0) at smtp.c:1758 #9 0x163b0 in smtpopen (SS=0xeffff998, host=0x454b0 "office.sob.tulane.edu", noMX=0) at smtp.c:1689 #10 0x13e28 in main (argc=0, argv=0x454b0) at smtp.c:669 (gdb) "koi.smtp.online.ru" is one of IP aliases of the local machine. selfaddrs.c:296 is gethostbyname() call. I *can* beleive that this is an error in Solaris... Though this happend in 2.4 as well. Also, I *can* write a wrapper around gethostbyname doing alarm(), but it sounds ugly. Any better ideas? Maybe the scheduler could kill letargic childs? Something else? Eugene From zmailer-owner@nic.funet.fi Wed Apr 2 20:44:42 1997 Received: from alpha.xerox.com ([13.1.64.93]) by nic.funet.fi with SMTP id <4517-19430>; Wed, 2 Apr 1997 20:42:18 +0300 Received: from Charon.PARC.Xerox.xns by alpha.xerox.com via XNS id <16628(5)>; Wed, 2 Apr 1997 09:33:55 PST X-NS-Transport-ID: 0000AA006EEA6200363E Date: Wed, 2 Apr 1997 09:33:16 PST From: Nicholas_Briggs.PARC@xerox.com Subject: Re: 2.99.48+p1 compile In-Reply-to: "mea@nic.funet:fi:Xerox's message of Wed, 2 Apr 1997 08:10:12 PST" To: mea@nic.funet.fi cc: data@amc.ab.ca, crosser@online.ru, zmailer@nic.funet.fi Message-Id: <97Apr2.093355pst."16628(5)"@alpha.xerox.com> Return-Path: In most of the cases you pointed out for multiple domains having the same MX host (or A record, or ...) i'd be willing to specify a channel/host pattern to match it -- thus something like smtp/*.nzl.xerox.com ... smtp -e ... $host I wouldn't expect the scheduler to just automagically detect that anything that is being scheduled on some SMTP channel might possibly be eligible for one of the MX checks. I think that you might want to reconsider creating the DNS cache -- you'll probably end up causing some trouble because while the DNS servers will get notification of zone updates (per the latest implementations) your cache won't. In my situation I can run an authoritative DNS server on the same machine as zmailer is running on, and that server collects its own large cache -- including the negative cache information (which zmailer doesn't have, and we could really do with as an option on some of its databases). Rather than put time into hacking too much more on the ZMailer DNS code, i'd be tempted to suggest that people run a forwarding DNS server, and rely on its cache. \nick From zmailer-owner@nic.funet.fi Wed Apr 2 20:45:22 1997 Received: from amc-edm.amc.ab.ca ([129.128.50.68]) by nic.funet.fi with ESMTP id <12930-15176>; Wed, 2 Apr 1997 20:42:39 +0300 Received: from amc-pc34.amc.ab.ca ([129.128.50.231]) by amc-edm.amc.ab.ca with SMTP id <17411-1649>; Wed, 2 Apr 1997 10:41:45 -0700 Date: Wed, 2 Apr 1997 10:41:54 -0700 () From: Daryle Tilroe Reply-To: Daryle Tilroe To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: 2.99.48+p1 compile In-Reply-To: <19970402161020Z15076-6745+5434@nic.funet.fi> Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 2 Apr 1997, Matti Aarnio wrote: > > I finally had a chance to give compiling zmailer under my > > SunOS 4.1.3_U1 system another whirl. It seems to compile with the > > following two kludges in dnsgetrr.c: > > > > At the beginning: > > > > #define HFIXEDSZ 12 > > #define NETDB_SUCCESS 0 > > #define NETDB_INTERNAL -1 > > Solaris 2.5(.0) did need the first of these three, the > Solaris 2.5.1 has them all. > > > and in the gethostbyname2_rz function: > > > > /*if (!strchr(name, '.') && (cp = __hostalias(name)))*/ > > if (!strchr(name, '.') ) > > name = cp; > > Ok, I have to look deeper into the code. Solaris 2.5.1 didn't > complain about it, because it apparently has bind 4.9.x resolver > library with __hostalias() routine.. > > > Now my question is should I go ahead and install this (other than the > > standard risks associated with a pre-release version)? Are these changes > > going to impact anything other than future "hooks" which are not > > currently used? > > These routines should not be affected with anything I will > introduce latter. And I am logging your notes for mergeing > into my developement source. Thanks for the reply! I am still not quite clear on one point however: Will either of the above changes (especially the removal of the &&) cause present problems? ie. is gethostbyname2_rz used right now? Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca #318, 11315-87 Avenue |----------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-5570 x227 | Home: (403) 439-4743 | Canada /_/\_ | Wfax: (403) 492-1643 | HFax: (403) 988-6454 | T6G 2T9 /__\ |----------------------------------------------------| From zmailer-owner@nic.funet.fi Wed Apr 2 21:38:50 1997 Received: from calzone.oit.unc.edu ([152.2.254.81]) by nic.funet.fi with ESMTP id <11968-28869>; Wed, 2 Apr 1997 21:35:55 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by calzone.oit.unc.edu (8.8.5/8.8.5) with ESMTP id NAA17955 for ; Wed, 2 Apr 1997 13:35:43 -0500 (EST) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id NAA12944 for ; Wed, 2 Apr 1997 13:35:43 -0500 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Wed, 2 Apr 1997 13:35:42 -0500 (EST) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: zmailer@nic.funet.fi Subject: question on using zmailer in a distributed environment Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Howdy, sorry to be asking such basics question to the guru list but I couldn't find out what I wanted in the list archive. Okay, so we've got a bunch of workstations and a big honking server. We want the server to be the only machine receiving mail and have users access their mailboxes via IMAP. If mail gets sent to a user at their workstation, we want the mail to be punted to the honking server. My guess is that defining the following in SiteConfig for the workstations will work: MAILSERVER= sunsite.unc.edu PUNTHOST= sunsite.unc.edu FORCEPUNT= sunsite.unc.edu SMARTHOST= sunsite.unc.edu but it seems a bit overkill. I don't want sunsite to necessarily send out all the mail, just receive it all. Second question: I want as much as possible of the zmailer stuff to live in our shared read-only /usr/local directory. Currently the setup looks like this: MAILBIN= /usr/local/mail/bin MAILSHARE= /usr/local/mail MAILVAR= /usr/local/mail prefix= /usr/local/mail but I strongly suspect that some of these things really _need_ to be local, not shared. Which ones, if any? Penultimate question: We're moving from icky NFS to slightly less icky AFS shortly. How well do zmailer and AFS coexist? Final question: There's another machine on campus which is running zmailer and has an interesting security hole. Commands run via the .forward file mechanism are executed with root permission! I was able to get myself a nice suidroot /bin/sh using my .forward file. Is this a misconfiguration issue or a flaw in zmailer? Thanks in advance for any answers. - donald From zmailer-owner@nic.funet.fi Wed Apr 2 22:05:09 1997 Received: by nic.funet.fi id <15868-19430>; Wed, 2 Apr 1997 22:02:02 +0300 Subject: Re: SMTP transport hangs From: Matti Aarnio To: crosser@online.ru Date: Wed, 2 Apr 1997 22:01:42 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 2, 97 09:23:43 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2778 Message-Id: <19970402190202Z15868-19430+5940@nic.funet.fi> Return-Path: > Matti, > > I'm experiencing this problem for a *long* time, but as this happens > rarely, it did not bother me much until recently. > > Sometimes smtp transport process hangs and stays in the process list for > a long time (days). This effectively stops the queue for the domain it > is contacting to. The process uses zero CPU and leaves *no* traces in > the smtp log. Netstat does not show any open connection on the receiving > end (and I cannot check if there is a connection associated with this > process on the originating end). > > Today, I caught such process and killed it with sig 11 (with -QUIT, it > leaves no core file?). Zmailer 2.99.47 + all patches, SPARC Solaris 2.5.1. > This is what gdb shows: If you started the scheduler with (as fairly usual default) ulimit -c 0 then schedulers children (smtp among them) will not drop cores.. However the trace you present below looks familiar, and I propably did fix it. "patch1c" at my workstation. ("patch1", version "c", on top of the 2.99.48 package there) Several bugs there, memory leaks, and writing past the end of an array... (last of which can cause mysterious errors...) Your debugger does not list symbols from within the dynamic libraries. Have you tried gdb 4.16 ? > Core was generated by `/usr/zmailer/bin/ta/smtp -s8H -l /var/log/zmailer/smtp'. > Program terminated with signal 11, Segmentation fault. > procfs (find_procinfo): Couldn't locate pid 0 > #0 0xef636d58 in _end () > (gdb) bt > #0 0xef636d58 in _end () > #1 0xef6732cc in _end () > #2 0xef6475f8 in _end () > #3 0xef6f2158 in _end () > #4 0xef6d77c8 in _end () > #5 0xef6f1bb8 in _end () > #6 0x1fec0 in stachmyaddress (host=0x44fc7 "koi.smtp.online.ru") > at selfaddrs.c:296 > #7 0x2011c in stachmyaddresses (host=0x44fd9 "") at selfaddrs.c:420 > #8 0x16604 in smtpconn (SS=0xeffff998, host=0x454b0 "office.sob.tulane.edu", > noMX=0) at smtp.c:1758 > #9 0x163b0 in smtpopen (SS=0xeffff998, host=0x454b0 "office.sob.tulane.edu", > noMX=0) at smtp.c:1689 > #10 0x13e28 in main (argc=0, argv=0x454b0) at smtp.c:669 > (gdb) > > "koi.smtp.online.ru" is one of IP aliases of the local machine. > selfaddrs.c:296 is gethostbyname() call. I *can* beleive that this is > an error in Solaris... Though this happend in 2.4 as well. > Also, I *can* write a wrapper around gethostbyname doing alarm(), > but it sounds ugly. > > Any better ideas? > Maybe the scheduler could kill letargic childs? Something else? Umm.. It does have the requisite trace information available for detecting non-active childs... Yes, it is possible. Now what is the longest time any process may be active waiting for a remote system ? (or subprocess in case of local pipe..) > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 2 23:07:33 1997 Received: by nic.funet.fi id <3296-19430>; Wed, 2 Apr 1997 23:06:24 +0300 Subject: Re: 2.99.48+p1 compile From: Matti Aarnio To: Nicholas_Briggs.PARC@xerox.com Date: Wed, 2 Apr 1997 23:06:14 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr2.093355pst."16628(5)"@alpha.xerox.com> from "Nicholas_Briggs.PARC@xerox.com" at Apr 2, 97 09:33:16 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1035 Message-Id: <19970402200624Z3296-19430+6260@nic.funet.fi> Return-Path: > In most of the cases you pointed out for multiple domains having the same > MX host (or A record, or ...) i'd be willing to specify a channel/host > pattern to match it -- thus something like > smtp/*.nzl.xerox.com ... > smtp -e ... $host In this kind of case you could do an explicite routeing via the "routes" file: .nzl.xerox.com smtp!nzlrelay.xerox.com However what I had listed were actual cases pulled from the queues at smtp.inet.fi, and smtp.tele.fi ... ( except names altered to protect the innocents(?) ) Worst offender that we detected, we solved with explicite route definition, but it isn't pretty... > I wouldn't expect the scheduler to just automagically detect that anything > that is being scheduled on some SMTP channel might possibly be eligible for > one of the MX checks. (DNS comments edited) Yes, you have valid comments. I must reconsider these things. (And do some profiling to see which parts are slowest these days.. --> where to put effort. ) ... > \nick /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 2 23:18:45 1997 Received: by nic.funet.fi id <11779-19430>; Wed, 2 Apr 1997 23:15:19 +0300 Subject: Re: 2.99.48+p1 compile From: Matti Aarnio To: data@amc.ab.ca Date: Wed, 2 Apr 1997 23:15:14 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Daryle Tilroe" at Apr 2, 97 10:41:54 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 494 Message-Id: <19970402201519Z11779-19430+6295@nic.funet.fi> Return-Path: > Thanks for the reply! I am still not quite clear on one point however: > Will either of the above changes (especially the removal of the &&) cause > present problems? ie. is gethostbyname2_rz used right now? It does not cause problems, it is not used right now (as some egrep:ing would have shown you :) ) > Daryle A. Tilroe ------------- Thin Film & System Admin ----- data@amc.ab.ca > Alberta Microelectronic Centre ------------------------ http://www.amc.ab.ca /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 2 23:23:29 1997 Received: from curly.oit.unc.edu ([152.2.22.41]) by nic.funet.fi with ESMTP id <12139-19430>; Wed, 2 Apr 1997 23:19:23 +0300 Received: by curly.oit.unc.edu id <231647-925>; Wed, 2 Apr 1997 16:17:13 -0500 Date: Wed, 2 Apr 1997 16:17:07 -0500 (EST) Sender: Eric Riehl From: Eric Riehl X-Sender: eriehl@curly.oit.unc.edu To: zmailer@nic.funet.fi Subject: Small (2 character) bug in smtp.c Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: In the file transports/smtp/smtp.c, near the end of the definition of the getmxrr routine, there's a line that should read: int l = ranny(len-1); instead of int l = ranny(len); That's all. Thanks, Eric Riehl Note: IMHO, you might also want to put the line: mx [0] . host = NULL at the beginning of the function instead of makeing the calling routing always do so. From zmailer-owner@nic.funet.fi Thu Apr 3 02:07:57 1997 Received: from zang.com ([204.119.251.100]) by nic.funet.fi with SMTP id <9200-21057>; Thu, 3 Apr 1997 02:06:28 +0300 Received: (from mark@localhost) by zang.com (8.6.9/zang) id NAA12213; Wed, 2 Apr 1997 13:06:14 -1000 From: Mark (Mookie) Message-Id: <199704022306.NAA12213@zang.com> Subject: Re: SMTP transport hangs To: mea@nic.funet.fi (Matti Aarnio) Date: Wed, 2 Apr 1997 13:06:12 -1000 (HST) Cc: crosser@online.ru, zmailer@nic.funet.fi In-Reply-To: <19970402190202Z15868-19430+5940@nic.funet.fi> from "Matti Aarnio" at Apr 2, 97 10:01:42 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 182 Return-Path: > Your debugger does not list symbols from within the dynamic > libraries. Have you tried gdb 4.16 ? You can get this from sunsite.unc.edu/pub/packages/solaris/sparc Cheers, Mark From zmailer-owner@nic.funet.fi Thu Apr 3 09:49:09 1997 Received: from trixie.mse.jhu.edu ([128.220.8.14]) by nic.funet.fi with ESMTP id <3199-21057>; Thu, 3 Apr 1997 09:47:43 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-31710>; Thu, 3 Apr 1997 01:47:21 -0500 Date: Thu, 3 Apr 1997 01:47:01 -0500 (Eastern Standard Time) From: Tim DiLauro To: zmailer distribution list Subject: smtpserver/router server init call problem Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: [zmailer-2.99.42 on DEC Unix 4.0] I encountered an interesting problem this evening. While fooling around with the smtpserver calls to the router, I noticed strange behavior. When smtpserver uses the router "server" function, it starts by calling server with key="init" with two arguments: a host name (potentially from the HELO command?) and the IP address of the socket. These are the same two values that end up on the "rcvdfrom" line in the router control file. The second argument is at least fairly reliable and I hope to use it as the basis of my filtering. So the problem: the IP address is passed in brackets ("[]"), just as it is in the router control file. Somewhere along the line, this gets evaluated as a "test" command in zmsh. It returns a value of ".". So, instead of getting: pembroke.mse.jhu.edu [128.220.8.107] in $* or $@,I get: pembroke.mse.jhu.edu . which is not very useful. Escaping the open bracket fixed the problem. Diff follows sig. -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 --- smtpserver.c Thu Apr 3 01:22:04 1997 +++ smtpserver.c.orig Mon Nov 18 02:55:20 1996 @@ -2488,7 +2488,7 @@ if (rhostname[strlen(rhostname)-1] == '\\') rhostname[strlen(rhostname)-1] = '?'; - fprintf(tofp, "PS1='%s' ; %s %s '%s' '\\%s'\n", promptbuf, + fprintf(tofp, "PS1='%s' ; %s %s '%s' '%s'\n", promptbuf, ROUTER_SERVER, RKEY_INIT, rhostname, ihostaddr); fflush(tofp); From zmailer-owner@nic.funet.fi Thu Apr 3 10:26:45 1997 Received: from zeronull.aszi.sztaki.hu ([192.84.225.33]) by nic.funet.fi with SMTP id <8241-10376>; Thu, 3 Apr 1997 10:24:16 +0300 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id ib26997; Thu, 3 Apr 1997 09:24:01 +0200 Date: Thu, 3 Apr 1997 09:24:01 +0200 From: Gabor Kiss Message-Id: <199704030724.ib26997@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: Re: smtpserver/router server init call problem X-Charset: US X-Char-Esc: 0 Return-Path: > When smtpserver uses the router "server" function, it starts by calling > server with key="init" with two arguments: a host name (potentially from > the HELO command?) and the IP address of the socket. These are the > same two values that end up on the "rcvdfrom" line in the router > control file. The second argument is at least fairly reliable and I hope > to use it as the basis of my filtering. > > So the problem: the IP address is passed in brackets ("[]"), just as it is > in the router control file. Somewhere along the line, this gets evaluated > as a "test" command in zmsh. It returns a value of ".". So, instead of getting: > Escaping the open bracket fixed the problem. Diff follows sig. > --- smtpserver.c Thu Apr 3 01:22:04 1997 > +++ smtpserver.c.orig Mon Nov 18 02:55:20 1996 > @@ -2488,7 +2488,7 @@ > if (rhostname[strlen(rhostname)-1] == '\\') > rhostname[strlen(rhostname)-1] = '?'; > > - fprintf(tofp, "PS1='%s' ; %s %s '%s' '\\%s'\n", promptbuf, > + fprintf(tofp, "PS1='%s' ; %s %s '%s' '%s'\n", promptbuf, > ROUTER_SERVER, RKEY_INIT, rhostname, ihostaddr); > fflush(tofp); > > > > A more elegant solution: server.cf: case $key in init) # If you want to log incoming connections, it can be done here #echo server "$@" >> /tmp/server # turn off filename globbing <--------------------- set -f <--------------------- add this # redefine the log function log () { } ;; I use this method in my policy based smtp filter without any problems. See details at ftp://ftp.sztaki.hu/pub/private/kissg/zmailer/ Regards Gabor From zmailer-owner@nic.funet.fi Thu Apr 3 10:50:32 1997 Received: from takthq.lipetsk.su ([193.232.237.17]) by nic.funet.fi with SMTP id <3247-21057>; Thu, 3 Apr 1997 10:49:10 +0300 Received: by takthq.lipetsk.su id KAA01830 (8.6.10); Thu, 3 Apr 1997 10:45:39 +0300 From: Victor Gamov Message-Id: <199704030745.KAA01830@takthq.lipetsk.su> Subject: Re: Problem with aliases to program To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Thu, 3 Apr 1997 11:45:39 +0400 (MSD) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at Apr 1, 97 10:48:40 am X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Content-Length: 580 Return-Path: > > On Tue, 1 Apr 1997, Victor Gamov wrote: > > Hi! > > > > I have problem when I try send mail to user with next > > record in aliases file: > > > > user: "|/usr/local/bin/myprogram" > > > > I've got error message: "Mail to program disallowed w/o proper > > privileges". > > > > Any suggestions? > > Yes, check the UID of the user. If it is something in the lower range > (below 100 or so), try upping it. I had this problem before and this > solved it for me. But I want add user in aliases file without adding in passwd file. -- CU, Victor Gamov From zmailer-owner@nic.funet.fi Thu Apr 3 12:51:17 1997 Received: by nic.funet.fi id <7070-6467>; Thu, 3 Apr 1997 12:49:53 +0300 Subject: Re: Problem with aliases to program From: Matti Aarnio To: vit@takthq.lipetsk.su (Victor Gamov) Date: Thu, 3 Apr 1997 12:49:51 +0300 (EET DST) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi In-Reply-To: <199704030745.KAA01830@takthq.lipetsk.su> from "Victor Gamov" at Apr 3, 97 11:45:39 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1264 Message-Id: <19970403094953Z7070-6467+2828@nic.funet.fi> Return-Path: > > On Tue, 1 Apr 1997, Victor Gamov wrote: > > > Hi! > > > > > > I have problem when I try send mail to user with next > > > record in aliases file: > > > user: "|/usr/local/bin/myprogram" > > > I've got error message: "Mail to program disallowed w/o proper > > > privileges". > > > Any suggestions? > > > > Yes, check the UID of the user. If it is something in the lower range > > (below 100 or so), try upping it. I had this problem before and this > > solved it for me. > > But I want add user in aliases file without adding in passwd file. This looks like security problem at the aliases file. Verify protections of: $MAILVAR/db/aliases* 644 or stricter $MAILVAR/db 755 or stricter If those conditions are not met, the address expanded thru the aliases will get privilege of "nobody", which is a magic token to disable several security sensitive operations, like writing to non-mailbox files, or running pipes. On related note, at machines where uid_t is unsigned, the "nobody" account should NOT be with uid -2 ! That MIGHT relate to a "easy to get a root shell" that was noted yesterday. (To know for sure, I do need more details of the setup..) > -- > CU, Victor Gamov /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 3 17:10:45 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <14622-6467>; Thu, 3 Apr 1997 17:07:54 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68045-27151>; Thu, 3 Apr 1997 18:07:44 +0400 Date: Thu, 3 Apr 1997 18:07:38 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: SMTP transport hangs To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970402190202Z15868-19430+5940@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > > Sometimes smtp transport process hangs and stays in the process list for > > a long time (days). > However the trace you present below looks familiar, and > I propably did fix it. "patch1c" at my workstation. > ("patch1", version "c", on top of the 2.99.48 package there) Thanks, I will try it. Happily enough, I found your workstation address in my bash history :) I will report if it helps. > > Maybe the scheduler could kill letargic childs? Something else? > > Umm.. It does have the requisite trace information available > for detecting non-active childs... Yes, it is possible. > > Now what is the longest time any process may be active waiting > for a remote system ? (or subprocess in case of local pipe..) Why not make it a configurable parameter in scheduler.conf? I would set it to 6 - 12 hours for smtp transport... Eugene From zmailer-owner@nic.funet.fi Thu Apr 3 18:59:28 1997 Received: from curly.oit.unc.edu ([152.2.22.41]) by nic.funet.fi with ESMTP id <16371-10376>; Thu, 3 Apr 1997 18:58:01 +0300 Received: by curly.oit.unc.edu id <231648-926>; Thu, 3 Apr 1997 11:55:48 -0500 Date: Thu, 3 Apr 1997 11:55:43 -0500 (EST) Sender: Eric Riehl From: Eric Riehl X-Sender: eriehl@curly.oit.unc.edu To: zmailer@nic.funet.fi Subject: Another bug in smtp.c Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Since smtpwrite calls smtpclose before returning EX_TEMPFAIL, the line in transports/smtp/smtp.c, function ssfgets(buf, bufsiz, inpfp, SS) that reads: if (i != EX_OK) { should read: if ((i != EX_OK) && (i != EX_TEMPFAIL)) { Thanks, Eric Riehl From zmailer-owner@nic.funet.fi Thu Apr 3 22:38:59 1997 Received: by nic.funet.fi id <12246-16787>; Thu, 3 Apr 1997 22:35:27 +0300 Subject: Re: Another bug in smtp.c From: Matti Aarnio To: eriehl@email.unc.edu (Eric Riehl) Date: Thu, 3 Apr 1997 22:35:13 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eric Riehl" at Apr 3, 97 11:55:43 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 656 Message-Id: <19970403193527Z12246-16787+5598@nic.funet.fi> Return-Path: > Since smtpwrite calls smtpclose before returning EX_TEMPFAIL, the line in > transports/smtp/smtp.c, function ssfgets(buf, bufsiz, inpfp, SS) that > reads: > > if (i != EX_OK) { > should read: > if ((i != EX_OK) && (i != EX_TEMPFAIL)) { Yes, that could work there too, but there were some way more serious errors in setjmp()/longjmp() handling. It took me whole day to get all those under control. Well, at least now the timeouts should happen in consistent manner. Everything still at my workstation. ( ftp://mea.tmt.tele.fi/ ) Solaris is -- interesting -- platform to do testing, and debugging. > Thanks, > Eric Riehl /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 06:54:05 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <7318-28091>; Fri, 4 Apr 1997 06:49:52 +0300 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <87177-13322>; Thu, 3 Apr 1997 22:48:06 -0500 Date: Thu, 3 Apr 1997 22:48:01 -0500 (EST) From: Andy Poling To: zmailer@nic.funet.fi Subject: SMTP ignoring MX RR's? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: We've encountered a problem lately with the SMTP client in version 2.99.38. In this specific case, mindspring.com has two MX RR's in place, but the SMTP client still *ocassionally* connects to mindspring.com's A RR to deliver the mail. Mindspring.com's SMTP port just replies 500-something immediately, causing the mail to immediately bounce. I'm using the "-e" option to smtp to do "expensive MX lookups" but it is still happening. Only some of the time though... I cannot discern a pattern to the boo-boo. Does this ring any bells? -Andy From zmailer-owner@nic.funet.fi Fri Apr 4 11:02:22 1997 Received: by nic.funet.fi id <9198-13834>; Fri, 4 Apr 1997 11:01:06 +0300 Subject: Re: Small (2 character) bug in smtp.c From: Matti Aarnio To: eriehl@email.unc.edu (Eric Riehl) Date: Fri, 4 Apr 1997 11:01:03 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eric Riehl" at Apr 2, 97 04:17:07 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 461 Message-Id: <19970404080106Z9198-13834+3006@nic.funet.fi> Return-Path: > In the file transports/smtp/smtp.c, near the end of the definition > of the getmxrr routine, there's a line that should read: > > int l = ranny(len-1); > instead of > int l = ranny(len); Indeed! How did you notice this ? It had escaped me entirely... Andy (and others): This explains the problem you (and I) see at random MX data rejections. > That's all. > > Thanks, > Eric Riehl My sincerest thanks! /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 11:17:47 1997 Received: by nic.funet.fi id <10352-7840>; Fri, 4 Apr 1997 11:16:09 +0300 Subject: Re: SMTP transport hangs From: Matti Aarnio To: crosser@online.ru Date: Fri, 4 Apr 1997 11:16:08 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 3, 97 06:07:38 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1597 Message-Id: <19970404081609Z10352-7840+2948@nic.funet.fi> Return-Path: > > > Sometimes smtp transport process hangs and stays in the process list for > > > a long time (days). > > > However the trace you present below looks familiar, and > > I propably did fix it. "patch1c" at my workstation. > > ("patch1", version "c", on top of the 2.99.48 package there) The naming schemes at my workstation are "challenging", perhaps you can find the lattest base source, and its batches :) Anyway, since this exchange I have rewritten the smtp alarm trappings, and now it works all right, sans one case which appeared today morning. -- Of course the process might have been running with old binary, though I seriously doubt that. I tried to figure out, why that one message had hung on read() of a responce within smtpwrite() of "." -- operation that should bail out after one hour, but had been sitting there for the past 4 hours.. Weird about it was that the process did not react on SIGALRM at all ? Any experience of such mal-functionality among you Solaris users ? (The binary does not have blocking of SIGALRM, which might be sensible thing to do while doing malloc() processing.) > > > Maybe the scheduler could kill letargic childs? Something else? ... > Why not make it a configurable parameter in scheduler.conf? > I would set it to 6 - 12 hours for smtp transport... I have now thought of it back and forth. It is interesting idea, however it is a kludge to handle a problem that sits somewhere else. Although in the case I mention above, it might be usefull backup for non-monitored systems. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 12:09:03 1997 Received: from tmp.idg.nl ([193.78.157.42]) by nic.funet.fi with ESMTP id <8672-28091>; Fri, 4 Apr 1997 12:07:42 +0300 Received: from kovoks.idg.nl (micha.idg.nl [194.178.113.87]) by tmp.idg.nl (8.7.4/970401rk) with SMTP id LAA01951 for ; Fri, 4 Apr 1997 11:06:34 +0200 (MET DST) Message-Id: <3.0.1.32.19970404110235.007cd820@tmp.idg.nl> X-Sender: micha@tmp.idg.nl X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Fri, 04 Apr 1997 11:02:35 +0200 To: zmailer@nic.funet.fi From: Micha Kersloot Subject: Huge amounts of resources Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Return-Path: Hello all, I've just installed Zmailer 2.99.47 on a pentium 75 machine with 32 MB Ram running linux 1.2.13. The problem is that Zmailer take huge amounts of resources when processing a mailing list with 30.000 subscribers. The load average is 32 and the swap-file usage is something like 60 MB's. The machine is also running an ftp-deamon with 50 concurent logins. Has anybody an idee how to restrict zmailer ? Kind regards, Micha Kersloot -------------------------------------------------- IDG Communications E-mail: shareware@idg.nl / micha@idg.nl Voice: +31-23-5461185 Fax: +31-23-5461155 http://www.idg.nl ftp://ftp.idg.nl Upload area: ftp.idg.nl/upload From zmailer-owner@nic.funet.fi Fri Apr 4 16:29:36 1997 Received: by nic.funet.fi id <14448-28091>; Fri, 4 Apr 1997 16:25:43 +0300 Subject: Re: Huge amounts of resources From: Matti Aarnio To: micha@idg.nl (Micha Kersloot) Date: Fri, 4 Apr 1997 16:25:34 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <3.0.1.32.19970404110235.007cd820@tmp.idg.nl> from "Micha Kersloot" at Apr 4, 97 11:02:35 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2101 Message-Id: <19970404132543Z14448-28091+3355@nic.funet.fi> Return-Path: > Hello all, > > I've just installed Zmailer 2.99.47 on a pentium 75 machine with 32 MB Ram > running linux 1.2.13. The problem is that Zmailer take huge amounts of > resources when processing a mailing list with 30.000 subscribers. The load > average is 32 and the swap-file usage is something like 60 MB's. The > machine is also running an ftp-deamon with 50 concurent logins. Has anybody > an idee how to restrict zmailer ? There are many ways on how to do it in scheduler, but which is your problem ? Huge router processes, or too many transporters ? Anyway, the ZMailer is not very good at expanding HUGE lists, a few hundred, or a few thousand will work fine, but tens of thousands is way too much in alias-like expansion. There is a program utils/listexpand.c which may help you. It is intended to be run at a pipe from 'aliases'. Program called "SmartLists" (I think it was) is able to sort, and split the list of addresses into smaller bits, which definitely will help at the routing phase. Depending on how many messages per day you want to send thru, you MAY be able to get it to run on this small a machine, but to give any real recommendations I would need more details about your lists -- how many addresses per list ? How many messages per list per day ? I would suggest that your machine should have at least 128 MB RAM -- a server like yours with less memory is really seriously undersized. Oh yes, you will also need my pre-release edition of ZMailer. It has several bugfixes in it which improve on smtp transporter reliability. Though I need to verbosify some aspects of the log reporting -- right now I don't always know (from the log) WHY some message is being reprocessed again and again.. (One case was a timeout error at DNS query for the target, and of course the message was just logged accepted for processing without details on what was done to it.. The scheduler gets "timeout on DNS query" report, though.) (Pre-releases are at: ftp://mea.tmt.tele.fi/ ) > Kind regards, > Micha Kersloot /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 16:41:41 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <14697-28091>; Fri, 4 Apr 1997 16:39:00 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109518-8953>; Fri, 4 Apr 1997 15:38:39 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <7828-204>; Fri, 4 Apr 1997 15:38:37 +0200 Date: Fri, 4 Apr 1997 15:38:21 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: Re: Huge amounts of resources In-Reply-To: <19970404132543Z14448-28091+3355@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: -----BEGIN PGP SIGNED MESSAGE----- On Fri, 4 Apr 1997, Matti Aarnio wrote: > > Hello all, > > > Oh yes, you will also need my pre-release edition of ZMailer. > It has several bugfixes in it which improve on smtp transporter > reliability. Though I need to verbosify some aspects of the Hi Matti, do you plan to fix 'mailbox' for HP-UX or better for non-MMAP machines aswell ? Also it seems as if newer zmailer versions don't care for MX records with HP-UX. Also some docs about new features (like filterings spam) would be nice. Greets, Sven - -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: latin1 iQCVAwUBM0UEUUgFYR9zZnaVAQHMhAP/UikAPPg4B/xyoq6VutwgfF/q8MP9oyrZ uWs3A+C1a7uq3yWbMsWd8ntRyOLwwJXydd86X73jM8/gmM3d8xSc7FA9JbDci8uj ZTTK0NftYx8ZJRxB0Q6DVbVXTZq3F2OdTW7YOpJXFA54XAKUoAUEtNaGjw6HCNB4 KDhbtqg7km0= =+FB/ -----END PGP SIGNATURE----- From zmailer-owner@nic.funet.fi Fri Apr 4 20:04:02 1997 Received: by nic.funet.fi id <5847-26392>; Fri, 4 Apr 1997 20:02:09 +0300 Subject: Re: Huge amounts of resources From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Fri, 4 Apr 1997 20:01:56 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Apr 4, 97 03:38:21 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1501 Message-Id: <19970404170209Z5847-26392+3726@nic.funet.fi> Return-Path: > > > Hello all, > > > > > Oh yes, you will also need my pre-release edition of ZMailer. > > It has several bugfixes in it which improve on smtp transporter > > reliability. Though I need to verbosify some aspects of the > > Hi Matti, > > do you plan to fix 'mailbox' for HP-UX or better for non-MMAP > machines aswell ? The MMAP (or lack of) does not affect the mailbox program per my quick tests at smtp.tele.fi (Solaris) where I simply removed the "HAVE_MMAP" from the config.h, and then compiled the mailbox program again. My tests at HP-UX 10.20 have been unsuccessfull. In our earlier private exchanges you mentioned the need for GNU gas, binutils, and gcc compiled with them for successfull debugging ? > Also it seems as if newer zmailer versions don't care for MX records > with HP-UX. ?? Perhaps this is one result of the ranny() call error at the smtp program ? It gave about 1/3 change of not giving MXes, if there were two active MXes. > Also some docs about new features (like filterings spam) would > be nice. Filtering of spams is not yet available (aside of Gabor Kiss' solution). However something else that Tele.FI production people wrote because one customer drove them nuts by probeing some external (and nonexistent) address for errors, and wanted to get the bounces to see that their email system is working: autoanswer@smtp.tele.fi That nice pe(a)rl will come to 2.99.48 :-) > Greets, Sven /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 22:31:56 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <5736-13834>; Fri, 4 Apr 1997 22:28:10 +0300 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86895-13322>; Fri, 4 Apr 1997 14:26:05 -0500 Date: Fri, 4 Apr 1997 14:26:00 -0500 (EST) From: Andy Poling To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Anti-spamming (was Re: Huge amounts of resources) In-Reply-To: <19970404170209Z5847-26392+3726@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 4 Apr 1997, Matti Aarnio wrote: > > Also some docs about new features (like filterings spam) would > > be nice. > > Filtering of spams is not yet available (aside of Gabor Kiss' > solution). I made the following changes to the 2.99.38 smtpserver for a WSP who has had increasing problems with spammers using their machines as spam exploders. Believe it or not, to add insult to injury, spammers were using them to send spam selling spam mail software for windoze PC's! Argh! That meta-spam did it for me and I started to work... I (ahem) would, of course, never suggest that you run such a modified SMTP server in real production use. :-) The changes do three things: - after 5 RCPT commands, sleep 5 minutes * numrecipients before ACking RCPT commands - after MAIL command, if there were more than 19 recipients specified, silently drop the spam on the floor - but only after wasting hours accepting their huge long recipient list - after 5 RSET commands in a single session, use the same sleeping strategy as above I believe that the first and third change are actually "legal" since they simply make the SMTP server respond slowly. The second is definitely not legal, but my client wasn't really concerned - they're a WSP, and they don't have to allow large volume mailings. Or just pretend it's a broken M$-MAIL gateway and blame M$... :-) The idea is to recognize behavior patterns of common spamming methodologies and to react, making it very unprofitable to use this SMTP server as an exploder. It will take approximately 7.5 hrs to submit mail to 20 recipients. At 100 recipients (quick calculation: about 8*10**146 hours later) each RCPT is resulting in an 8 hr wait. Needless to say, this will not appeal to a spammer as a high-performance delivery method. :-) It also prevents that spammer from loading down your system by hammering it with their drivel. These changes combat two types of spammers we were seeing - those that send a single message to N (20 to 1000) addresses, and those that send N (20 to 10000) identical and individually addressed messages in a single SMTP session/connection. The second type of spamming appeared after we made the recipient counting change (which actually surprised me - they actually adapted their methods in two days - scary), necessitating the RSET counter. I'm sure we'll soon have to add a simple database to track abusers and stop them. I'm just providing this as an example of the horrid gruesome hacks you can use to combat spammers. I reiterate that I do not recommend the use of this code on a production server. No, certainly not... that would be irresponsible... Context diff for 2.99.38 smtpserver/smtpserver.c below, though you might be able to apply it against more recent versions. Attached below that is a simple shell script (we call it show_senders) that will perform a crude scan for the biggest senders (by msg count - not recipient count) on your router log. Edit it to change the router log location to that for your system before test-driving it. It's crude and dumb, but it will give you an idea what's going on. Attached below it is sample output from one of our systems early in the day (before the numbers get too big). Before I modified the smtpserver, we used to see numbers like 4160 for certain bogus spamming addresses in the output of this shell script... now they're not there at all. Success! -Andy PS - These smtpserver changes also give you leisurely *hours* to track down the offending spammer's ISP and catch the offender red-handed if you're so inclined... ============================================================================== *** smtpserver.c.orig Tue Apr 1 00:12:51 1997 --- smtpserver.c Fri Apr 4 13:40:10 1997 *************** *** 191,196 **** --- 191,197 ---- int pid, routerpid; FILE *logfp = NULL; int D_alloc = 0; + int ResetsSeen = 0; static char *m200 = "2.0.0"; static char *m400 = "4.0.0"; *************** *** 702,707 **** --- 703,711 ---- if (msgfd > 1) close(msgfd); + /* anti-spamming measure - ASP/MSI */ + ResetsSeen = 0; + smtpserver(1); if (routerpid > 0) killr(routerpid); *************** *** 1400,1413 **** } else if (sizeoptsum > availspace) { type(452, "4.3.1", "insufficient storage space, try again later"); } else if (s) { ! type(atoi(s), s+4, "Ok", "Ok"); ! ++rcpt_count; } else { ! if (sizeoptval) ! type(250, "2.1.5", "Ok; can accomodate %d byte message",sizeoptval); else ! type(250, "2.1.5", (char *)NULL); ! ++rcpt_count; } if (s) free(s); --- 1404,1440 ---- } else if (sizeoptsum > availspace) { type(452, "4.3.1", "insufficient storage space, try again later"); } else if (s) { ! /* anti-spamming measures - ASP/MSI */ ! if (rcpt_count >= 4) ! { ! if (rcpt_count++ > 4) ! sleep(rcpt_count * 5 * 60); /* sleep 5 minutes * numrecipients */ ! type(atoi(s), s+4, "Ok spammer... now go away", "Ok spammer... now go away"); ! } ! else ! { ! type(atoi(s), s+4, "Ok", "Ok"); ! ++rcpt_count; ! } } else { ! /* anti-spamming measures - ASP/MSI */ ! if (rcpt_count >= 4) ! { ! if (rcpt_count++ > 4) ! sleep(rcpt_count * 5 * 60); /* sleep 5 minutes * numrecipients */ ! if (sizeoptval) ! type(250, "2.1.5", "Ok spammer... now go away; can accomodate %d byte message",sizeoptval); ! else ! type(250, "2.1.5", "Ok spammer... now go away"); ! } else ! { ! if (sizeoptval) ! type(250, "2.1.5", "Ok; can accomodate %d byte message",sizeoptval); ! else ! type(250, "2.1.5", (char *)NULL); ! ++rcpt_count; ! } } if (s) free(s); *************** *** 1522,1527 **** --- 1549,1565 ---- mfp = NULL; type(552, "5.3.4", "Message size exceeded fixed maximum size for acceptable email"); fflush(stdout); + } else if (rcpt_count >= 19) { + /* anti-spamming measure - ASP/MSI */ + mail_abort(mfp); + mfp = NULL; + #ifdef LOG_INFO + syslog(LOG_INFO, + "pretended to accept (but silently dropped) spam (%drecipients>19) from %s:%d", + rcpt_count, rhostname, rport); + #endif + type(250, "2.6.0", "I thought I told you to go away..."); + fflush(stdout); } else if (mail_close(mfp) == EOF) { type(452,m430,(char *)NULL); fflush(stdout); *************** *** 1923,1928 **** --- 1961,1975 ---- if (mfp != NULL) { mail_abort(mfp); mfp = NULL; + } + /* anti-spamming measure - ASP/MSI */ + if (++ResetsSeen > 4) + { + #ifdef LOG_INFO + syslog(LOG_INFO, "multiple (%d) RSET cmds in one session from %s:%d", + ResetsSeen, rhostname, rport); + #endif + sleep(5 * 60 * ResetsSeen); } if (state != Hello) state = MailOrHello; ============================================================================== #!/bin/sh nawk ' ($2 == "file:") { x = 4; while ( $x != "=>" && x < 20) { printf ("%s ", $x); x++; } printf ("\n"); } ' < /var/adm/mail/router \ | sort | uniq -c | sort -rn | head -23 ============================================================================== 68 owner-freebsd-questions@freefall.freebsd.org 66 MailProcessor@lasergo.com 49 48 48 <70741.3507@CompuServe.COM> 34 staff@quote.com 34 34 33 23 owner-year2000-discuss@year2000.com 23 21 20 scopus-users-owner@gateway.scopus.com 19 rmoore@fourthtier.com 19 Stephen Oskoui 16 otle@chat.carleton.ca 16 communicate@jit.com 15 error 15 London Virtual Mall 14 webmaster-l-owner@cybercorp.net 14 philhbh@fourthtier.com 14 cte@rammaps.com 14 From zmailer-owner@nic.funet.fi Fri Apr 4 23:06:03 1997 Received: from alpha.xerox.com ([13.1.64.93]) by nic.funet.fi with SMTP id <14910-26392>; Fri, 4 Apr 1997 23:03:31 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <18541(3)>; Fri, 4 Apr 1997 12:00:48 PST X-NS-Transport-ID: 0800200AF21F36ED9C7D Date: Fri, 4 Apr 1997 11:59:12 PST From: Nicholas_Briggs.PARC@xerox.com Subject: Re: Anti-spamming (was Re: Huge amounts of resources) In-Reply-to: "andy@realbig:com:Xerox's message of Fri, 4 Apr 1997 11:26:00 PST" To: andy@realbig.com cc: mea@nic.funet.fi, zmailer@nic.funet.fi Message-Id: <97Apr4.120048pst."18541(3)"@alpha.xerox.com> Return-Path: I watch the bounces -- since the spammers almost always have bogus Xerox addresses on their list, and I have a (small, but growing) list of addresses I just refuse to accept mail from (both domains for the HELO, and envelope From addresses). The latest trick i've seen is that they are now sending MAIL FROM:<> which you can't afford to block, if you expect your users to get back error messages. That combined with the "drop 20-1000 messages on each server you can find" trick is getting really annoying. \nick From zmailer-owner@nic.funet.fi Fri Apr 4 23:22:47 1997 Received: by nic.funet.fi id <5601-13834>; Fri, 4 Apr 1997 23:20:55 +0300 Subject: Re: Anti-spamming (was Re: Huge amounts of resources) From: Matti Aarnio To: Nicholas_Briggs.PARC@xerox.com Date: Fri, 4 Apr 1997 23:20:44 +0300 (EET DST) Cc: andy@realbig.com, zmailer@nic.funet.fi In-Reply-To: <97Apr4.120048pst."18541(3)"@alpha.xerox.com> from "Nicholas_Briggs.PARC@xerox.com" at Apr 4, 97 11:59:12 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 948 Message-Id: <19970404202055Z5601-13834+4598@nic.funet.fi> Return-Path: > I watch the bounces -- since the spammers almost always have bogus Xerox > addresses on their list, and I have a (small, but growing) list of > addresses I just refuse to accept mail from (both domains for the HELO, > and envelope From addresses). Gabor's ideas help here too quite nicely -- I just need to get them codified without running router in parallel.. > The latest trick i've seen is that they are now sending > MAIL FROM:<> > which you can't afford to block, if you expect your users to get back > error messages. That combined with the "drop 20-1000 messages on each > server you can find" trick is getting really annoying. Actually why not ? If the incoming address is a "box", and there are more than N ( as low as 2 ? 3 ? ) recipients, start rejecting the message. As my luck runs, people of course have multiple non-local people as postmasters, and all get copies ? > \nick /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 4 23:34:46 1997 Received: from netcreations.com ([206.243.201.4]) by nic.funet.fi with ESMTP id <5730-13834>; Fri, 4 Apr 1997 23:32:03 +0300 Received: from rscott.nc ([206.243.201.10]) by NetCreations.com with SMTP id <4523232-8105>; Fri, 4 Apr 1997 15:19:32 -0500 Comments: Authenticated sender is From: "Ryan Scott" Organization: NetCreations, Inc To: Andy Poling Date: Fri, 4 Apr 1997 14:53:46 +0000 Subject: Re: Anti-spamming (was Re: Huge amounts of resources) Reply-to: rscott@netcreations.com CC: zmailer@nic.funet.fi Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-Id: <97Apr4.151932-0500_est.4523232-8105+162@NetCreations.com> Return-Path: from what I understand, using someone elses server to send out your mail without their permission is called Theft of Service and is quite illegal. You are under no obligation to deliver such mail. So dropping it on the floor is not illegal. As a matter of fact, if you can get a snail mail address for the offending party, and you can have your lawyer send a cease and desist, and it continues to happen, the sheriffs dept will go in and take their machines. We're attempting to get that to happen with one of CyberPromotions' clients, the sandkconsulting scamspam. I'm not a lawyer, but this is what my lawyer tells me. Just some fyi. We've not delivered lots of their mail, kept it for evidence. We'll continue to do so. Ryan Scott NetCreations, Inc http://www.netcreations.com/ > On Fri, 4 Apr 1997, Matti Aarnio wrote: > > > Also some docs about new features (like filterings spam) would > > > be nice. > > > > Filtering of spams is not yet available (aside of Gabor Kiss' > > solution). > > I made the following changes to the 2.99.38 smtpserver for a WSP who > has had increasing problems with spammers using their machines as > spam exploders. Believe it or not, to add insult to injury, spammers > were using them to send spam selling spam mail software for windoze > PC's! Argh! > > That meta-spam did it for me and I started to work... > > I (ahem) would, of course, never suggest that you run such a > modified SMTP server in real production use. :-) > > The changes do three things: > > - after 5 RCPT commands, sleep 5 minutes * numrecipients before > ACking RCPT commands > > - after MAIL command, if there were more than 19 recipients > specified, silently drop the spam on the floor - but only > after wasting hours accepting their huge long recipient list > > - after 5 RSET commands in a single session, use the same sleeping > strategy as above > > I believe that the first and third change are actually "legal" since > they simply make the SMTP server respond slowly. The second is > definitely not legal, but my client wasn't really concerned - > they're a WSP, and they don't have to allow large volume mailings. > Or just pretend it's a broken M$-MAIL gateway and blame M$... :-) > > The idea is to recognize behavior patterns of common spamming > methodologies and to react, making it very unprofitable to use this > SMTP server as an exploder. It will take approximately 7.5 hrs to > submit mail to 20 recipients. At 100 recipients (quick calculation: > about 8*10**146 hours later) each RCPT is resulting in an 8 hr wait. > Needless to say, this will not appeal to a spammer as a > high-performance delivery method. :-) It also prevents that spammer > from loading down your system by hammering it with their drivel. > > These changes combat two types of spammers we were seeing - those > that send a single message to N (20 to 1000) addresses, and those > that send N (20 to 10000) identical and individually addressed > messages in a single SMTP session/connection. > > The second type of spamming appeared after we made the recipient > counting change (which actually surprised me - they actually adapted > their methods in two days - scary), necessitating the RSET counter. > I'm sure we'll soon have to add a simple database to track abusers > and stop them. > > I'm just providing this as an example of the horrid gruesome hacks > you can use to combat spammers. I reiterate that I do not recommend > the use of this code on a production server. No, certainly not... > that would be irresponsible... > > Context diff for 2.99.38 smtpserver/smtpserver.c below, though you > might be able to apply it against more recent versions. > > Attached below that is a simple shell script (we call it > show_senders) that will perform a crude scan for the biggest senders > (by msg count - not recipient count) on your router log. Edit it to > change the router log location to that for your system before > test-driving it. It's crude and dumb, but it will give you an idea > what's going on. Attached below it is sample output from one of our > systems early in the day (before the numbers get too big). Before I > modified the smtpserver, we used to see numbers like 4160 for > certain bogus spamming addresses in the output of this shell > script... now they're not there at all. Success! > > -Andy > > PS - These smtpserver changes also give you leisurely *hours* to > track down the offending spammer's ISP and catch the offender > red-handed if you're so inclined... > > ==================================================================== > ========= > > *** smtpserver.c.orig Tue Apr 1 00:12:51 1997 > --- smtpserver.c Fri Apr 4 13:40:10 1997 > *************** > *** 191,196 **** > --- 191,197 ---- > int pid, routerpid; > FILE *logfp = NULL; > int D_alloc = 0; > + int ResetsSeen = 0; > > static char *m200 = "2.0.0"; > static char *m400 = "4.0.0"; > *************** > *** 702,707 **** > --- 703,711 ---- > if (msgfd > 1) > close(msgfd); > > + /* anti-spamming measure - ASP/MSI */ > + ResetsSeen = 0; > + > smtpserver(1); > if (routerpid > 0) > killr(routerpid); > *************** > *** 1400,1413 **** > } else if (sizeoptsum > availspace) { > type(452, "4.3.1", "insufficient storage space, try again > later"); > } else if (s) { > ! type(atoi(s), s+4, "Ok", "Ok"); > ! ++rcpt_count; > } else { > ! if (sizeoptval) > ! type(250, "2.1.5", "Ok; can accomodate %d byte > ! message",sizeoptval); > else > ! type(250, "2.1.5", (char *)NULL); > ! ++rcpt_count; > } > if (s) > free(s); > --- 1404,1440 ---- > } else if (sizeoptsum > availspace) { > type(452, "4.3.1", "insufficient storage space, try again > later"); > } else if (s) { > ! /* anti-spamming measures - ASP/MSI */ > ! if (rcpt_count >= 4) > ! { > ! if (rcpt_count++ > 4) > ! sleep(rcpt_count * 5 * 60); /* sleep 5 minutes * numrecipients > ! */ type(atoi(s), s+4, "Ok spammer... now go away", "Ok > ! spammer... now go away"); } else { type(atoi(s), s+4, > ! "Ok", "Ok"); ++rcpt_count; } > } else { > ! /* anti-spamming measures - ASP/MSI */ > ! if (rcpt_count >= 4) > ! { > ! if (rcpt_count++ > 4) > ! sleep(rcpt_count * 5 * 60); /* sleep 5 minutes * numrecipients > ! */ if (sizeoptval) type(250, "2.1.5", "Ok spammer... now go > ! away; can accomodate %d byte message",sizeoptval); else > ! type(250, "2.1.5", "Ok spammer... now go away"); } > else > ! { > ! if (sizeoptval) > ! type(250, "2.1.5", "Ok; can accomodate %d byte > ! message",sizeoptval); else type(250, "2.1.5", (char *)NULL); > ! ++rcpt_count; } > } > if (s) > free(s); > *************** > *** 1522,1527 **** > --- 1549,1565 ---- > mfp = NULL; > type(552, "5.3.4", "Message size exceeded fixed maximum size > for acceptable email"); fflush(stdout); > + } else if (rcpt_count >= 19) { > + /* anti-spamming measure - ASP/MSI */ > + mail_abort(mfp); > + mfp = NULL; > + #ifdef LOG_INFO > + syslog(LOG_INFO, > + "pretended to accept (but silently dropped) spam > + (%drecipients>19) from %s:%d", rcpt_count, rhostname, > + rport); > + #endif > + type(250, "2.6.0", "I thought I told you to go away..."); > + fflush(stdout); > } else if (mail_close(mfp) == EOF) { > type(452,m430,(char *)NULL); > fflush(stdout); > *************** > *** 1923,1928 **** > --- 1961,1975 ---- > if (mfp != NULL) { > mail_abort(mfp); > mfp = NULL; > + } > + /* anti-spamming measure - ASP/MSI */ > + if (++ResetsSeen > 4) > + { > + #ifdef LOG_INFO > + syslog(LOG_INFO, "multiple (%d) RSET cmds in one session from > + %s:%d", ResetsSeen, rhostname, rport); > + #endif > + sleep(5 * 60 * ResetsSeen); > } > if (state != Hello) > state = MailOrHello; > > > ==================================================================== > ========= > > #!/bin/sh > > nawk ' > ($2 == "file:") { > x = 4; > while ( $x != "=>" && x < 20) > { > printf ("%s ", $x); > x++; > } > printf ("\n"); > } > ' < /var/adm/mail/router \ > | sort | uniq -c | sort -rn | head -23 > > > ==================================================================== > ========= > > 68 owner-freebsd-questions@freefall.freebsd.org > 66 MailProcessor@lasergo.com > 49 > 48 > 48 <70741.3507@CompuServe.COM> > 34 staff@quote.com > 34 > 34 > 33 > 23 owner-year2000-discuss@year2000.com > 23 > 21 > 20 scopus-users-owner@gateway.scopus.com > 19 rmoore@fourthtier.com > 19 Stephen Oskoui > 16 otle@chat.carleton.ca > 16 communicate@jit.com > 15 error > 15 London Virtual Mall > 14 webmaster-l-owner@cybercorp.net > 14 philhbh@fourthtier.com > 14 cte@rammaps.com > 14 ______________________________________________________________________ Ryan Scott - rscott@netcreations.com - 718 522 1531 - Reinventing Direct Marketing on the Net - NetCreations, Inc - http://www.netcreations.com/ - Targeted 100% OPT IN Email - PostMaster Direct Response - http://www.postmasterdirect.com/ *be sure to quote me in your reply* From zmailer-owner@nic.funet.fi Sat Apr 5 00:02:54 1997 Received: from ws.realbig.com ([206.161.253.65]) by nic.funet.fi with ESMTP id <6122-26392>; Sat, 5 Apr 1997 00:00:31 +0300 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86908-13325>; Fri, 4 Apr 1997 15:40:05 -0500 Date: Fri, 4 Apr 1997 15:39:59 -0500 (EST) From: Andy Poling To: Nicholas_Briggs.PARC@xerox.com cc: mea@nic.funet.fi, zmailer@nic.funet.fi Subject: Re: Anti-spamming (was Re: Huge amounts of resources) In-Reply-To: <97Apr4.120048pst."18541(3)"@alpha.xerox.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 4 Apr 1997 Nicholas_Briggs.PARC@xerox.com wrote: > I watch the bounces -- since the spammers almost always have bogus Xerox > addresses on their list, and I have a (small, but growing) list of addresses I > just refuse to accept mail from (both domains for the HELO, and envelope From > addresses). Our spammers sometimes use real addresses from the domains they're spamming through as the sender. We thus cannot block them that way. > The latest trick i've seen is that they are now sending > MAIL FROM:<> Yes - I've seen growing use of this. Pretty soon it's going to get alot harder to stop them. Especially since there is now a growing industry of increasingly clever/obnoxious software that will allow any idiot to drive us insane. I am firmly committed, however, to two goals. First I want to concentrate on defeating the spam-specific software since I think that's a industry that should just *die* a flaming death (and I'll be more than happy to help it along). Second I want to prevent more amateurish attempts at mail and mailbox abuse. If anyone has an idea how we can figure out how the spamming software works without buying it, I'd love to hear it. There is no substitute for studying the target software, but I really don't want to make them rich (plus some packages are like $300US). -Andy From zmailer-owner@nic.funet.fi Sat Apr 5 01:09:36 1997 Received: from netcreations.com ([206.243.201.4]) by nic.funet.fi with ESMTP id <5276-23120>; Sat, 5 Apr 1997 01:08:29 +0300 Received: from rscott.nc ([206.243.201.10]) by NetCreations.com with SMTP id <4611206-8106>; Fri, 4 Apr 1997 16:56:02 -0500 Comments: Authenticated sender is From: "Ryan Scott" Organization: NetCreations, Inc To: Andy Poling Date: Fri, 4 Apr 1997 16:30:08 +0000 Subject: Spamming reports wanted Reply-to: rscott@netcreations.com CC: mea@nic.funet.fi, zmailer@nic.funet.fi Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-Id: <97Apr4.165602-0500_est.4611206-8106+257@NetCreations.com> Return-Path: > If anyone has an idea how we can figure out how the spamming > software works without buying it, I'd love to hear it. There is no > substitute for studying the target software, but I really don't want > to make them rich (plus some packages are like $300US). > > -Andy I may have a copy of one such piece of software shortly and will let you know. We're going to attack this problem from the legal end. Apparently the FTC is very much interested in these issues and it looks doubtful they are going to support spammers tactics. While this will not necessarily prevent the hardcore illegal spammer, the true criminal, it will put a major chill in the many thousands of spammers out there if the $500 unsolicited fax fine gets transfered online (for real), and/or it gets easier to procecute those who steal service and bounce their mail through other servers. Put one of two of these guys in jail and you'll see theft of service attacks drop off in a big way. The FTC is VERY interested in getting real life stories about system abuse via spam. If you have any information to supply, PLEASE send it to me. We're going to be generating a lot of media noise about this. The stories really help the media and the FTC. Almost no one in the real world knows how much work it is when your system gets attacked like this. Because of the reputation we're trying to build as 100% opt-in emailers (never spamming), I had to reply to all complaints we got from the recent spam someone sent to let them know that we had nothing to do with it. I've gotten about 100 complaints so far. One thing that is particularly annoying is the people getting spammed feeling compelled to forward the spam to the postmaster. Ok, maybe most sysadmins don't look at that mail, I don't know. But its as bad as the spamming. I got one spam, and 100 complaints about spam. I can't blame them for trying, though. Anyway, send me real world stories of damage or work you need to do to prevent that damage. It will make a big difference. Ryan ______________________________________________________________________ Ryan Scott - rscott@netcreations.com - 718 522 1531 - Reinventing Direct Marketing on the Net - NetCreations, Inc - http://www.netcreations.com/ - Targeted 100% OPT IN Email - PostMaster Direct Response - http://www.postmasterdirect.com/ *be sure to quote me in your reply* From zmailer-owner@nic.funet.fi Sat Apr 5 18:33:22 1997 Received: from press-gopher.uchicago.edu ([128.135.234.36]) by nic.funet.fi with ESMTP id <3152-574>; Sat, 5 Apr 1997 18:32:16 +0300 Received: from localhost by press-gopher.uchicago.edu with SMTP id <26448-16289>; Sat, 5 Apr 1997 09:31:59 -0600 Date: Sat, 5 Apr 1997 09:31:44 -0600 (CST) Sender: Roy C Bixler From: Roy C Bixler To: zmailer@nic.funet.fi Subject: Procmail as local delivery agent In-Reply-To: <97Apr4.165602-0500_est.4611206-8106+257@NetCreations.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I have just been tinkering with setting up 'procmail' as the local mail delivery agent using the example entries in 'scheduler.conf' and 'sm.conf'. However, I find it fails to deliver local mail since the '$u' in 'sm.conf' expands to 'user@host' and 'procmail' is called like this: procmail -d user@host If I remove the '@host' part, the delivery is successful. But if the message gets to the 'procmail' delivery method at all, then it's already local and '@host' is superfluous anyway. Is there any way in 'sm.conf' to specify to pass only the username to 'procmail'? Roy rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Sat Apr 5 19:35:56 1997 Received: by nic.funet.fi id <3238-23120>; Sat, 5 Apr 1997 19:35:01 +0300 Subject: Re: Procmail as local delivery agent From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Sat, 5 Apr 1997 19:34:52 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at Apr 5, 97 09:31:44 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 947 Message-Id: <19970405163501Z3238-23120+3309@nic.funet.fi> Return-Path: > I have just been tinkering with setting up 'procmail' as the local mail > delivery agent using the example entries in 'scheduler.conf' and > 'sm.conf'. However, I find it fails to deliver local mail since the '$u' > in 'sm.conf' expands to 'user@host' and 'procmail' is called like this: > > procmail -d user@host > > If I remove the '@host' part, the delivery is successful. But if the > message gets to the 'procmail' delivery method at all, then it's already > local and '@host' is superfluous anyway. Is there any way in 'sm.conf' to > specify to pass only the username to 'procmail'? See router configuration files: crossbar.cf Around line 128 these is a multi-choice label for "local". If you remove that instance, and activate (uncomment) the "uucp|local)" in place of "uucp)", then you should get plain "users" into the recipient address. > Roy > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Sun Apr 6 23:09:39 1997 Received: from trixie.mse.jhu.edu ([128.220.8.14]) by nic.funet.fi with ESMTP id <7081-15946>; Sun, 6 Apr 1997 23:06:36 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5311>; Sun, 6 Apr 1997 16:06:33 -0400 Date: Sun, 6 Apr 1997 16:06:12 -0400 (Eastern Daylight Time) From: Tim DiLauro To: zmailer distribution list Subject: makendbm minor, but NASTY, error Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: [zmailer-2.99.42] BEWARE! makendbm has a minor error that smacks you upside the head if config.h has both HAVE_NDBM_H and HAVE_DB_H. The error causes the database file (the one that has all your data in it) to be truncated. That's because makedbm calls dbm_open() for the NDBM support(which creates the empty .pag and .dir files) and then procedes to call dbopen() for the BTREE support (which creams the input file). Diff follows... ======================================================================= --- makendbm.c Sun Apr 6 15:53:50 1997 +++ makendbm.c.orig Sun Apr 6 14:09:39 1997 @@ -221,7 +221,7 @@ } #endif #ifdef HAVE_DB_H - if (typ == 3) { + if (typ == 1) { dbfile = dbopen(dbasename, O_RDWR|O_CREAT|O_TRUNC, 0644, DB_BTREE, NULL); dbf = dbfile; ======================================================================= -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Mon Apr 7 00:41:59 1997 Received: by nic.funet.fi id <17880-15705>; Mon, 7 Apr 1997 00:40:18 +0300 Subject: Re: makendbm minor, but NASTY, error From: Matti Aarnio To: timmo@RacerX.mse.jhu.edu (Tim DiLauro) Date: Mon, 7 Apr 1997 00:40:15 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Tim DiLauro" at Apr 6, 97 04:06:12 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1519 Message-Id: <19970406214018Z17880-15705+242@nic.funet.fi> Return-Path: > [zmailer-2.99.42] Thanks for mentioning the version. > BEWARE! makendbm has a minor error that smacks you upside the head if > config.h has both HAVE_NDBM_H and HAVE_DB_H. The error causes the > database file (the one that has all your data in it) to be truncated. > > That's because makedbm calls dbm_open() for the NDBM support(which creates > the empty .pag and .dir files) and then procedes to call dbopen() for the > BTREE support (which creams the input file). This one took me a while to understand. Mainly because the diff below is reversed! Yes indeed there was an error, I have ChangeLogged entries on 2nd of Dec -96 about NDBM/BTREE parametrization error in the makendbm.c It has been fixed since 2.99.44. /Matti Aarnio > Diff follows... > > ======================================================================= > --- makendbm.c Sun Apr 6 15:53:50 1997 > +++ makendbm.c.orig Sun Apr 6 14:09:39 1997 > @@ -221,7 +221,7 @@ > } > #endif > #ifdef HAVE_DB_H > - if (typ == 3) { > + if (typ == 1) { > dbfile = dbopen(dbasename, O_RDWR|O_CREAT|O_TRUNC, 0644, > DB_BTREE, NULL); > dbf = dbfile; > ======================================================================= > > -timmo > > Tim DiLauro Milton S. Eisenhower Library > Library Systems Jack Johns Hopkins University > (410) 516-5263 3400 N. Charles Street > timmo@jhu.edu Baltimore, MD 21218 > > > From zmailer-owner@nic.funet.fi Mon Apr 7 10:29:11 1997 Received: from zeronull.aszi.sztaki.hu ([193.225.12.33]) by nic.funet.fi with SMTP id <16022-15705>; Mon, 7 Apr 1997 10:27:12 +0300 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id OK06585; Mon, 7 Apr 1997 09:27:06 +0200 Date: Mon, 7 Apr 1997 09:27:06 +0200 From: Gabor Kiss Message-Id: <199704070727.OK06585@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: Re: Anti-spamming (was Re: Huge amounts of resources) X-Charset: US X-Char-Esc: 0 Return-Path: > > I watch the bounces -- since the spammers almost always have bogus Xerox > > addresses on their list, and I have a (small, but growing) list of > > addresses I just refuse to accept mail from (both domains for the HELO, > > and envelope From addresses). > > Gabor's ideas help here too quite nicely -- I just need > to get them codified without running router in parallel.. Just a reminder. The URL is ftp://ftp.sztaki.hu/pub/private/kissg/zmailer Gabor From zmailer-owner@nic.funet.fi Mon Apr 7 22:33:21 1997 Received: by nic.funet.fi id <1797-13604>; Mon, 7 Apr 1997 22:31:47 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: weekend datadump at mea.tmt.tele.fi Message-Id: <19970407193147Z1797-13604+68@nic.funet.fi> Date: Mon, 7 Apr 1997 22:31:44 +0300 Return-Path: Hello, I downloaded my weekend products to my office workstation. I am still not happy, but that does not mean it is bad -- just that I try to achieve perfectionism (sometimes).. At Solaris I see the smtp-clients occasionally freezing, and not reacting on SIGALRM as if it were set a block on the signal. (That is: 'truss -p nn' does not see the SIGALRM signal to the process :-? ) The smtpserver's policy filtering is now mere hooks, no actual code behind them.. Documents are -- not updated to all of the new features. That is, man-pages. (The main document is for some different version of the software :) ) Any comments (about the package) ? /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Apr 7 22:52:49 1997 Received: from press-gopher.uchicago.edu ([128.135.234.36]) by nic.funet.fi with ESMTP id <1674-24570>; Mon, 7 Apr 1997 22:51:08 +0300 Received: from localhost by press-gopher.uchicago.edu with SMTP id <26519-16291>; Mon, 7 Apr 1997 14:50:52 -0500 Date: Mon, 7 Apr 1997 14:50:51 -0500 (CDT) Sender: Roy C Bixler From: Roy C Bixler To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: weekend datadump at mea.tmt.tele.fi In-Reply-To: <19970407193147Z1797-13604+68@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 7 Apr 1997, Matti Aarnio wrote: > At Solaris I see the smtp-clients occasionally > freezing, and not reacting on SIGALRM as if it > were set a block on the signal. > (That is: 'truss -p nn' does not see the SIGALRM > signal to the process :-? ) > > Any comments (about the package) ? Interesting ... I never upgraded to 2.99.48 on my Solaris systems because I never saw the stuck 'smtp' processes as on Linux. I am still happily running 2.99.47 here. Also, since running 2.99.48p3 on Linux, I have not seen the stuck 'smtp' processes. Just a datapoint ... but it would seem to be saying that something that's recently changed is causing the problem. Roy rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Mon Apr 7 23:35:54 1997 Received: by nic.funet.fi id <1257-24570>; Mon, 7 Apr 1997 23:34:52 +0300 Subject: Re: weekend datadump at mea.tmt.tele.fi From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Mon, 7 Apr 1997 23:34:50 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at Apr 7, 97 02:50:51 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1530 Message-Id: <19970407203452Z1257-24570+82@nic.funet.fi> Return-Path: [talking of smtp becoming stuck, SIGALRM ineffectivity] > Interesting ... I never upgraded to 2.99.48 on my Solaris systems because > I never saw the stuck 'smtp' processes as on Linux. I am still happily > running 2.99.47 here. Also, since running 2.99.48p3 on Linux, I have not > seen the stuck 'smtp' processes. Just a datapoint ... but it would seem > to be saying that something that's recently changed is causing the > problem. The Solaris is weird. Same code works fine on Linux, SGI, and FreeBSD, but becomes stuck on Solaris... Recently our main mail-relay's smtp-server caused an alert on the periodic service quality monitor by presenting following at the log (and doing an exit() after it..): 00000# started server pid 460 at Thu, 3 Apr 1997 17:51:59 +0300 000000# accept(): No child processes 00000# started server pid 27408 at Sat, 5 Apr 1997 18:57:16 +0300 The smtpserver did until then consider such a major catastrophe, but then I decided that as the accept() at Solaris can get weird errors not listed at its documentation, who am I to argue ? Sigh... Perhaps the freeze-out of the smtp-client is similar Solaris weirdo... Hmm.. or.. It is possible that the Solaris uses POSIX sigaction(), and does REQUIRE usage of flag SA_NODEFER. The ALRM handler does after all do a longjump, and will not return from the signal handler per se... In such a case the first timeout works, and the second one jams.. > Roy > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 8 03:56:41 1997 Received: from pccross.average.org ([194.67.9.6]) by nic.funet.fi with ESMTP id <1708-14774>; Tue, 8 Apr 1997 03:55:15 +0300 Received: from localhost(really [127.0.0.1]) by pccross.average.org via smtpd with esmtp id for ; Tue, 8 Apr 1997 05:15:12 +0400 (MSD) (Smail-3.2 1996-Jul-4 #12 built 1996-Jul-20) Date: Tue, 8 Apr 1997 05:15:12 +0400 (MSD) From: Eugene Crosser Sender: crosser@pccross.average.org Reply-To: Eugene Crosser Subject: Re: weekend datadump at mea.tmt.tele.fi To: Matti Aarnio cc: Roy C Bixler , zmailer@nic.funet.fi In-Reply-To: <19970407203452Z1257-24570+82@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; LANGUAGE=ru Return-Path: > The Solaris is weird. Same code works fine on Linux, > SGI, and FreeBSD, but becomes stuck on Solaris... > > Recently our main mail-relay's smtp-server caused an alert > on the periodic service quality monitor by presenting following > at the log (and doing an exit() after it..): > > 00000# started server pid 460 at Thu, 3 Apr 1997 17:51:59 +0300 > 000000# accept(): No child processes > 00000# started server pid 27408 at Sat, 5 Apr 1997 18:57:16 +0300 This is probably a Solaris bug. We recently got similar error - "No child processes" after select() with RADIUS daemon. Solaris 2.5.1. Also, these days I am getting mad smtp agents eating 10 - 30 - 50 megs of virtual memory. And their process numbers are not shown in mailq output. This is with 2.99.47, I think that I need to install .48 on the main relay since you told that you've fixed some bugs in smtp since then... Eugene From zmailer-owner@nic.funet.fi Tue Apr 8 10:35:38 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <2130-29641>; Tue, 8 Apr 1997 10:34:43 +0300 Received: by redhotmomma.ssr.com id <36096-26889>; Tue, 8 Apr 1997 03:52:12 -0400 Sender: Scott Ballantyne From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: Unfortunate behavior for 2.99.47 Message-Id: <97Apr8.035212-0400_edt.36096-26889+2@redhotmomma.ssr.com> Date: Tue, 8 Apr 1997 03:52:11 -0400 Return-Path: I tried upgrading zmailer today, but perhaps I did not get the most recent version? In any case, the router in 2.88.47 eats messages. The log message is <19970408073828Z36096-26599+1@redhotmomma.ssr.com>: recipient: local sdb@ssr.com sdb router: rename(36096-26599,../deferred/36096.ctrlfile): No such file or directory There is certainly a deferred directory --- the messages are simply lost. This portion of code needs protecting, somehow... Has this behavior been fixed? Thanks, sdb -- sdb@ssr.com From zmailer-owner@nic.funet.fi Tue Apr 8 10:55:24 1997 Received: by nic.funet.fi id <1277-21911>; Tue, 8 Apr 1997 10:54:53 +0300 Subject: Re: Unfortunate behavior for 2.99.47 From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Tue, 8 Apr 1997 10:54:49 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr8.035212-0400_edt.36096-26889+2@redhotmomma.ssr.com> from "Scott Ballantyne" at Apr 8, 97 03:52:11 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 704 Message-Id: <19970408075453Z1277-21911+97@nic.funet.fi> Return-Path: > I tried upgrading zmailer today, but perhaps I did not get the most > recent version? In any case, the router in 2.88.47 eats messages. The > log message is > > <19970408073828Z36096-26599+1@redhotmomma.ssr.com>: recipient: local sdb@ssr.com sdb > router: rename(36096-26599,../deferred/36096.ctrlfile): No such file or directory > > There is certainly a deferred directory --- the messages are simply > lost. This portion of code needs protecting, somehow... The lib/esyslib.c: erename() routine had lost its ending "return 0" at some point. > Has this behavior been fixed? At my betas: ftp://mea.tmt.tele.fi/zmailer-2.99.48-*.tar.gz > Thanks, > sdb > -- > sdb@ssr.com /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 8 11:11:54 1997 Received: by nic.funet.fi id <2874-14774>; Tue, 8 Apr 1997 11:10:47 +0300 Subject: Re: weekend datadump at mea.tmt.tele.fi From: Matti Aarnio To: crosser@online.ru Date: Tue, 8 Apr 1997 11:10:42 +0300 (EET DST) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 8, 97 05:15:12 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 762 Message-Id: <19970408081047Z2874-14774+121@nic.funet.fi> Return-Path: ... > Also, these days I am getting mad smtp agents eating 10 - 30 - 50 megs > of virtual memory. And their process numbers are not shown in mailq > output. This is with 2.99.47, I think that I need to install .48 on > the main relay since you told that you've fixed some bugs in smtp > since then... Not shown ? It must mean you have restarted the scheduler ? The smtp agent did (in 47) occasionally longjmp() back to a stack-frame that had been exited already -- which yields unpredictable results... Even the .48 betas do not (yet) protect all dynamic structure buildups against the SIGALRM doing a longjmp from midst of their code -- though most likely the alarm will occur during write() or read() of a socket... > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 8 12:15:39 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <2569-21911>; Tue, 8 Apr 1997 12:14:20 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109568-227>; Tue, 8 Apr 1997 11:14:09 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <7828-204>; Tue, 8 Apr 1997 11:13:48 +0200 Date: Tue, 8 Apr 1997 11:13:42 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: weekend datadump at mea.tmt.tele.fi In-Reply-To: <19970407203452Z1257-24570+82@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: -----BEGIN PGP SIGNED MESSAGE----- On Mon, 7 Apr 1997, Matti Aarnio wrote: > [talking of smtp becoming stuck, SIGALRM ineffectivity] > > It is possible that the Solaris uses POSIX sigaction(), > and does REQUIRE usage of flag SA_NODEFER. The ALRM > handler does after all do a longjump, and will not return > from the signal handler per se... > Hi, not that i have looked at the code, but ... do you define _POSIX_SOURCE ? Also the man page says: On entry to the signal handler, that set of signals is added to the set of signals already being blocked when the signal is delivered. In addition, the signal that caused the handler to be executed will also be blocked, unless the SA_NODEFER flag has been specified. Greets, Sven - -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: latin1 iQCVAwUBM0oMSUgFYR9zZnaVAQFFggP+NLXeD/2gatKex/hN08XIcC218CFrELDk M5saPxHbnioZ8hLYYK0K8FV/8YJbsLrGyfBFVfCm0ejShSH/kQPJT37ny68T0tn9 gPPw7cpFg895eX6CnJEjP5UTooFyH6TgWhvUUdlZ6vnvN/lCQTlbJaIOLbSKn7Ma njso4YxDa2I= =mzih -----END PGP SIGNATURE----- From zmailer-owner@nic.funet.fi Tue Apr 8 15:09:50 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2510-14774>; Tue, 8 Apr 1997 15:08:40 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-23466>; Tue, 8 Apr 1997 16:08:32 +0400 Date: Tue, 8 Apr 1997 16:08:23 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: weekend datadump at mea.tmt.tele.fi To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970408081047Z2874-14774+121@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN Return-Path: > > Also, these days I am getting mad smtp agents eating 10 - 30 - 50 megs > > of virtual memory. And their process numbers are not shown in mailq > > output. This is with 2.99.47, I think that I need to install .48 on > > the main relay since you told that you've fixed some bugs in smtp > > since then... > > Not shown ? It must mean you have restarted the scheduler ? Very well might be. I restarted the scheduler to put a "ulimit -v 30000" ;-) I will try .48 and see what happens. Eugene From zmailer-owner@nic.funet.fi Tue Apr 8 15:57:57 1997 Received: from american.com ([204.253.96.2]) by nic.funet.fi with ESMTP id <2177-21911>; Tue, 8 Apr 1997 15:56:40 +0300 Received: from hotsprings.american.com (hotsprings.american.com [204.253.96.21]) by american.com (8.7.6/8.6.12) with ESMTP id IAA21039 for ; Tue, 8 Apr 1997 08:56:23 -0400 (EDT) Received: from localhost (pgf@localhost) by hotsprings.american.com (8.6.12/8.6.9) with SMTP id IAA12243 for ; Tue, 8 Apr 1997 08:56:22 -0400 Message-Id: <199704081256.IAA12243@hotsprings.american.com> X-Authentication-Warning: hotsprings.american.com: Host localhost didn't use HELO protocol To: zmailer@nic.funet.fi Subject: Re: weekend datadump at mea.tmt.tele.fi In-reply-to: goldt's message of Tue, 08 Apr 1997 11:13:42 +0200. Reply-to: pgf@american.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <12241.860504182.1@hotsprings> Date: Tue, 08 Apr 1997 08:56:22 -0400 From: Paul Fox Return-Path: > > > [talking of smtp becoming stuck, SIGALRM ineffectivity] > > > > It is possible that the Solaris uses POSIX sigaction(), > > and does REQUIRE usage of flag SA_NODEFER. The ALRM > > handler does after all do a longjump, and will not return > > from the signal handler per se... remember that BSD longjmp has traditionally unblocked signals, for precisely this reason. the Solaris version, true to heritage, does not. you must use sigsetjmp/siglongjmp to get the BSD behavior. paul --------------------- paul fox american internet corporation pgf@american.com (home: pgf@foxharp.boston.ma.us) From zmailer-owner@nic.funet.fi Tue Apr 8 19:39:56 1997 Received: by nic.funet.fi id <2461-26897>; Tue, 8 Apr 1997 19:38:26 +0300 Subject: Re: weekend datadump at mea.tmt.tele.fi From: Matti Aarnio To: pgf@american.com Date: Tue, 8 Apr 1997 19:38:22 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199704081256.IAA12243@hotsprings.american.com> from "Paul Fox" at Apr 8, 97 08:56:22 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1058 Message-Id: <19970408163826Z2461-26897+224@nic.funet.fi> Return-Path: > > > [talking of smtp becoming stuck, SIGALRM ineffectivity] > > > > > > It is possible that the Solaris uses POSIX sigaction(), > > > and does REQUIRE usage of flag SA_NODEFER. The ALRM > > > handler does after all do a longjump, and will not return > > > from the signal handler per se... > > remember that BSD longjmp has traditionally unblocked signals, for > precisely this reason. the Solaris version, true to heritage, does not. > you must use sigsetjmp/siglongjmp to get the BSD behavior. I have now using two approaches: using SA_NODEFER, and releasing the sigmask explicitely, though I presume the release in systems that know of SA_NODEFER is unnecessary. I haven't yet checked on what purpose serves the explicite definition of _POSIX_SOURCE in Solaris systems. I feel it should be something that the compiler defines, like: gcc -posix > paul > --------------------- > paul fox american internet corporation > pgf@american.com (home: pgf@foxharp.boston.ma.us) /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 8 19:56:01 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2724-29641>; Tue, 8 Apr 1997 19:54:51 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-22147>; Tue, 8 Apr 1997 20:54:38 +0400 Date: Tue, 8 Apr 1997 20:54:35 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: weekend datadump at mea.tmt.tele.fi To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970407193147Z1797-13604+68@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Just a note: in .48-970407 something changed in handling addresses like @somehwere:somebody@elswhere, but this still does not work. It results in [non]delivery reports with empty reason, and the message is not delivered. A sidenote: as the smtpserver code was split, I had to redo all my on-the-fly translation patches by hand :-( Well, I understand that there are not many users of this code and you probably won't want to include it into mainstram... Ehh... Eugene From zmailer-owner@nic.funet.fi Tue Apr 8 20:31:52 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2151-14774>; Tue, 8 Apr 1997 20:29:59 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-23357>; Tue, 8 Apr 1997 21:29:50 +0400 Date: Tue, 8 Apr 1997 21:29:34 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: bug in .48-970407 reports To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Just a few minutes ago I installed .48-970407 and had to rollback immediately because of the following trouble: non-delivery reports look like this: --A41E5.860519815=_/deimos Content-Type: text/plain This is a collection of reports about email delivery process concerning a message you originated: : --A41E5.860519815=_/deimos Content-Type: message/delivery-status Reporting-MTA: dns; deimos Arrival-Date: Mon, 7 Apr 1997 14:34:57 +0400 Original-Recipient: rfc822;Yas@kvazar.kmicro.com Final-Recipient: RFC822;Yas@kvazar.kmicro.com Action: Status: Diagnostic-Code: Last-Attempt-Date: Tue, 8 Apr 1997 21:16:43 +0400 --A41E5.860519815=_/deimos Content-Type: message/rfc822 m Received: from marceau.fm.intel.com ([132.233.247.8]) by deimos.sovam.com with E SMTP id <228153-268>; Mon, 7 Apr 1997 14:35:01 +0400 Received: from fmmail.fm.intel.com by marceau.fm.intel.com (8.8.4/10.0i); Mon, 7 Apr 1997 10:34:43 GMT [skipped] First, there is no error description in the report. Second, there is a line with a single character 'm' in the attached original message (and it is thus an invalid rfc822 message). Eugene From zmailer-owner@nic.funet.fi Tue Apr 8 21:31:27 1997 Received: by nic.funet.fi id <2298-21911>; Tue, 8 Apr 1997 21:29:58 +0300 Subject: Re: bug in .48-970407 reports From: Matti Aarnio To: crosser@online.ru Date: Tue, 8 Apr 1997 21:29:49 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 8, 97 09:29:34 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1584 Message-Id: <19970408182958Z2298-21911+247@nic.funet.fi> Return-Path: > Just a few minutes ago I installed .48-970407 and had to rollback > immediately because of the following trouble: non-delivery reports > look like this: Hmm... Yes, transports/libta/ctlopen.c has one instance of setting the headeroffset -- that is "two under". But why does the status info be blank ? What the real error was ? There might (after all) be a loophole where the information for the messages were not set for the notary. /Matti Aarnio > --A41E5.860519815=_/deimos > Content-Type: text/plain > > This is a collection of reports about email delivery > process concerning a message you originated: > > : > > --A41E5.860519815=_/deimos > Content-Type: message/delivery-status > > Reporting-MTA: dns; deimos > Arrival-Date: Mon, 7 Apr 1997 14:34:57 +0400 > > Original-Recipient: rfc822;Yas@kvazar.kmicro.com > Final-Recipient: RFC822;Yas@kvazar.kmicro.com > Action: > Status: > Diagnostic-Code: > Last-Attempt-Date: Tue, 8 Apr 1997 21:16:43 +0400 > > --A41E5.860519815=_/deimos > Content-Type: message/rfc822 > > m > Received: from marceau.fm.intel.com ([132.233.247.8]) by deimos.sovam.com with > E > SMTP id <228153-268>; Mon, 7 Apr 1997 14:35:01 +0400 > Received: from fmmail.fm.intel.com by marceau.fm.intel.com (8.8.4/10.0i); Mon, > 7 > Apr 1997 10:34:43 GMT > > [skipped] > > First, there is no error description in the report. Second, there is a > line with a single character 'm' in the attached original message (and > it is thus an invalid rfc822 message). > > Eugene > From zmailer-owner@nic.funet.fi Tue Apr 8 21:38:42 1997 Received: by nic.funet.fi id <2048-14774>; Tue, 8 Apr 1997 21:37:26 +0300 Subject: Re: weekend datadump at mea.tmt.tele.fi From: Matti Aarnio To: crosser@online.ru Date: Tue, 8 Apr 1997 21:37:11 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 8, 97 08:54:35 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 661 Message-Id: <19970408183726Z2048-14774+298@nic.funet.fi> Return-Path: > Just a note: in .48-970407 something changed in handling addresses like > @somehwere:somebody@elswhere, but this still does not work. It results in > [non]delivery reports with empty reason, and the message is not delivered. Eh, no. I don't think I changed anything in that regard. I must look at it. > A sidenote: as the smtpserver code was split, I had to redo all my > on-the-fly translation patches by hand :-( Well, I understand that > there are not many users of this code and you probably won't want > to include it into mainstram... Ehh... I might well integrate it, I just have waited you to submit it to me :-) > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 9 00:24:06 1997 Received: by nic.funet.fi id <1191-22902>; Wed, 9 Apr 1997 00:22:58 +0300 Subject: Re: hung smtp processes again From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Wed, 9 Apr 1997 00:22:54 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at Apr 8, 97 01:33:49 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1787 Message-Id: <19970408212258Z1191-22902+6@nic.funet.fi> Return-Path: > On Tue, 8 Apr 1997, Matti Aarnio wrote: > > Yes, that one has gone over the deep end. It is also from the > > time before the great setjmp()/longjmp() rewamp. > > > > I betcha you are not using my 2.99.48-970407.tar.gz, > > as it does not have 'p3' for it... > > > > Yes, the version numbers at my workstation are -- challenging.. > > OK, I just compiled up your latest 2.99.48-970407 version and I got the > same thing again. Oooo... > Also, on Linux, I notice something disturbing. Remembering the security > alert from unc.edu about how it was easy to get 'root' with the right > .forward file, I set up a .forward file in my directory and mailed myself > a message with the verbose option on. While I did not try to actually get > 'root' with my .forward file (why would I? I know the passwd ... :-) ), > the transport agent said it executed the pipe with uid=0 and gid=0. Is > there a fix for this or have I set something up incorrectly? Three bugs. 'mailbox' prints out data from uninitialized variables. When initialized, they were zero :-( As to why -- the third component on the recipient address contained spaces, but without proper quotes: |cat > /dev/null when it should have been: "|cat > /dev/null" First fix was to modify the first instance of: \|.+ at the rrouter.cf to: |.+ so that it matches the pipes... But I went ahead and 'overdid' the fix by armoring the router to always to produce a proper quoted string, but also do some tests at the mailbox to see that the input values make sense.. ..... Ok, I made a diff (patch, 7.4kB) containing my current source state. Available now. Back to figuring out why you see the process stuck. Datapoints are welcomed. > Roy > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 9 10:52:27 1997 Received: by nic.funet.fi id <2633-8235>; Wed, 9 Apr 1997 10:51:15 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: Addresses of @a:b@c are now acceptable in all forms Message-Id: <19970409075115Z2633-8235+48@nic.funet.fi> Date: Wed, 9 Apr 1997 10:51:14 +0300 Return-Path: The problem was that the 'localnames' table was not used for the analysis of those address components, rather just a few elements out of /etc/mail.conf The change for rrouter.cf is rather small, and is provided in the zmailer-2.99.48-970407.patch1b at my workstation ( ftp://mea.tmt.tele.fi/ ) /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 9 11:22:38 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2159-8235>; Wed, 9 Apr 1997 11:22:15 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68045-23357>; Wed, 9 Apr 1997 12:22:00 +0400 Date: Wed, 9 Apr 1997 12:21:55 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: bug in .48-970407 reports To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970408182958Z2298-21911+247@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > Just a few minutes ago I installed .48-970407 and had to rollback > > immediately because of the following trouble: non-delivery reports > > look like this: > But why does the status info be blank ? > What the real error was ? There might (after all) > be a loophole where the information for the messages > were not set for the notary. This error was "No MX for the domain, are we the best?" (something of that meaning). The same occurs when when you try <@a:b@c> type of address. Eugene From zmailer-owner@nic.funet.fi Wed Apr 9 12:05:25 1997 Received: from tuminfo2.informatik.tu-muenchen.de ([131.159.0.81]) by nic.funet.fi with ESMTP id <2221-3577>; Wed, 9 Apr 1997 12:04:43 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109574-227>; Wed, 9 Apr 1997 11:04:15 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <7828-204>; Wed, 9 Apr 1997 11:03:47 +0200 Date: Wed, 9 Apr 1997 11:03:41 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt To: Matti Aarnio cc: pgf@american.com, zmailer@nic.funet.fi Subject: Re: weekend datadump at mea.tmt.tele.fi In-Reply-To: <19970408163826Z2461-26897+224@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: -----BEGIN PGP SIGNED MESSAGE----- On Tue, 8 Apr 1997, Matti Aarnio wrote: > I have now using two approaches: using SA_NODEFER, and > releasing the sigmask explicitely, though I presume the > release in systems that know of SA_NODEFER is unnecessary. > > I haven't yet checked on what purpose serves the explicite > definition of _POSIX_SOURCE in Solaris systems. I feel > it should be something that the compiler defines, like: > gcc -posix > Hi Matti, i just thought you would program in POSIX standard where sigaction is prefered and where a signal handler doesn't has to be reset. If you use POSIX (which nearly all UNIX systems support and even OS/2) then you have to tell it your system by defining _POSIX_SOURCE which often is used in the system header files for #ifdef switches. This really reduces the problems to make zmailer run on different UNIX flavors. Greets, Sven - -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: latin1 iQCVAwUBM0tbcEgFYR9zZnaVAQH5AwP9E7W4lIBMwILSg6noZ/KzHS9fCxBgOqHe yKqeXXNeJC3JR+gikl0CBE9n9LghCB19An8490YD/1S2c8gxr2dtZ2/RTGJaup2S p4AyBWzQNRZ+kVKgDS6SgLcoM11g7fP01HTui5BmRhmCxuwFAsiXFQKdPI48jACr iRQvxPIpx/Q= =OibA -----END PGP SIGNATURE----- From zmailer-owner@nic.funet.fi Wed Apr 9 14:52:42 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <3028-3577>; Wed, 9 Apr 1997 14:51:24 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68048-23601>; Wed, 9 Apr 1997 15:51:02 +0400 Date: Wed, 9 Apr 1997 15:50:57 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: .48-970407.patch1b is OK To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: .48-970407.patch1b seems to work correctly so far. Non-delivery reports are good. @a:b@c addresses resolved correctly. Even long broken identd code works now. I'll see if the problems with crasy smtp processes are over or not. Solaris 2.5.1 Eugene From zmailer-owner@nic.funet.fi Wed Apr 9 18:40:49 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2632-8235>; Wed, 9 Apr 1997 18:36:50 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68048-23598>; Wed, 9 Apr 1997 19:36:36 +0400 Date: Wed, 9 Apr 1997 19:36:25 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: .48-970407.patch1b is OK To: zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > .48-970407.patch1b seems to work correctly so far. Non-delivery > reports are good. @a:b@c addresses resolved correctly. > Even long broken identd code works now. > I'll see if the problems with crasy smtp processes are over or not. Although, I noticed a big (18Meg) smtp process now and started investigation. It revealed that the process is trying to deliver a hughe message (17M). Matti, does it mean that smtp agent keeps the whole message in memory? Is it expected behavior? (No, in my previous reports, smtp processes grew to 50Meg while delivering average several Kb size messages). Eugene From zmailer-owner@nic.funet.fi Wed Apr 9 19:39:59 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2368-31831>; Wed, 9 Apr 1997 19:38:26 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68048-23598>; Wed, 9 Apr 1997 20:38:12 +0400 Date: Wed, 9 Apr 1997 20:38:06 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: domain name case sensitivity?! To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Hmm, looks like a bug? === This is a collection of reports about email delivery process concerning a message you originated: : smtp; 500 (DNS: no such domain: 0nline.ru) === while there is domain "online.ru". According to rfc822, domain matching should be case insensitive? Eugene From zmailer-owner@nic.funet.fi Wed Apr 9 20:07:09 1997 Received: from wit393101.student.utwente.nl ([130.89.235.11]) by nic.funet.fi with ESMTP id <2135-3577>; Wed, 9 Apr 1997 20:05:54 +0300 Received: from localhost ([127.0.0.1]) by wit393101.student.utwente.nl with SMTP id <23845-20308>; Wed, 9 Apr 1997 19:06:01 +0200 Date: Wed, 9 Apr 1997 19:05:49 +0200 (MET DST) From: Florian Overkamp Reply-To: sim@wit393101.student.utwente.nl To: Eugene Crosser cc: zmailer@nic.funet.fi Subject: Re: domain name case sensitivity?! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 9 Apr 1997, Eugene Crosser wrote: > Hmm, looks like a bug? > > === > This is a collection of reports about email delivery > process concerning a message you originated: > > : smtp; 500 (DNS: no such domain: > 0nline.ru) > === > > while there is domain "online.ru". According to rfc822, domain matching > should be case insensitive? Looks like a zero ? 0nline != Online == online /* Florian Overkamp */ /* PGP Key-ID 233FC67D */ /* PGP Fingerprint 00 3F F5 63 DD FB 5D B7 0A 16 0A EC F9 BD 01 97 */ From zmailer-owner@nic.funet.fi Wed Apr 9 20:12:09 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <2058-31831>; Wed, 9 Apr 1997 20:10:25 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68049-23601>; Wed, 9 Apr 1997 21:10:16 +0400 Date: Wed, 9 Apr 1997 21:10:11 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: domain name case sensitivity?! To: sim@wit393101.student.utwente.nl cc: zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > : smtp; 500 (DNS: no such domain: > > 0nline.ru) > Looks like a zero ? > > 0nline != Online == online Oh, probaly it is! Sorry. Eugene From zmailer-owner@nic.funet.fi Wed Apr 9 20:13:51 1997 Received: from rainier.tpdinc.com ([206.136.148.3]) by nic.funet.fi with ESMTP id <2501-8235>; Wed, 9 Apr 1997 20:12:16 +0300 Received: from orcas.tpdinc.com by rainier.tpdinc.com (8.7.5/TPD-Sendmail-2.0-072396) id KAA09704; Wed, 9 Apr 1997 10:12:17 -0700 (PDT) Date: Wed, 9 Apr 1997 10:14:38 -0700 (Pacific Daylight Time) From: Nick Silberstein To: Eugene Crosser cc: zmailer@nic.funet.fi Subject: Re: domain name case sensitivity?! In-Reply-To: Message-ID: X-X-Sender: nick@rainier.tpdinc.com MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Eugene, On Wed, 9 Apr 1997, Eugene Crosser wrote: > Hmm, looks like a bug? > > === > This is a collection of reports about email delivery > process concerning a message you originated: > > : smtp; 500 (DNS: no such domain: > 0nline.ru) > === > > while there is domain "online.ru". According to rfc822, domain matching > should be case insensitive? Correct; however, that seems to read 0nline.ru (zero) rather than Online.ru (capital 'o'). Nick From zmailer-owner@nic.funet.fi Wed Apr 9 21:10:29 1997 Received: from alpha.xerox.com ([13.1.64.93]) by nic.funet.fi with SMTP id <2849-3577>; Wed, 9 Apr 1997 21:09:32 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <23713(6)>; Wed, 9 Apr 1997 11:08:17 PDT X-NS-Transport-ID: 0800200AF21F36ED9F44 Date: Wed, 9 Apr 1997 11:07:38 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: domain name case sensitivity?! In-Reply-to: "crosser@online:ru:Xerox's message of Wed, 9 Apr 1997 09:38:06 PDT" To: crosser@online.ru cc: Nicholas_Briggs.PARC@xerox.com, zmailer@nic.funet.fi Message-Id: <97Apr9.110817pdt."23713(6)"@alpha.xerox.com> Return-Path: It's not a case difference -- it's because it is a zero instead of an O in "0nline.ru" \nick From zmailer-owner@nic.funet.fi Thu Apr 10 00:19:09 1997 Received: by nic.funet.fi id <16260-20009>; Thu, 10 Apr 1997 00:17:53 +0300 Subject: Re: .48-970407.patch1b is OK From: Matti Aarnio To: crosser@online.ru Date: Thu, 10 Apr 1997 00:17:52 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 9, 97 07:36:25 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1253 Message-Id: <19970409211753Z16260-20009+4@nic.funet.fi> Return-Path: > > .48-970407.patch1b seems to work correctly so far. Non-delivery > > reports are good. @a:b@c addresses resolved correctly. > > Even long broken identd code works now. > > I'll see if the problems with crasy smtp processes are over or not. > > Although, I noticed a big (18Meg) smtp process now and started > investigation. It revealed that the process is trying to deliver > a hughe message (17M). Matti, does it mean that smtp agent keeps > the whole message in memory? Is it expected behavior? No, it PAGES the message in. If you try to send a message that is way too large (2-3 GB?) then average 32-bit system can't send it, but it should not be needed either.. The message file (and the transport-descriptor file) are opened RDONLY (and RW respectively), and then mmap()ed MAP_FILE|MAP_SHARED. It gives process often enlarged apparent size, which really is not that big in memory/swap. > (No, in my previous reports, smtp processes grew to 50Meg while > delivering average several Kb size messages). Propably an error in lib/selfaddr.c. I spotted it even in my home machine... "weird, the machine went to a swap frenzy, what is going on ? smtp growing rapidly ? Oooo..." Fixed in patch1b > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 10 01:54:51 1997 Received: from weaver-gw.netapp.com ([198.95.224.2]) by nic.funet.fi with ESMTP id <16260-20009>; Thu, 10 Apr 1997 01:52:10 +0300 Received: from netapp.com ([192.9.200.1]) by weaver.netapp.com with SMTP id <246374-11191>; Wed, 9 Apr 1997 15:50:51 -0700 Received: by netapp.com (4.1/SMI-4.1) id AA21493; Wed, 9 Apr 97 15:53:37 PDT From: sterling@netapp.com (Bruce Sterling Woodcock) Message-Id: <9704092253.AA21493@netapp.com> Subject: Problems with zmailer 2.99.25 To: zmailer@nic.funet.fi Date: Wed, 9 Apr 1997 15:53:37 -0700 (PDT) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 4376 Return-Path: I'm having a couple of crashes in my zmailer and was curious if they were known problems or not. I'm still running zmailer 2.99.25, which might be the problem, but I am reluctant to upgrade given that all the later versions seem to have stability problems of their own. For some time, I have been having crashes in both the router and the smtp transport agent. The smtp is not so bad, as the mail gets sent later; the router crash kills all processing of the messages and requires me to delete the offending message from the queue(s) and restart zmailer. Originally, I thought some of these problems were due to our old BSDI 1.1 system and a failing disk, but I continue to see problems after upgrading to 3.0. The router crash is fairly straightforward: weaver-gw# gdb /usr/local/lib/mail/bin/router router/router.core GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.16 (i386-unknown-bsdi3.0), Copyright 1996 Free Software Foundation, Inc... Core was generated by `router'. Program terminated with signal 11, Segmentation fault. #0 0xebf1 in prctladdr (info=0xd7608, fp=0x76580, cfflag=115, comment=0xbb60 "sender") at rfc822.c:1945 1945 if (*x->string == '\0') (gdb) where #0 0xebf1 in prctladdr (info=0xd7608, fp=0x76580, cfflag=115, comment=0xbb60 "sender") at rfc822.c:1945 #1 0xe731 in sequencer (e=0xd5e6c, file=0x9a9fc "15972-174") at rfc822.c:1789 #2 0xa72e in run_rfc822 (argc=2, argv=0x9aa08) at rfc822.c:123 #3 0x1a5ac in execute (c=0xefbfcd90, caller=0xefbfd708, oretcode=0, name=0x34d0 "rfc822") at execute.c:384 #4 0x1ce8e in runcommand (c=0xefbfcd90, pc=0xefbfd708, retcodep=0xefbfd704) at interpret.c:690 #5 0x1ec96 in interpret (code=0xae3ec "\r", eocode=0xae68e "\016*,", entry=0xae3fb "2#file", caller=0xefbfd708, retcodep=0xefbfd704, cdp=0xaa98c) at interpret.c:1643 #6 0x20adc in lapply (fname=0x34d7 "process", l=0xd5e3c) at interpret.c:2602 #7 0x20b72 in apply (argc=2, argv=0xefbfd998) at interpret.c:2626 #8 0x93ff in s_apply (argc=2, argv=0xefbfd998) at shliaise.c:53 #9 0x3fbc in rd_doit (filename=0xac4e0 "15972-178", dirs=0x844e8 "") at functions.c:637 #10 0x433c in rd_instability (dirp=0x980a0, dirs=0x844e8 "") at functions.c:768 #11 0x462d in run_daemon (argc=1, argv=0xefbfdee4) at functions.c:903 #12 0x1a27 in main (argc=3, argv=0xefbfdf64) at router.c:309 The headers on the offending message (in the above example) look like: external rcvdfrom gateway1.isite.net ([198.207.204.66]) with SMTP channel error to Received: from proxy1.noc.isite.net (proxy1.noc.isite.net [172.16.1.11]) by gateway1.isite.net (8.8.5/8.8) with ESMTP id JAA22477 for ; Tue, 8 Apr 1997 09:54:38 -0700 (PDT) Received: from smtp-gateway.asyst.com (smtp-gateway.asyst.com [172.26.12.12]) by proxy1.noc.isite.net (8.8.5/8.8) with SMTP id KAA10721 for ; Tue, 8 Apr 1997 10:56:22 -0700 (PDT) Message-Id: <199704081756.KAA10721@proxy1.noc.isite.net> Received: by smtp-gateway.asyst.com via MsMail with Mimetic; Tue, 8 Apr 1997 09:49 -0700 MIME-Version: 1.0 Date: Tue, 8 Apr 1997 09:49 -0700 From: Adminstrator <> To: lmay@netapp.com Subject: Mail failure Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable The smtp core is a little more obscure: weaver-gw# gdb /usr/local/lib/mail/bin/ta/smtp transport/smtp.core GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.16 (i386-unknown-bsdi3.0), Copyright 1996 Free Software Foundation, Inc... Core was generated by `smtp'. Program terminated with signal 11, Segmentation fault. #0 0x2f274 in fclose () (gdb) where #0 0x2f274 in fclose () #1 0x1b2b in main (argc=3, argv=0xefbfdf00) at smtp.c:523 Is it possible to get patches applied to 2.99.25 that fix these issues? Also, can anyone recommend a later version that is actually stable? Bruce -- Bruce Sterling Woodcock ------ Systems Administrator ][ sterling@netapp.com Network & Systems Administration - Network Appliance ][ sterling@netcom.com From zmailer-owner@nic.funet.fi Thu Apr 10 07:01:00 1997 Received: from roadrunner.realbig.com ([206.161.253.66]) by nic.funet.fi with ESMTP id <15872-5378>; Thu, 10 Apr 1997 06:55:10 +0300 Received: from ws.realbig.com ([206.161.253.65]) by roadrunner.realbig.com with ESMTP id <131087-25076>; Wed, 9 Apr 1997 22:53:52 -0500 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86756-15296>; Wed, 9 Apr 1997 22:53:44 -0500 Date: Wed, 9 Apr 1997 22:53:32 -0500 (EST) From: Andy Poling To: zmailer@nic.funet.fi Subject: ZMAILER smtp server and client at odds? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I'm running 2.99.38 on two Linux systems, with one (ws) configured to forward all outgoing SMTP mail to the other (roadrunner) since roadrunner is the powerhouse of the two. Roadrunner's disk filled up today and all of the mail from ws to roadrunner (all outgoing mail in this case) started bouncing with a very misleading error. Here's an example from a sendmail -v: ============================================================================== router done processing 86756-15296 scheduler processing 86756-15296 SMTP: Connecting to host: roadrunner.realbig.com 220 roadrunner.realbig.com ZMailer Server 2.99.38 #1 ESMTP+IDENT ready at Wed, 9 Apr 1997 22:23:59 -0500 EHLO ws.realbig.com 250-roadrunner.realbig.com Hello ws.realbig.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANGEDSTATUSCODES 250-EXPN 250-VRFY 250-XDSN 250-X-RCPTLIMIT 10000 250-X-TURNME 250-ETRN 250 HELP EHLO responce flags = 0x0f, rcptlimit=10000 MAIL From: BODY=8BITMIME SIZE=408 RCPT To: DATA 452 4.3.1 Try again later, insufficient storage available at the moment 452 4.3.1 insufficient storage space, try again later 550 5.1.7 No valid sender, rejecting all recipients Processed headers: ContentKind=0, CvtMode=0 ------ Received: from ws.realbig.com by ws.realbig.com id <86756-15296>; Wed, 9 Apr 1997 22:23:55 -0500 From: root To: andy@request.net Subject: test Message-Id: <97Apr9.222355-0500edt.86756-15296+30@ws.realbig.com> Date: Wed, 9 Apr 1997 22:23:54 -0500 . 550 5.5.2 Syntax error ->> 550 5.5.2 Syntax errorm andy@request.net 32765: error scheduler done processing 86756-15296 ============================================================================== Notice that the smtp client apparently didn't grok the 452 replies to the MAIL and RCPT commands. Instead it chose to pay attention to the last error it saw (550 - after the smtpserver was already thoroughly pissed off at it). I guess the pipelining was the reason, but it seems that the zmailer SMTP client and server should at least be able to co-exist. :-) The only thing in the bounce message transcripts was this last error and a complaint about a syntax error. Until I did this "sendmail -v" test I had no idea that the problem was a full disk. Is this a known/understood problem with the SMTP client? I realize that this version is a little bit out of date, but it's pretty solid and I've been waiting for Matti to decide to cut a really stable release before updating it. That's not a shot at you Matti - it just seems like the bugs have been keeping pace with the bugfixes lately, so I've decided to wait until good eventually (and inevitably) triumphs. :-) If you tell me that this is fixed in the current release, I might consider updating though... -Andy PS - I'm sure you've fixed the mis-spelling of ENHANCEDSTATUSCODES by now. Global Auctions From zmailer-owner@nic.funet.fi Thu Apr 10 15:09:38 1997 Received: by nic.funet.fi id <15635-24407>; Thu, 10 Apr 1997 15:08:30 +0300 Subject: SPAM and eBGP4 real-time map ? From: Matti Aarnio To: zmailer@nic.funet.fi Date: Thu, 10 Apr 1997 15:08:21 +0300 (EET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 5349 Message-Id: <19970410120830Z15635-24407+162@nic.funet.fi> Return-Path: I commented on something that Paul said on bind-workers, and as usual with elm, group-reply goes also to the sender. I am curious as to what Paul means with "eBGP4" feed, I must see his pages... /Matti Aarnio Forwarded message: Date: Wed, 9 Apr 1997 15:08:52 -0700 Message-Id: <199704092208.PAA19812@gw.home.vix.com> To: mea@nic.funet.fi From: Paul Vixie Subject: Paul Vixie did not spam you (this is an automated response) X-Agent: BSD vacation daemon Precedence: bulk Some weeks ago, I started receiving a massive number of e-mail bounces and complaints about spam. I immediately realized that someone had abused the network in my name; sure enough, I shortly received the evidence shown below. I apologize for this form letter response, but I'm expecting another 10,000 complaints and I do not plan to send personalized replies to each one. Your message has been filed and will be read by me shortly. I'm thankful, in a way, to see this occur. It underscores the terror that the spammers must be feeling based on the excellent effect we've produced with our web page at http://www.vix.com/spam/, and the wonderful feedback and sign up rate I've received on my "real time spammer black hole" eBGP4 feed. Yesterday they forged threats in my name to every member of the U.S.Senate and today I see even more evidence that I'm hurting spammers everywhere. I'm sad if you got spammed, but I assure you that I don't ever send unsolicited e-mail other than person-to-person for a specific purpose. You'll be spammed another dozen times this week, probably by Sanford Wallace of Cyberpromotions or some other fun-loving jerk who wants to steal your time and money. Please visit the http://www.vix.com/spam/ web page and consider signing our petition. If you are an ISP, please send me private e-mail asking about my real time blackhole feed, that makes known spammer networks disappear from your address space. PS. Here's the first human report I received about Friday's forged spam. The second report (further below) shows a second spam, apparently from someone named "Alan K Baker" who claims responsibility for both spams. | Date: Sat, 22 Feb 1997 00:58:26 +0000 (GMT) | From: xxx | To: scott@zorch.sf-bay.org | cc: paul@vix.com | Subject: Someone's faking your address in spam re: FREE GOLD!!!! | In-Reply-To: | | Someone has been sending junk mail with your names on it (at least, having | looked at your web pages I guess it's not from you :-). Here's the | message, including full headers. I've complained to glasscity.net. I | suggest you use what ever legal measures you can to make them divulge the | details of the user who forged this mail. Keep fighting the good fight. | | Mail follows below sig. | | -- | xxx | | Return-path: | Envelope-to: pw201@hermes.cam.ac.uk | Delivery-date: Sat, 22 Feb 1997 00:12:30 +0000 | Received: from mauve.csi.cam.ac.uk [131.111.8.38] (exim) | by black.csi.cam.ac.uk with smtp (Exim 1.58 #1) | id 0vy54n-0001c9-00; Sat, 22 Feb 1997 00:12:29 +0000 | Received: from home.glasscity.net [208.13.0.1] | by mauve.csi.cam.ac.uk with smtp (Exim 1.58 #1) | id 0vy54m-0002UV-00; Sat, 22 Feb 1997 00:12:28 +0000 | Received: from home.glasscity.net (194.72.198.24) by home.glasscity.net | (EMWAC SMTPRS 0.81) with SMTP id ; | Fri, 21 Feb 1997 19:12:25 -0500 | Date: Fri, 21 Feb 1997 19:12:25 -0500 | Message-ID: | Subject: FREE GOLD!!!! | X-UIDL: 20934820873692863905867902837082 | Priority: normal | X-Flags: PXU2-4R65-2264:R | Apparently-To: | From: scott@zorch.sf-bay.org | Bcc: | | Free gold! Phone, fax or email! | | Guaranteed FREE gold! 99.9% pure for first 1,000 replies! | | Phone +1 415 747 0204 | Fax +1 415 747 0106 | | email: | scott@zorch.sf-bay.org | or | buglady@ability.net | | Promotion by: | | Vixie Enterprises | Star Route Box 159A | Woodside, CA 94062 | | Be QUICK!!! | | Paul Vixie | paul@VIX.COM | +1 415 747 0204 ...and #2: | From: xxx | To: | Subject: Fw: Dear spam hater. | Date: Fri, 21 Feb 1997 19:30:42 -0700 | | I wonder if you can tell me how you show up as the source/return path of | this message below? I checked some of your news group postings and you | hardly seem like the type to mail this kind of nonsense. Any advise? | | Regards, | xxx | | ---------- | > From: | > To: | > Subject: Dear spam hater. | > Date: Friday, February 21, 1997 6:52 PM | > | > HA HA HA HA | > | > You are on a published list of peolple who hate commercial mail! | > | > How STUPID can you get! Commercial email is just taking off - and I'm | going to distribute this list everywhere | > I can to shut you whining people with nothing better to do than complain | up! | > | > It's BASTARDS like you who got me shut down last time - but demon won't | touch me! HA HA HA! | > | > I can go thru DIFFERENT servers! ANYTIME........... | > | > Alan the Infowarrior | > 'Let me read your palm!' | > -- | > Alan K Baker (MailWurliTzer Willy) | > www.senora.demon.co.uk | > alan@senora.demon.co.uk | > | > PS glasscity.net are not responsible for this......... | > I will continue working to eradicate spam from the Internet no matter what these "people" (term applies loosely) do. From zmailer-owner@nic.funet.fi Thu Apr 10 15:12:52 1997 Received: by nic.funet.fi id <15879-8497>; Thu, 10 Apr 1997 15:12:15 +0300 Subject: Re: Problems with zmailer 2.99.25 From: Matti Aarnio To: sterling@netapp.com (Bruce Sterling Woodcock) Date: Thu, 10 Apr 1997 15:12:05 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <9704092253.AA21493@netapp.com> from "Bruce Sterling Woodcock" at Apr 9, 97 03:53:37 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1950 Message-Id: <19970410121215Z15879-8497+192@nic.funet.fi> Return-Path: > I'm having a couple of crashes in my zmailer and was curious if they > were known problems or not. > > I'm still running zmailer 2.99.25, which might be the problem, but I am > reluctant to upgrade given that all the later versions seem to have > stability problems of their own. For some time, I have been having > crashes in both the router and the smtp transport agent. The smtp > is not so bad, as the mail gets sent later; the router crash kills > all processing of the messages and requires me to delete the offending > message from the queue(s) and restart zmailer. > > Originally, I thought some of these problems were due to our old BSDI 1.1 > system and a failing disk, but I continue to see problems after upgrading > to 3.0. It is a problem on "From: <>" -header, and things relating to it. Fixed 'ages' ago. ... (about smtp) > weaver-gw# gdb /usr/local/lib/mail/bin/ta/smtp transport/smtp.core > GDB is free software and you are welcome to distribute copies of it > under certain conditions; type "show copying" to see the conditions. > There is absolutely no warranty for GDB; type "show warranty" for details. > GDB 4.16 (i386-unknown-bsdi3.0), > Copyright 1996 Free Software Foundation, Inc... > Core was generated by `smtp'. > Program terminated with signal 11, Segmentation fault. > #0 0x2f274 in fclose () > (gdb) where > #0 0x2f274 in fclose () > #1 0x1b2b in main (argc=3, argv=0xefbfdf00) at smtp.c:523 > > Is it possible to get patches applied to 2.99.25 that fix these issues? > Also, can anyone recommend a later version that is actually stable? Apparently my ..48 betas are increasingly more stable, and since ..25, the smtp transporter has changes so much, I can't really give simple answers. > Bruce > > -- > Bruce Sterling Woodcock ------ Systems Administrator ][ sterling@netapp.com > Network & Systems Administration - Network Appliance ][ sterling@netcom.com /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 10 17:03:33 1997 Received: by nic.funet.fi id <15824-5378>; Thu, 10 Apr 1997 17:02:45 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: plans on release of 2.99.48 Message-Id: <19970410140245Z15824-5378+211@nic.funet.fi> Date: Thu, 10 Apr 1997 17:02:35 +0300 Return-Path: Hello, My internal (@tele.fi) users are asking for a stable release, and my intention is to make 2.99.48 such. Timewise the intention is to make that release on 14th of April (next Monday). After that, 2.99.48 will get bug-fixes if needed. (but hopefully none are -- feature addition will be the smtpserver policy filtering, unless I can finish it by monday.) I will cut short the day at the office (I spent it in lots of meetings anyway - without any feeling of getting something accomplished..) and get to my home machine for ZMailer coding... /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 10 17:24:49 1997 Received: from roger.interlynx.net ([204.50.196.97]) by nic.funet.fi with ESMTP id <16119-24407>; Thu, 10 Apr 1997 17:23:08 +0300 Received: from boris.interlynx.net (boris [204.50.196.17]) by roger.interlynx.net (8.7.5/8.6.12) with ESMTP id KAA20521 for ; Thu, 10 Apr 1997 10:22:39 -0400 (EDT) Received: from localhost (colins@localhost) by boris.interlynx.net (8.7.5/8.6.12) with SMTP id KAA00388 for ; Thu, 10 Apr 1997 10:23:02 -0400 (EDT) Date: Thu, 10 Apr 1997 10:23:02 -0400 (EDT) From: Colin Smillie To: zmailer@nic.funet.fi Subject: unsubscribe In-Reply-To: <19970410140245Z15824-5378+211@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: unsubscribe From zmailer-owner@nic.funet.fi Thu Apr 10 17:50:45 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <16319-24407>; Thu, 10 Apr 1997 17:48:39 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68048-23601>; Thu, 10 Apr 1997 18:48:30 +0400 Date: Thu, 10 Apr 1997 18:48:25 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: .48-970407.patch1b is OK To: Matti Aarnio cc: crosser@online.ru, zmailer@nic.funet.fi In-Reply-To: <19970409211753Z16260-20009+4@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > Although, I noticed a big (18Meg) smtp process now and started > > investigation. It revealed that the process is trying to deliver > > a hughe message (17M). Matti, does it mean that smtp agent keeps > > the whole message in memory? Is it expected behavior? ... > The message file (and the transport-descriptor file) are > opened RDONLY (and RW respectively), and then mmap()ed > MAP_FILE|MAP_SHARED. Ah, I understand. No need to worry about the resources. > > (No, in my previous reports, smtp processes grew to 50Meg while > > delivering average several Kb size messages). ... > Fixed in patch1b It is. 1b is running almost 24 hours and things look good. I slept well this night, no pager messages about the "mail system broken" :-) A candidate for a "stable" release? Eugene From zmailer-owner@nic.funet.fi Thu Apr 10 18:34:09 1997 Received: by nic.funet.fi id <15824-24407>; Thu, 10 Apr 1997 18:32:49 +0300 Subject: Re: .48-970407.patch1b is OK From: Matti Aarnio To: crosser@online.ru Date: Thu, 10 Apr 1997 18:32:44 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 10, 97 06:48:25 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 385 Message-Id: <19970410153249Z15824-24407+208@nic.funet.fi> Return-Path: > > > (No, in my previous reports, smtp processes grew to 50Meg while > > > delivering average several Kb size messages). > ... > > Fixed in patch1b > > It is. 1b is running almost 24 hours and things look good. I slept well > this night, no pager messages about the "mail system broken" :-) > A candidate for a "stable" release? Yes, very near so. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 11 01:13:32 1997 Received: from caipfs.rutgers.edu ([128.6.19.100]) by nic.funet.fi with ESMTP id <1860-55>; Fri, 11 Apr 1997 01:12:26 +0300 Received: from jenolan.caipgeneral (jenolan.rutgers.edu [128.6.111.5]) by caipfs.rutgers.edu (8.8.5/8.8.5) with SMTP id SAA13076; Thu, 10 Apr 1997 18:12:20 -0400 (EDT) Received: by jenolan.caipgeneral (SMI-8.6/SMI-SVR4) id SAA00599; Thu, 10 Apr 1997 18:11:04 -0400 Date: Thu, 10 Apr 1997 18:11:04 -0400 Message-Id: <199704102211.SAA00599@jenolan.caipgeneral> From: "David S. Miller" To: mea@nic.funet.fi CC: zmailer@nic.funet.fi In-reply-to: <19970410120830Z15635-24407+162@nic.funet.fi> (message from Matti Aarnio on Thu, 10 Apr 1997 15:08:21 +0300 (EET DST)) Subject: Re: SPAM and eBGP4 real-time map ? Return-Path: From: Matti Aarnio Date: Thu, 10 Apr 1997 15:08:21 +0300 (EET DST) I commented on something that Paul said on bind-workers, and as usual with elm, group-reply goes also to the sender. I am curious as to what Paul means with "eBGP4" feed, I must see his pages... That "feed" is a mailing list, when a large scale spammer is found Paul Vixie sends PGP signed BGP4 routing rules which those on the list add to their BGP4 router configurations, these rules block out any packets sent from the site where the spammer is found to originate from. This has been shown to be truly effective and piss spammers off especially since many people using that list run the primary backbones in the US, most of the time this cuts the spammers life lines off entirely. ---------------------------------------------//// Yow! 11.26 MB/s remote host TCP bandwidth & //// 199 usec remote TCP latency over 100Mb/s //// ethernet. Beat that! //// -----------------------------------------////__________ o David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ >< From zmailer-owner@nic.funet.fi Fri Apr 11 19:04:12 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <1212-22203>; Fri, 11 Apr 1997 19:03:11 +0300 Received: by redhotmomma.ssr.com id <36098-27097>; Fri, 11 Apr 1997 12:20:47 -0400 From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: 2.99.48 on SunOS 4.1.3_U1 Message-Id: <97Apr11.122047-0400_edt.36098-27097+22@redhotmomma.ssr.com> Date: Fri, 11 Apr 1997 12:20:44 -0400 Return-Path: Configure doesn't find tcpwrapper, and insists that SunOS 4.1.3_U1 doesn't have a modern dbm, so leaves out the dbm type --- looks like it is in libc, so no go with new zmailer... This tcpwrapper stuff is new --- how is it used in zmailer? sdb From zmailer-owner@nic.funet.fi Fri Apr 11 19:19:35 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <1970-22203>; Fri, 11 Apr 1997 19:18:24 +0300 Received: by redhotmomma.ssr.com id <36097-27107>; Fri, 11 Apr 1997 12:36:08 -0400 From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: 2.99.48 on SunOS 4.1.3_U1 Message-Id: <97Apr11.123608-0400_edt.36097-27107+32@redhotmomma.ssr.com> Date: Fri, 11 Apr 1997 12:35:57 -0400 Return-Path: Hmmm... update to my last message, it appears that the config.h file is generated correctly, with #define HAVE_NDBM_H and #define HAVE_DBM_ERROR set, so the warning is simply misleading. sdb From zmailer-owner@nic.funet.fi Fri Apr 11 19:34:23 1997 Received: from antares.utu.fi ([130.232.1.2]) by nic.funet.fi with ESMTP id <3014-13823>; Fri, 11 Apr 1997 19:33:15 +0300 Received: by utu.fi id <33345-989>; Fri, 11 Apr 1997 19:32:13 +0300 Subject: Re: 2.99.48 on SunOS 4.1.3_U1 In-Reply-To: <97Apr11.122047-0400_edt.36098-27097+22@redhotmomma.ssr.com> from Scott Ballantyne at "Apr 11, 97 12:20:44 pm" From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Fri, 11 Apr 1997 19:32:03 +0300 (EET DST) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 557 Message-Id: <19970411163213Z33345-989+1758@utu.fi> Return-Path: > Configure doesn't find tcpwrapper, and insists that SunOS 4.1.3_U1 > doesn't have a modern dbm, so leaves out the dbm type --- looks like > it is in libc, so no go with new zmailer... > > This tcpwrapper stuff is new --- how is it used in zmailer? It is not a new thing, just rewritten at the autoconfig. Following may be a bit cryptic for an explanation, though: :) hosts.allow: smtp-receiver : ALL : mailq : 127.0.0.1 195.1.2. hosts.deny: ALL : ALL : DENY (very lightly modified from smtp.tele.fi/smtp.inet.fi systems) > sdb /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 11 19:39:02 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <2231-13823>; Fri, 11 Apr 1997 19:38:07 +0300 Received: by redhotmomma.ssr.com id <36097-27105>; Fri, 11 Apr 1997 12:55:43 -0400 From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: -lresolv required in 2.99.48 on SunOS 4.1.3_U1 Message-Id: <97Apr11.125543-0400_edt.36097-27105+43@redhotmomma.ssr.com> Date: Fri, 11 Apr 1997 12:55:33 -0400 Return-Path: As I've mentioned before, the test for whether or not -lresolv is needed is inadquate. Many systems (netbsd, sunos, solaris) have the resolver routines in a shared library, but not the structure _res, for reasons explained in the Bind distribution, so I won't repeat them here. So present test for -lresolv concludes -lresolv not needed, because finds res_init() in shared lib, but then linking fails because no _res struct found... Please consider adding a _res.options = blah and a link as part of the configure test. sdb From zmailer-owner@nic.funet.fi Fri Apr 11 20:34:58 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <2156-55>; Fri, 11 Apr 1997 20:33:19 +0300 Received: by redhotmomma.ssr.com id <36097-27106>; Fri, 11 Apr 1997 13:50:53 -0400 From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: more on 2.99.48 on SunOS 4.1.3_U1 Message-Id: <97Apr11.135053-0400_edt.36097-27106+32@redhotmomma.ssr.com> Date: Fri, 11 Apr 1997 13:50:50 -0400 Return-Path: In zmsignal.h you have defined: #define SIGNAL_RELEASE(SIG) \ sigsetmask(~sigmask(SIG) & sigblock(SIG_HOLD(sig))) This doesn't work on SunOS 4.1.3_U1, since SIG_HOLD is defined as #define SIG_HOLD (void (*)())3 Not sure what to suggest here... sdb From zmailer-owner@nic.funet.fi Fri Apr 11 21:31:50 1997 Received: from american.com ([204.253.96.2]) by nic.funet.fi with ESMTP id <2591-18589>; Fri, 11 Apr 1997 21:29:39 +0300 Received: from hotsprings.american.com (hotsprings.american.com [204.253.96.21]) by american.com (8.7.6/8.6.12) with ESMTP id OAA06608 for ; Fri, 11 Apr 1997 14:29:22 -0400 (EDT) Received: from localhost (pgf@localhost) by hotsprings.american.com (8.6.12/8.6.9) with SMTP id OAA17832 for ; Fri, 11 Apr 1997 14:29:22 -0400 Message-Id: <199704111829.OAA17832@hotsprings.american.com> X-Authentication-Warning: hotsprings.american.com: Host localhost didn't use HELO protocol To: zmailer@nic.funet.fi Subject: Re: 2.99.48 on SunOS 4.1.3_U1 In-reply-to: mea's message of Fri, 11 Apr 1997 19:32:03 +0300. <19970411163213Z33345-989+1758@utu.fi> Reply-to: pgf@american.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <17830.860783361.1@hotsprings> Date: Fri, 11 Apr 1997 14:29:22 -0400 From: Paul Fox Return-Path: > > > > This tcpwrapper stuff is new --- how is it used in zmailer? > > It is not a new thing, just rewritten at the autoconfig. > Following may be a bit cryptic for an explanation, though: :) be sure tcpd was built with -DPROCESS_OPTIONS, or your examples won't work, i believe. see hosts_access(5) and hosts_options(5). paul --------------------- paul fox american internet corporation pgf@american.com (home: pgf@foxharp.boston.ma.us) From zmailer-owner@nic.funet.fi Fri Apr 11 21:44:24 1997 Received: by nic.funet.fi id <3020-13823>; Fri, 11 Apr 1997 21:42:52 +0300 Subject: Re: more on 2.99.48 on SunOS 4.1.3_U1 From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Fri, 11 Apr 1997 21:42:41 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr11.135053-0400_edt.36097-27106+32@redhotmomma.ssr.com> from "Scott Ballantyne" at Apr 11, 97 01:50:50 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 336 Message-Id: <19970411184252Z3020-13823+312@nic.funet.fi> Return-Path: > In zmsignal.h you have defined: > > #define SIGNAL_RELEASE(SIG) \ > sigsetmask(~sigmask(SIG) & sigblock(SIG_HOLD(sig))) > > This doesn't work on SunOS 4.1.3_U1, since SIG_HOLD is defined as > > #define SIG_HOLD (void (*)())3 > > Not sure what to suggest here... Oops... It should be SIGNAL_HOLD()... > sdb /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 11 22:51:19 1997 Received: from redhotmomma.ssr.com ([199.4.235.2]) by nic.funet.fi with ESMTP id <2984-13823>; Fri, 11 Apr 1997 22:50:00 +0300 Received: by redhotmomma.ssr.com id <36097-27097>; Fri, 11 Apr 1997 16:07:25 -0400 From: Scott Ballantyne To: mea@utu.fi CC: zmailer@nic.funet.fi In-reply-to: <19970411163213Z33345-989+1758@utu.fi> (message from Matti Aarnio on Fri, 11 Apr 1997 19:32:03 +0300 (EET DST)) Subject: Re: 2.99.48 on SunOS 4.1.3_U1 Message-Id: <97Apr11.160725-0400_edt.36097-27097+24@redhotmomma.ssr.com> Date: Fri, 11 Apr 1997 16:07:25 -0400 Return-Path: It is not a new thing, just rewritten at the autoconfig. Following may be a bit cryptic for an explanation, though: :) No, that's good. Why not just grep through inetd.conf for tcpd, instead of the way it is checked for now, in the configure script? sdb From zmailer-owner@nic.funet.fi Sat Apr 12 01:05:24 1997 Received: by nic.funet.fi id <2087-31899>; Sat, 12 Apr 1997 01:04:20 +0300 Subject: Re: 2.99.48 on SunOS 4.1.3_U1 From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Sat, 12 Apr 1997 01:04:18 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr11.160725-0400_edt.36097-27097+24@redhotmomma.ssr.com> from "Scott Ballantyne" at Apr 11, 97 04:07:25 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 736 Message-Id: <19970411220420Z2087-31899+15@nic.funet.fi> Return-Path: > It is not a new thing, just rewritten at the autoconfig. > Following may be a bit cryptic for an explanation, though: :) > > No, that's good. Why not just grep through inetd.conf for tcpd, > instead of the way it is checked for now, in the configure script? Because it is COMPILED IN facility, not external something that is run from the inetd, and which then execs the smtpserver. (The way the tcpd is usually used.) Both the smtpserver, and the scheduler have code that check on if the connection comes from an acceptable IP address. (which might not be worth of much...) If the header (tcpd.h) is present, but the libwrap.a can't be found, it isn't worth claiming having found it... > sdb /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Apr 12 01:50:33 1997 Received: by nic.funet.fi id <15965-31899>; Sat, 12 Apr 1997 01:49:20 +0300 Subject: Re: 2.99.48 on SunOS 4.1.3_U1 From: Matti Aarnio To: sdb@redhotmomma.ssr.com (Scott Ballantyne) Date: Sat, 12 Apr 1997 01:49:05 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr11.122047-0400_edt.36098-27097+22@redhotmomma.ssr.com> from "Scott Ballantyne" at Apr 11, 97 12:20:44 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 332 Message-Id: <19970411224920Z15965-31899+30@nic.funet.fi> Return-Path: > Configure doesn't find tcpwrapper, and insists that SunOS 4.1.3_U1 > doesn't have a modern dbm, so leaves out the dbm type --- looks like > it is in libc, so no go with new zmailer... Yes, the message is surprising. I will see if I can come up with anything to give a genuine diagnostics on this. .... > sdb /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 15 12:12:18 1997 Received: from falco.its.uu.se ([130.238.131.113]) by nic.funet.fi with ESMTP id <19675-12359>; Tue, 15 Apr 1997 12:11:16 +0300 Received: from alba.its.uu.se ([130.238.131.102]) by falco.its.uu.se with ESMTP id <2056-19230> convert rfc822-to-8bit; Tue, 15 Apr 1997 11:08:39 +0100 X-Mailer: exmh version 2.0gamma 1/27/96 To: zmailer@nic.funet.fi cc: Torbjorn.Wictorin@its.uu.se Subject: Bug in sm truncates messages on AIX Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8BIT Date: Tue, 15 Apr 1997 11:11:12 +0200 From: Martin Wendel Message-Id: <97Apr15.110839+0100msz.2056-19230+8@falco.its.uu.se> Return-Path: Bug report Affected OS: AIX zmailer versions: the later versions. Description (just discovered): The zmailer specific message buffering stuff in sm, using the readalready size parameter, does not work on AIX. Since file IO is handled by the paging system on AIX, the OS handles file buffering perfectly by itself. When zmailer tries to do buffering on its own, this comes in conflict with the OS buffering resulting in a 4096 byte (one page) size limit on the zmailer buffers. Unfortunately zmailer is unaware of this limit and happily sends the contents of the buffer, a truncated message. I've seen this happen on one of our production systems where readalready never exceeded 4096 over several thousands messages. Users complained about truncated messages with ca 3k attachments. When looking at cases where readalready equaled 4096 messages where truncated. I've commented out the stuff in my version, probably this will fix the problem. -- E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS Tel: 018 18 77 80 Box 887 Int: +46 18 18 77 80 S-751 08 Uppsala Fax: +46 18 18 79 10 Sweden http://www.uu.se/Internservice/ue/personal/MartinWendel.html From zmailer-owner@nic.funet.fi Tue Apr 15 17:58:36 1997 Received: by nic.funet.fi id <19652-12358>; Tue, 15 Apr 1997 17:56:45 +0300 Subject: Re: Bug in sm truncates messages on AIX From: Matti Aarnio To: martin.wendel@its.uu.se (Martin Wendel) Date: Tue, 15 Apr 1997 17:56:41 +0300 (EET DST) Cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se In-Reply-To: <97Apr15.110839+0100msz.2056-19230+8@falco.its.uu.se> from "Martin Wendel" at Apr 15, 97 11:11:12 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1517 Message-Id: <19970415145645Z19652-12358+22@nic.funet.fi> Return-Path: > Bug report > Affected OS: AIX > zmailer versions: the later versions. > > Description (just discovered): > The zmailer specific message buffering stuff in sm, using > the readalready size parameter, does not work on AIX. Since > file IO is handled by the paging system on AIX, the OS handles > file buffering perfectly by itself. When zmailer tries to do > buffering on its own, this comes in conflict with the OS > buffering resulting in a 4096 byte (one page) size limit on > the zmailer buffers. Unfortunately zmailer is unaware of this > limit and happily sends the contents of the buffer, a truncated > message. The 'readalready' was not reset in all cases after the branch of (convertmode == _CONVERT_NONE) had done its reading of MULTIPLE buffer-fulls. ("multiple" meaning: "more than one") > I've seen this happen on one of our production systems where > readalready never exceeded 4096 over several thousands messages. > Users complained about truncated messages with ca 3k attachments. > When looking at cases where readalready equaled 4096 messages > where truncated. I've commented out the stuff in my version, > probably this will fix the problem. My current snapshot patch2 does address this problem. (patch1 handled another problem that kept me from releasing 2.99.48 yesterday..) ( Snapshots are at location: ftp://mea.tmt.tele.fi/ ) > -- > E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS > http://www.uu.se/Internservice/ue/personal/MartinWendel.html /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 15 20:26:10 1997 Received: from chronos ([194.67.3.135]) by nic.funet.fi with ESMTP id <19490-12360>; Tue, 15 Apr 1997 20:24:11 +0300 Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-23601>; Tue, 15 Apr 1997 21:23:59 +0400 Date: Tue, 15 Apr 1997 21:23:48 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: still hangover smtp processes To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Matti, I hate to say it, but with .48-970407 patch 1b I still get long-hanging smtp transporters. Today I killed a few that started on Apr 11. Here is a sample backtrace: (gdb) bt #0 0xef636d58 in _end () #1 0xef6732cc in _end () #2 0xef6475f8 in _end () #3 0xef6f2158 in _end () #4 0xef6d77c8 in _end () #5 0xef6f1bb8 in _end () #6 0x1ad9c in getmxrr (SS=0xeffff980, host=0x4a638 "arstel.ru", mx=0xefffebb0, maxmx=128, depth=0, mxcntp=0xefffe5ac) at smtp.c:3720 #7 0x16ec0 in smtpconn (SS=0xeffff980, host=0x4a638 "arstel.ru", noMX=0) at smtp.c:2137 #8 0x16a34 in smtpopen (SS=0xeffff980, host=0x4a638 "arstel.ru", noMX=0) at smtp.c:1939 #9 0x14254 in main (argc=0, argv=0x4a638) at smtp.c:862 smtp.c:3720 is gethostbyname() call. Solaris 2.5.1. Eugene From zmailer-owner@nic.funet.fi Wed Apr 16 18:10:42 1997 Received: from pobox3.funet.fi ([128.214.46.64]) by nic.funet.fi with ESMTP id <11862-1049>; Wed, 16 Apr 1997 18:05:34 +0300 Received: from chronos (chronos.sovam.com [194.67.3.135]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with ESMTP id QAA02758 for ; Wed, 16 Apr 1997 16:54:35 +0300 (EET DST) Received: from ariel ([194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-322>; Wed, 16 Apr 1997 17:54:27 +0400 Date: Wed, 16 Apr 1997 17:54:25 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Hanging smtp transport processes To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Following up to my previous posting, I beleive that smtp program is not at fault for hanging gethostbyname() - I have other programs (ipop3d) hanging the same way. I think this is a bug in NIS+ client code: get...by...() functions tend to hang if NIS+ was not properly working at the moment. Eugene From zmailer-owner@nic.funet.fi Wed Apr 16 20:43:06 1997 Received: by nic.funet.fi id <2311-1049>; Wed, 16 Apr 1997 20:36:50 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: Still new snapshots... Message-Id: <19970416173650Z2311-1049+70@nic.funet.fi> Date: Wed, 16 Apr 1997 20:32:30 +0300 Return-Path: ... and no actual release. If you pick my snapshots (ftp://mea.tmt.tele.fi/), after you have applied the patches, you need to do: touch configure That because both 'configure.in', and 'configure' change, but your machine might be a bit too fast, and patch both files at the same time -- or 'configure.in' before 'configure', and thus causing instant need for 'autoconf' run -- which needs GNU autoconf 2.12.. Anyway, nic.funet.fi had also some weird problems, and it may have affected the email that has traversed thru the server... /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 22 12:01:11 1997 Received: from qmailr@cam053212.student.utwente.nl (port 11562 [130.89.226.142]) by nic.funet.fi with SMTP id <460-560>; Tue, 22 Apr 1997 11:59:25 +0300 Received: (qmail 9771 invoked by uid 1000); 22 Apr 1997 08:59:21 -0000 Date: Tue, 22 Apr 1997 10:59:01 +0200 (CEST) From: Remco van de Meent To: zmailer@nic.funet.fi Subject: zmailer on Debian systems Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hey I'm trying to install ZMailer 2.99.46 on my Debian 1.2.9 box. David Trueman posted a message about the router segfaulting on his Debian box at Thu, 6 Feb 1997, using .45. There was a reply on that message which stated that the problem should be resolved in .46. But I got those segfauls even using .46 :( I think the cause of the problem is in the 'strange' *db*.h files and libraries, which come with Debian. So, my question is: is anyone using ZMailer on a Debian machine successfully? Or maybe anyone has a solution, or some ideas about how to solve it? Thanks! bye -R. // Remco van de Meent // email: remco@oloon.student.utwente.nl // www: http://oloon.student.utwente.nl // " Never make any mistaeks. " From zmailer-owner@nic.funet.fi Tue Apr 22 12:22:42 1997 Received: from root@iug.org (port 13912 [194.86.187.30]) by nic.funet.fi with ESMTP id <311-5477>; Tue, 22 Apr 1997 12:21:39 +0300 Received: by the.iug.org id <20407-30633>; Tue, 22 Apr 1997 12:21:31 +0300 Subject: ndbm database problems with fullnames. From: Niko Lilja To: zmailer@nic.funet.fi Date: Tue, 22 Apr 1997 12:21:27 +0300 (EET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 341 Message-Id: <97Apr22.122131+0300_eet_dst.20407-30633+1@the.iug.org> Return-Path: Hello. I have following problem with zmailer fullname-database. When I run makendbm ndbm fullnames it works great, but it makes gdbm database not ndbm. And of course, the router can't open it ndbm-database. So, does someone have solution for this how to get router looking also gdbm-database? Regarding -- (Niko Lilja ) From zmailer-owner@nic.funet.fi Tue Apr 22 18:00:17 1997 Received: by nic.funet.fi id <1262-5291>; Tue, 22 Apr 1997 17:58:15 +0300 Subject: Re: ndbm database problems with fullnames. From: Matti Aarnio To: nmk@iug.org (Niko Lilja) Date: Tue, 22 Apr 1997 17:58:09 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr22.122131+0300_eet_dst.20407-30633+1@the.iug.org> from "Niko Lilja" at Apr 22, 97 12:21:27 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 779 Message-Id: <19970422145815Z1262-5291+182@nic.funet.fi> Return-Path: > Hello. > > I have following problem with zmailer fullname-database. > > When I run makendbm ndbm fullnames it works great, but > it makes gdbm database not ndbm. Really ? Which zmailer version you have ? ... because the 'makendbm' used to have some problems in this regard until fairly recently. Which platform ( i.e.: Linux kernel + libc + distribution versions ) you have ? Which databases does it have ? > And of course, the router can't open it ndbm-database. > > So, does someone have solution for this how to get router > looking also gdbm-database? Sure, look at the 'relation' entry defining the 'fullnames' database. There is the '-t ndbm' (or whatever) definition. > Regarding > -- > (Niko Lilja ) /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Apr 22 18:48:31 1997 Received: from root@iug.org (port 14610 [194.86.187.30]) by nic.funet.fi with ESMTP id <2028-31344>; Tue, 22 Apr 1997 18:47:37 +0300 Received: by the.iug.org id <20462-1>; Tue, 22 Apr 1997 18:47:28 +0200 Subject: Re: ndbm database problems with fullnames. In-Reply-To: <19970422145815Z1262-5291+182@nic.funet.fi> from Matti Aarnio at "Apr 22, 97 05:58:09 pm" From: Niko Lilja To: mea@nic.funet.fi (Matti Aarnio) Date: Tue, 22 Apr 1997 18:47:24 +0300 (EET DST) Cc: zmailer@nic.funet.fi MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1019 Message-Id: <97Apr22.184728+0200eet_dst.20462-1+2@the.iug.org> Return-Path: > > Hello. > > > > I have following problem with zmailer fullname-database. > > > > When I run makendbm ndbm fullnames it works great, but > > it makes gdbm database not ndbm. > > Really ? Which zmailer version you have ? > ... because the 'makendbm' used to have some problems in > this regard until fairly recently. > I have zmailer version 2.99.45. > Which platform ( i.e.: Linux kernel + libc + distribution > versions ) you have ? Which databases does it have ? > Linux 2.0.30 / libc version 5.3.12 / redhat 4.1. gdbm, db, and "ndbm-emulation" in libdb.so.2.0.0. > > And of course, the router can't open it ndbm-database. > > > > So, does someone have solution for this how to get router > > looking also gdbm-database? > > Sure, look at the 'relation' entry defining the 'fullnames' > database. There is the '-t ndbm' (or whatever) definition. > I get it work when I get router binary from other machine :) > > /Matti Aarnio -- (Niko Lilja ) From zmailer-owner@nic.funet.fi Wed Apr 23 00:24:51 1997 Received: from NO-IDENT-SERVICE@hermes.hrz.uni-giessen.de (port 12458 [134.176.2.15]) by nic.funet.fi with ESMTP id <179-28309>; Wed, 23 Apr 1997 00:18:16 +0300 Received: from netserv.lahn.de by hermes.hrz.uni-giessen.de; Tue, 22 Apr 1997 22:54:59 +0200 Received: from newton.lahn.de by netserv.lahn.de via rsmtp with bsmtp id for ; Tue, 22 Apr 1997 22:54:57 +0200 (CEST) (Smail-3.2 1996-Jul-4 #1 built 1997-Jan-12) Received: by newton.lahn.de (UMSRFC 0.12); Tue, 22 Apr 1997 22:53:12 +0200 To: "zmailer" From: "Michael Leun" Date: Tue, 22 Apr 1997 20:13:01 +0200 X-Mailer: IntuiNews 1.4 (23.7.96) Subject: Batched SMTP Message-ID: <75208821@newton.lahn.de> Organization: UUCP-Freunde Lahn e.V. Return-Path: Hello! Maybe a FAQ, but I'm new to zmailer and I have even no idea where to get the FAQ :-) Is it possible to send mail as batched SMTP via UUCP with zmailer? I searched the sources for the string "MAIL FROM:" which is used for batched SMTP and only found it in smtp.c and smtpserver.c, I think those are for SMTP over TCP/IP only. --- tschuess, Michael PGP Key available - use it From zmailer-owner@nic.funet.fi Wed Apr 23 01:27:31 1997 Received: by nic.funet.fi id <135-2517>; Wed, 23 Apr 1997 01:24:03 +0300 Subject: Re: Batched SMTP From: Matti Aarnio To: ml@newton.lahn.de (Michael Leun) Date: Wed, 23 Apr 1997 01:23:52 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <75208821@newton.lahn.de> from "Michael Leun" at Apr 22, 97 08:13:01 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1503 Message-Id: <19970422222403Z135-2517+28@nic.funet.fi> Return-Path: > Hello! > > Maybe a FAQ, but I'm new to zmailer and I have even no idea where to get > the FAQ :-) Nobody has written one ? (I haven't anyway..) > Is it possible to send mail as batched SMTP via UUCP with zmailer? > > I searched the sources for the string "MAIL FROM:" which is used for > batched SMTP and only found it in smtp.c and smtpserver.c, I think those > are for SMTP over TCP/IP only. The 'sm' program can do it. Following is an adaptation from the 'uucp' channel specification at the standard distribution. Define a 'uucpb' channel, which you configure in the MAILSHARE/scheduler.conf as: uucpb/* ... command="sm -c uucpb uucpb" Then at the 'MAILSHARE/sm.conf' have: uucpb Ub /usr/bin/uux uux - -r -a$g -gC some-bsmtp-cmd At the "Flags" field the important one is the 'b' for the Batch-SMTP. You can also have 'B' or 'BB' for two grades of ESMTP. More details in the 'man sm'. Hmm.. A new 'channel' needs to be defined in MAILSHARE/cf/crossbar.cf too to select proper way to header rewrite rules. Propably it should be more like smtp* -channels, than like uucp, though.. For the BSMTP reception the smtpserver does pretty well. MAILBIN/smtpserver -i > /dev/null < uucp.input.file If you want smarter bsmtp error reporter, tell me how to do it. My handicap is the lack of any UUCP links at my systems... (I do have an NJE link, but the BITNET is dying...) > --- > tschuess, Michael > PGP Key available - use it /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 23 09:05:32 1997 Received: from root@iug.org (port 41025 [194.86.187.30]) by nic.funet.fi with ESMTP id <303-2517>; Wed, 23 Apr 1997 09:04:15 +0300 Received: by the.iug.org id <20528-1452>; Wed, 23 Apr 1997 09:04:04 +0300 Subject: database problem solved. From: Niko Lilja To: zmailer@nic.funet.fi Date: Wed, 23 Apr 1997 09:03:59 +0300 (EET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 479 Message-Id: <97Apr23.090404+0300_eet_dst.20528-1452+1@the.iug.org> Return-Path: Heh, just solved the mystery puzzle of database. When you're linking the binary of router and makendbm, be sure that you heva changed the libs order in Makefile in those sources directory to: ROUTER_LIB= -lndbm -lgdbm -ldb and similar order in makendbm's Makefile. Just tested it and it works fine now, so it depends what order you link the libraries in the binary :) Anyway, now I am happy (after 10 hours head banging to the wall) :) -- (Niko Lilja ) From zmailer-owner@nic.funet.fi Wed Apr 23 12:52:08 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 55698 [194.67.3.135]) by nic.funet.fi with ESMTP id <156-1776>; Wed, 23 Apr 1997 12:50:03 +0300 Received: from NO-IDENT-SERVICE@ariel (port 34432 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-15330>; Wed, 23 Apr 1997 13:49:38 +0400 Date: Wed, 23 Apr 1997 13:49:37 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Syslog reporting To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Matti, in recent betas, syslog reporting made a big step forward. With the current scheme, it is (at last!) possible to track down any message passing thru the system with time stamps. Although, I'd like to make a few comments: 1. I think that it was a *good* idea to log the *arrival* of a message, as it was done in patch5. Somehow, this was commented out in patch6. If arrival reports where done in smtpserver and sendmail, the logging would cover the whole lifetime of the message, not just the period from routing till delivery, as in patch6+ 2. As there are several different transport programs, it is not trivial to distinguash delivery records for an automated syslog processor. I would suggest to add a "function" tag to the log format, e.g.: vvvvvvvv Apr 23 13:16:28 deimos smtpserver[14939]: accepted .nLRBg227587: \ from ts5-a13.dial.sovam.com/1031 vvvvvv Apr 23 13:16:31 deimos router[4224]: routed .nLRBg227587: \ from=, relay=ts5-a13.dial.sovam.com \ ([194.67.14.173]) size=868, nrcpts=1, \ msgid=<01BC4FE8.EA184F80@ts5-a13.dial.sovam.com> vvvvvvvvv Apr 23 13:16:38 deimos smtp[14638]: delivered .nLRBg227587: \ to=embassy@embas-kyrg.msk.ru, delay=00:00:10, \ xdelay=00:00:06, mailer=smtp, relay=relay1.kiae.su, stat=ok2 Eugene From zmailer-owner@nic.funet.fi Wed Apr 23 17:45:36 1997 Received: by nic.funet.fi id <12783-28309>; Wed, 23 Apr 1997 17:44:16 +0300 Subject: Re: database problem solved. From: Matti Aarnio To: nmk@iug.org (Niko Lilja) Date: Wed, 23 Apr 1997 17:44:08 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Apr23.090404+0300_eet_dst.20528-1452+1@the.iug.org> from "Niko Lilja" at Apr 23, 97 09:03:59 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1298 Message-Id: <19970423144416Z12783-28309+178@nic.funet.fi> Return-Path: > Heh, just solved the mystery puzzle of database. > > When you're linking the binary of router and makendbm, be sure that > you heva changed the libs order in Makefile in those sources directory to: > > ROUTER_LIB= -lndbm -lgdbm -ldb Using order of: -lndbm -ldb -lgdbm should work equally well, but apparently it didn't ? Ok, generally speaking the problem is that both the DB, and the GDBM have NDBM-like compability routines in them, which does complicate things considerably. To top it off, any/all of these can be in the system libc too.. The 'configure.in' source has following comment, which I added after one instance of juggling these libraries around: # GDBM must be last as if we have LIBDB at the system instead of real # LIBNDBM, we still have ndbm routines, but want to use (likely) # ones from LIBDB instead of LIBGDBM... Any comments ? (If you want to edit the 'configure.in', you do need GNU autoconf-2.12 package to regenerate the 'configure' script.) > and similar order in makendbm's Makefile. > > Just tested it and it works fine now, so it depends what order you > link the libraries in the binary :) > > Anyway, now I am happy (after 10 hours head banging to the wall) :) > > -- > (Niko Lilja ) /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 23 17:56:29 1997 Received: by nic.funet.fi id <12676-2517>; Wed, 23 Apr 1997 17:54:46 +0300 Subject: Re: zmailer on Debian systems From: Matti Aarnio To: remco@oloon.student.utwente.nl (Remco van de Meent) Date: Wed, 23 Apr 1997 17:54:43 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Remco van de Meent" at Apr 22, 97 10:59:01 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1238 Message-Id: <19970423145446Z12676-2517+193@nic.funet.fi> Return-Path: > Hey > > I'm trying to install ZMailer 2.99.46 on my Debian 1.2.9 box. > > David Trueman posted a message about the router segfaulting on his Debian > box at Thu, 6 Feb 1997, using .45. There was a reply on that message which > stated that the problem should be resolved in .46. But I got those segfauls > even using .46 :( > > I think the cause of the problem is in the 'strange' *db*.h files and > libraries, which come with Debian. > > So, my question is: is anyone using ZMailer on a Debian machine > successfully? Or maybe anyone has a solution, or some ideas about how to > solve it? Getting GDB backtraces would be most instructive, and perhaps if you could try my 2.99.48beta snapshots ? ftp://mea.tmt.tele.fi/ Also getting the transport- and rfc-822 envelopes of the message in processing when the crash occurs would tell something to me. ... and if this crash occurs only with one particular message, while others run thru successfully, or does it happen with all messages ? (I am holding the release of 2.99.48, because I haven't had time to go over the changes, and to log them properly..) > Thanks! > > bye -R. > > // Remco van de Meent > // email: remco@oloon.student.utwente.nl /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 23 18:07:36 1997 Received: from NO-IDENT-SERVICE@roger.interlynx.net (port 7645 [204.50.196.97]) by nic.funet.fi with ESMTP id <1405-1776>; Wed, 23 Apr 1997 18:06:09 +0300 Received: from boris.interlynx.net (boris [204.50.196.17]) by roger.interlynx.net (8.7.5/8.6.12) with ESMTP id LAA02494 for ; Wed, 23 Apr 1997 11:05:23 -0400 (EDT) Received: from localhost (colins@localhost) by boris.interlynx.net (8.7.5/8.6.12) with SMTP id LAA13221 for ; Wed, 23 Apr 1997 11:05:52 -0400 (EDT) Date: Wed, 23 Apr 1997 11:05:49 -0400 (EDT) From: Colin Smillie To: zmailer@nic.funet.fi Subject: unsubscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: unsubscribe From zmailer-owner@nic.funet.fi Wed Apr 23 18:33:22 1997 Received: from NO-IDENT-SERVICE@antares.utu.fi (port 39158 [130.232.1.2]) by nic.funet.fi with ESMTP id <797-1776>; Wed, 23 Apr 1997 18:31:36 +0300 Received: by utu.fi id <34121-1343>; Wed, 23 Apr 1997 18:30:30 +0300 Subject: Re: unsubscribe In-Reply-To: from Colin Smillie at "Apr 23, 97 11:05:49 am" From: Matti Aarnio To: colins@interlynx.net (Colin Smillie) Date: Wed, 23 Apr 1997 18:30:17 +0300 (EET DST) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 238 Message-Id: <19970423153030Z34121-1343+286@utu.fi> Return-Path: > unsubscribe The administrative support for the zmailer -list is on addresses: Actually the first is an alias to the second. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Apr 23 20:08:35 1997 Received: from root@cam053212.student.utwente.nl (port 34941 [130.89.226.142]) by nic.funet.fi with ESMTP id <248-1776>; Wed, 23 Apr 1997 20:05:02 +0300 Received: by cam053212.student.utwente.nl id <166928-7801>; Wed, 23 Apr 1997 19:04:29 +0200 Date: Wed, 23 Apr 1997 19:04:22 +0200 (CEST) From: Remco van de Meent To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: zmailer on Debian systems In-Reply-To: <19970423145446Z12676-2517+193@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 23 Apr 1997, Matti Aarnio wrote: [about the router segfaulting in .46] > > Getting GDB backtraces would be most instructive, and perhaps > if you could try my 2.99.48beta snapshots ? > > ftp://mea.tmt.tele.fi/ > > Also getting the transport- and rfc-822 envelopes of the message > in processing when the crash occurs would tell something to me. > ... and if this crash occurs only with one particular message, > while others run thru successfully, or does it happen with all > messages ? It happened with all messages. I didn't get any messages through the router. But I wanted to install zmailer, and while looking at my system, I said "why not installing correct libraries, which everyone is using", so I removed the Debian include files and lib's, and replaced them by the ones you can find at every GNU-mirror. And now it all works fine. I'm sorry, but I cannot give you any traces. I didn't use GDB, but I did a 'strace' once. I remember that it was complaining about 'seek' before it got killed by segfault. But I don't remember the exact calls. bye -R. // Remco van de Meent // email: remco@oloon.student.utwente.nl // www: http://oloon.student.utwente.nl // " Never make any mistaeks. " From zmailer-owner@nic.funet.fi Wed Apr 23 22:19:14 1997 Received: from root@mea.tmt.tele.fi (port 63331 [194.251.252.33]) by nic.funet.fi with ESMTP id <1404-28309>; Wed, 23 Apr 1997 22:13:25 +0300 Received: from mea@localhost by mea.tmt.tele.fi id <231462-29944>; Wed, 23 Apr 1997 22:14:45 +0300 Sender: Matti Aarnio From: Matti Aarnio To: zmailer@nic.funet.fi Subject: New snapshot -- last of 2.99.48 ? Message-Id: <19970423191445Z231462-29944+4@mea.tmt.tele.fi> Date: Wed, 23 Apr 1997 22:14:36 +0300 Return-Path: Ok, I spent a day at finalizing a distribution, but I still call this 'beta' snapshot. In a few days I will modify the version name to be '2.99.48', and store it to ftp.funet.fi, but until then, I am interested in your reports. Snapshot location: ftp://mea.tmt.tele.fi/ /Matti Aarnio Wed Apr 23 17:58:19 1997 Matti Aarnio * Makefile.in: Version 2.99.48-970423 * configure.in: - Rearranged libresolv autoconfiguration - Test for possible need of -lresolv to satisfy the access to the structure: _res - Corrected the RFC822TABS (--without-rfc822-tabs) option - Do GCC optimizer resistant test on existence of external objects, e.g. char *sys_siglist[]; * include/ta.h, transports/libta/tasyslog.c: New parameter (msg) to the tasyslog() routine * include/zsyslog.h: Yield an error if the system does not have proper (TODO: Have acceptable substitutes for that case..) * libc/mail.c, smtpserver/smtpdata.c: The mail_close() routine is a wrapper for an internal routine that returns information regarding the last status of the spool file for the smtpserver to be able to report that at the message reception. * router/libdb/header.c, man/zmailer.3: Define the "rcvdfrom" to require a "user@address" type of parameter * smtpserver/smtpcmds.c: Store the "rcvdfrom" information from the IDENT information. * router/rfc822.c, router/router.c: If the system can't resolve internal uname->uid/uid->uname mappings (getpwnam() fails ?), then defer the message. If the message has a "rcvdfrom" header, and the message is not submitted by any of the "trusted" UIDs, replace the "rcvdfrom" information with a new one we fabricate * router/rfc822.c, router/rtsyslog.c: router's syslog (rtsyslog()) has a new parameter * smtpserver/smtpserver.c: - Don't syslog() anything, unless indicating an error! - Recognize IN6_IS_ADDR_V4MAPPED() type of addresses, and do reversal lookup in proper IPv4 manner. * transports/libta/diagnostics.c, transports/libta/tasyslog(): - notary_setwttip() to behave in similar manner to notary_setwtt(), but to define the IP address that the host has, and we have made a connection with. - tasyslog() takes that information as a parameter - tasyslog() has new sprintf() format strings with output width limitations * transports/libta/mimeheaders.c: Rewrote (somewhat) the text of the "X-Warning:" -lines * transports/mailbox/mailbox.c: - The recipient address is properly dequoted, and localized for feeding to the local processing (to a file, pipe, or to store at somebodys mailbox) - HP-UX 9.x oriented: 'fflush() returns an error in case the FP points to a PIPE, but lets hope it won't set ferror() unless it really is an error!' * transports/sm/sm.c: - Process bigger let_buffer[]s than before - If the system reads more than one bufferfull, always remember to zero the 'readalready'. * transports/smtp/smtp.c: - Report the notary_setwttip() along with notary_setwtt() data for the diagnostics - smtp pipelining code rearrangement completion * transports/libta/tasyslog.c: Don't be shy about telling what the return message of the transporter was when tasyslog()ing it, and the code was OK. From zmailer-owner@nic.funet.fi Wed Apr 23 22:29:04 1997 Received: from NO-IDENT-SERVICE@press-gopher.uchicago.edu (port 42671 [128.135.234.36]) by nic.funet.fi with ESMTP id <12437-2517>; Wed, 23 Apr 1997 22:25:37 +0300 Received: from localhost by press-gopher.uchicago.edu with SMTP id <26521-16297>; Wed, 23 Apr 1997 14:25:22 -0500 Date: Wed, 23 Apr 1997 14:25:17 -0500 (CDT) Sender: Roy C Bixler From: Roy C Bixler To: Remco van de Meent cc: zmailer@nic.funet.fi Subject: Re: zmailer on Debian systems In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 23 Apr 1997, Remco van de Meent wrote: > It happened with all messages. I didn't get any messages through the router. > But I wanted to install zmailer, and while looking at my system, I said "why > not installing correct libraries, which everyone is using", so I removed the > Debian include files and lib's, and replaced them by the ones you can find > at every GNU-mirror. > And now it all works fine. I use Debian Linux and Z-mailer (version 2.99.48-970407 + patch 1b) with no problems. In fact, such a system is now in production here as our primary mailserver processing a good number of messages each day without crashing. What version of the db libraries did you have installed? Here is the list from the mailserver: ii libdb1 1.85.4-3 The Berkeley database routines (runtime vers ion) ii libgdbm1 1.7.3-19 GNU dbm database routines (runtime version). This is more or less on the current 'bo' (aka 'frozen') version. Roy rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Thu Apr 24 00:43:17 1997 Received: from root@pccross.average.org (port 51540 [194.67.9.6]) by nic.funet.fi with ESMTP id <1502-24419>; Thu, 24 Apr 1997 00:41:26 +0300 Received: from localhost(really [127.0.0.1]) by pccross.average.org via smtpd with esmtp id for ; Thu, 24 Apr 1997 01:44:01 +0400 (MSD) (Smail-3.2 1996-Jul-4 #12 built 1996-Jul-20) Date: Thu, 24 Apr 1997 01:44:01 +0400 (MSD) From: Eugene Crosser Sender: crosser@pccross.average.org Reply-To: Eugene Crosser Subject: router coredump in logit() To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; LANGUAGE=ru Return-Path: Matti, the code of logit() function around router.c:570 is incorrect. It tries to cut down the "from" string if it is too long, but if the "id" string is too long, it happens that baselen is bigger that strlen(from)+MAXSAFESIZE, and offset in the line 574 "c = *(from+MAXSAFESIZE-baselen);" is negative. This causes segfault. This happend when someone put a junk in the Message-ID header: Message-ID: .........................something Zmailer converted every dot into the text "(illegal something blah-blah)" and "id" became longer than 1Kb. Eugene From zmailer-owner@nic.funet.fi Thu Apr 24 13:17:53 1997 Received: from root@tuminfo2.informatik.tu-muenchen.de (port 43447 [131.159.0.81]) by nic.funet.fi with ESMTP id <12601-24419>; Thu, 24 Apr 1997 13:16:42 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109523-3568>; Thu, 24 Apr 1997 12:16:17 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <18949-216>; Thu, 24 Apr 1997 12:16:05 +0200 Date: Thu, 24 Apr 1997 12:15:48 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: new zmailer beta Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: -----BEGIN PGP SIGNED MESSAGE----- Hello there, the new beta version of zmailer doesn't even compile on HP-UX. Isn't there someone who runs zmailer on HP-UX ? I still can't find a version that works flawlessly on HP-UX. This time the problem is in bsdbtree.c where nowhere BTREEINFO is defined. Greets, Sven - -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: latin1 iQCVAwUBM18y10gFYR9zZnaVAQEMSAQAnOSkgq5EAGXJGJy0ai2IwsCmB3vmIZws b5MIeK9V7lY2wpoN3gGLBs/bWd3bEHkFcYRejtwpudyVag2z4VRmPa3xz2LSisLl y0QCt4gQSxXF4mtxNd5pX0eb1Vcwer6rB/kdwT6kdO7Gzur3Dw4U8pHuGjdR+wGj Bw6rl0B5ltM= =6AhI -----END PGP SIGNATURE----- From zmailer-owner@nic.funet.fi Thu Apr 24 16:15:12 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 2730 [194.67.3.135]) by nic.funet.fi with ESMTP id <12511-24297>; Thu, 24 Apr 1997 16:14:00 +0300 Received: from NO-IDENT-SERVICE@ariel (port 36831 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-15325>; Thu, 24 Apr 1997 17:13:08 +0400 Date: Thu, 24 Apr 1997 17:13:02 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: .48-970423 glitch To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: The previous beta (970414 patch7) compiled clean, but this one bombs with: gcc -g -O2 -DHAVE_CONFIG_H -I.. -I../include -c whathost.c whathost.c:40: mntent.h: No such file or directory *** Error code 1 on SPARC Solaris 2.5.1 Eugene From zmailer-owner@nic.funet.fi Thu Apr 24 16:36:16 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 28330 [194.67.3.135]) by nic.funet.fi with ESMTP id <2010-24419>; Thu, 24 Apr 1997 16:34:26 +0300 Received: from NO-IDENT-SERVICE@ariel (port 36832 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-15330>; Thu, 24 Apr 1997 17:34:10 +0400 Date: Thu, 24 Apr 1997 17:34:08 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: re: the glitch To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Followup to my previous post: after I removed site.config and reran ./configure, I got a 'correct' config.h. Eugene From zmailer-owner@nic.funet.fi Thu Apr 24 17:43:14 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 27308 [194.67.3.135]) by nic.funet.fi with ESMTP id <1365-26696>; Thu, 24 Apr 1997 17:41:41 +0300 Received: from NO-IDENT-SERVICE@ariel (port 36977 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68047-15859>; Thu, 24 Apr 1997 18:41:27 +0400 Date: Thu, 24 Apr 1997 18:41:13 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Another glitch with latest beta To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Another thing with .48-970423: pipe works `strange' in the mailbox transport. I have an entry in /etc/zmailer/db/aliases: mailping: "|exec /usr/local/bin/mailping" and it used to work, but now the script *is* executed, but the message is not removed from the queue and the following is in the syslog: Apr 24 18:39:12 chronos mailbox[15922]: S.nLr/p68048: to="|exec /usr/local/bin/mailping", delay=00:00:27, xdelay=00:00:00, mailer=local, stat=deferred message write to ""|exec /usr/local/bin/mailping"" failed: Error 0 Eugene From zmailer-owner@nic.funet.fi Thu Apr 24 17:51:42 1997 Received: from NO-IDENT-SERVICE@antares.utu.fi (port 59281 [130.232.1.2]) by nic.funet.fi with ESMTP id <1845-24419>; Thu, 24 Apr 1997 17:48:57 +0300 Received: by utu.fi id <34225-1343>; Thu, 24 Apr 1997 17:47:57 +0300 Subject: Re: new zmailer beta In-Reply-To: from Sven Goldt at "Apr 24, 97 12:15:48 pm" From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Thu, 24 Apr 1997 17:47:52 +0300 (EET DST) Cc: zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1156 Message-Id: <19970424144757Z34225-1343+1010@utu.fi> Return-Path: > Hello there, > > the new beta version of zmailer doesn't even compile on HP-UX. > Isn't there someone who runs zmailer on HP-UX ? I still can't find > a version that works flawlessly on HP-UX. > This time the problem is in bsdbtree.c where nowhere > BTREEINFO is defined. Yeah, I have been playing with it at HP-UX 10.20 with gcc, gas, and GNU-make. I got it compiling, and even routing/delivering mail when I configure it as: CC="gcc -D_POSIX_SOURCE" ./configure ... However it still barfs at me when I try to run a pipe from my .forward :-( Oh yes, I had to do some changes at the Makefile.in's sed scripts to satisfy HP-UX sed ... I will provide diffs of things I am working at at the moment. > Greets, Sven > -- > PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey /Matti Aarnio "A hacker dressed in a three piece suit isn't very relaxed one" (That's me -- I dressed up today for going out with one of our bigger bosses to meet other suited people. The bosses had their short meeting, and then I spent hours discussing technical details with sales support engineers...) From zmailer-owner@nic.funet.fi Thu Apr 24 20:37:05 1997 Received: by nic.funet.fi id <12601-24297>; Thu, 24 Apr 1997 20:34:17 +0300 Subject: Re: Another glitch with latest beta From: Matti Aarnio To: crosser@online.ru Date: Thu, 24 Apr 1997 20:34:07 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Apr 24, 97 06:41:13 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1046 Message-Id: <19970424173417Z12601-24297+230@nic.funet.fi> Return-Path: > Another thing with .48-970423: pipe works `strange' in the mailbox transport. > I have an entry in /etc/zmailer/db/aliases: > > mailping: "|exec /usr/local/bin/mailping" > > and it used to work, but now the script *is* executed, but the message > is not removed from the queue and the following is in the syslog: > > Apr 24 18:39:12 chronos mailbox[15922]: S.nLr/p68048: to="|exec > /usr/local/bin/mailping", delay=00:00:27, xdelay=00:00:00, mailer=local, > stat=deferred message write to ""|exec /usr/local/bin/mailping"" failed: Error > 0 I am not exactly sure there was a problem, but nevertheless I have recoded parts of the mailbox diagnostics routines due to these reports. The "nobodypipe" alias works now ok at Linux, but constantly fails at HP-UX by reporting an error on the FILE* stream. That is, the ferror(fp) yields non-zero for pipe-writes. There the fp has been created with statement: fp = fdopen(writefd, "w"); Hmm... Fails on Solaris 2.5.1 too. Same report !? > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 24 21:00:37 1997 Received: from NO-IDENT-SERVICE@antares.utu.fi (port 51628 [130.232.1.2]) by nic.funet.fi with ESMTP id <13134-24419>; Thu, 24 Apr 1997 20:56:49 +0300 Received: by utu.fi id <34276-1343>; Thu, 24 Apr 1997 20:55:51 +0300 Subject: Re: Another glitch with latest beta In-Reply-To: <19970424173417Z12601-24297+230@nic.funet.fi> from Matti Aarnio at "Apr 24, 97 08:34:07 pm" From: Matti Aarnio To: mea@nic.funet.fi (Matti Aarnio) Date: Thu, 24 Apr 1997 20:55:49 +0300 (EET DST) Cc: crosser@online.ru, zmailer@nic.funet.fi X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 699 Message-Id: <19970424175551Z34276-1343+1085@utu.fi> Return-Path: > > Apr 24 18:39:12 chronos mailbox[15922]: S.nLr/p68048: to="|exec > > /usr/local/bin/mailping", delay=00:00:27, xdelay=00:00:00, mailer=local, > > stat=deferred message write to ""|exec /usr/local/bin/mailping"" failed: Error > > 0 ... > The "nobodypipe" alias works now ok at Linux, but constantly > fails at HP-UX by reporting an error on the FILE* stream. ... > Hmm... Fails on Solaris 2.5.1 too. Same report !? Found it! Urgle... I am blind to my own code at the times. It didn't like to do fseek() on a pipe, and did set ferror() flag on it... Now it recognizes file being a pipe, and won't try to be clever with ending newlines. > > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 24 21:23:31 1997 Received: by nic.funet.fi id <2040-24297>; Thu, 24 Apr 1997 21:21:06 +0300 Subject: Re: new zmailer beta From: Matti Aarnio To: goldt@informatik.tu-muenchen.de (Sven Goldt) Date: Thu, 24 Apr 1997 21:21:01 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Sven Goldt" at Apr 24, 97 12:15:48 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 568 Message-Id: <19970424182106Z2040-24297+246@nic.funet.fi> Return-Path: > Hello there, > > the new beta version of zmailer doesn't even compile on HP-UX. > Isn't there someone who runs zmailer on HP-UX ? I still can't find > a version that works flawlessly on HP-UX. > This time the problem is in bsdbtree.c where nowhere > BTREEINFO is defined. Unless your host has available, you should not get that compiled in at all. While comparing yesterdays snapshot, and the release (yes!) I did notice there file 'config.cache', which may have been the cause for your problems.. > Greets, Sven /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Apr 24 21:31:16 1997 Received: by nic.funet.fi id <13254-24419>; Thu, 24 Apr 1997 21:27:47 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.48 - the release! Message-Id: <19970424182747Z13254-24419+238@nic.funet.fi> Date: Thu, 24 Apr 1997 21:27:39 +0300 Return-Path: Ok, finally :-) On the last stretch of work, I did add runtime features to allow local definition of what the locally used DEFCHARSET= is (into zmailer.conf file), I also got the beast compiling, and running properly at HP-UX 10.20. I added several input string length limiting format strings at various printf()s, and at the syslog() statements. The package is available at the usual location: ftp://ftp.funet.fi/pub/unix/mail/zmailer/ plus also at my workstation, though why should you access it, when the FUNET server has a lot faster connection :) /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 25 01:17:10 1997 Received: by nic.funet.fi id <1399-16499>; Fri, 25 Apr 1997 01:15:37 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.48 patch1 -- proto/cf/rrouter.cf regexp bug Message-Id: <19970424221537Z1399-16499+8@nic.funet.fi> Date: Fri, 25 Apr 1997 01:15:27 +0300 Return-Path: If you install the router scripts from 2.99.48, you will need the "patch1". :-/ Not perfect... /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 25 02:32:42 1997 Received: from IDENT-NONSENSE@claven.cse.psu.edu (port 56499 [130.203.3.50]) by nic.funet.fi with ESMTP id <12946-15564>; Fri, 25 Apr 1997 02:30:14 +0300 Received: from jovial.cse.psu.edu (groenvel@jovial.cse.psu.edu [130.203.10.24]) by cse.psu.edu (8.8.5/8.7.3) with ESMTP id TAA01462 for ; Thu, 24 Apr 1997 19:27:56 -0400 (EDT) Message-Id: <199704242327.TAA01462@cse.psu.edu> X-Mailer: exmh version 1.6.9 8/22/96 X-uri: X-work-address: Manufacturing Technology - Systems Development Group PO Box 30 The Applied Research Lab State College, PA 16804 X-work-phone: +1 814 863 9896 (Voice) +1 814 863 1183 (Fax) To: zmailer@nic.funet.fi Subject: 2.99.48 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 24 Apr 1997 19:27:49 EDT From: John D Groenveld Return-Path: Solaris 2.5, cc: SC4.0 18 Oct 1995 C 4.0 cc -g -DHAVE_CONFIG_H -I../.. -I../../include -c smtp.c "smtp.c", line 2177: operands have incompatible types: pointer to uchar ":" pointer to char "smtp.c", line 3544: cannot recover from previous errors cc: acomp failed for smtp.c *** Error code 2 make: Fatal error: Command failed for target `smtp.o' Current working directory /home/elvis/jdg117/zmailer-2.99.48/transports/smtp From zmailer-owner@nic.funet.fi Fri Apr 25 11:43:09 1997 Received: by nic.funet.fi id <12517-15564>; Fri, 25 Apr 1997 11:42:01 +0300 Subject: Re: 2.99.48 From: Matti Aarnio To: groenvel@cse.psu.edu (John D Groenveld) Date: Fri, 25 Apr 1997 11:41:59 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199704242327.TAA01462@cse.psu.edu> from "John D Groenveld" at Apr 24, 97 07:27:49 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 856 Message-Id: <19970425084201Z12517-15564+180@nic.funet.fi> Return-Path: > Solaris 2.5, cc: SC4.0 18 Oct 1995 C 4.0 > cc -g -DHAVE_CONFIG_H -I../.. -I../../include -c smtp.c > "smtp.c", line 2177: operands have incompatible types: > pointer to uchar ":" pointer to char > "smtp.c", line 3544: cannot recover from previous errors > cc: acomp failed for smtp.c Ok, this should fix it. This is also appended into the "patch1" -file. /Matti Aarnio --- zmailer-2.99.48/transports/smtp/smtp.c~ Thu Apr 24 18:45:17 1997 +++ zmailer-2.99.48/transports/smtp/smtp.c Fri Apr 25 11:38:39 1997 @@ -2174,7 +2174,7 @@ else fprintf(SS->verboselog, " rc=%d, mxh[0].host=%.200s (host=%.200s) mxcnt=%d\n", - rc, (mxh[0].host == NULL) ? mxh[0].host : "", + rc, (! mxh[0].host) ? (char*)mxh[0].host : "", host, mxcnt); hp = NULL; /* Ruined 'hp' datas within getmxrr() */ switch (rc) { From zmailer-owner@nic.funet.fi Fri Apr 25 12:34:05 1997 Received: from TIMEDOUT@gaston.pivoine.septgie.fr (port 25224 [194.199.139.2]) by nic.funet.fi with ESMTP id <689-4375>; Fri, 25 Apr 1997 12:33:04 +0300 Received: from Bad.EHLO.Input ([192.190.131.68]) by gaston.pivoine.septgie.fr with ESMTP id <313447-18054>; Fri, 25 Apr 1997 11:32:11 +0200 Message-ID: <33607AC9.77F5F81B@gaston.pivoine.septgie.fr> Date: Fri, 25 Apr 1997 11:35:05 +0200 From: Jean-Luc Tardivel X-Mailer: Mozilla 4.0b3 [en] (Win95; I) MIME-Version: 1.0 To: zmailer@nic.funet.fi Subject: Mail to News GateWay X-Priority: 3 (Normal) Content-Type: text/plain; charset=iso-8859-1 Return-Path: Wich program would you suggest to make a mail to news gateway whith zmailer. thanks. From zmailer-owner@nic.funet.fi Fri Apr 25 13:40:51 1997 Received: by nic.funet.fi id <769-16499>; Fri, 25 Apr 1997 13:39:35 +0300 Subject: Re: Mail to News GateWay From: Matti Aarnio To: tardivel@gaston.pivoine.septgie.fr (Jean-Luc Tardivel) Date: Fri, 25 Apr 1997 13:39:25 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <33607AC9.77F5F81B@gaston.pivoine.septgie.fr> from "Jean-Luc Tardivel" at Apr 25, 97 11:35:05 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 500 Message-Id: <19970425103935Z769-16499+225@nic.funet.fi> Return-Path: > Wich program would you suggest to make a mail to news gateway whith > zmailer. > thanks. It depends... There is one bundled way to do it, but I am not fully convinced that it is always a good way. See file utils/usenet/usenet.sh in the source. It is intended to run on channel 'usenet'. It creates one way gateway: email -> usenet There are also couple other scripts which may (or may not) be worth looking. All of them under the 'utils/' subdir. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Apr 25 13:41:25 1997 Received: from root@[131.159.0.81] (port 28107 [131.159.0.81]) by nic.funet.fi with ESMTP id <819-15564>; Fri, 25 Apr 1997 13:40:22 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109546-3565>; Fri, 25 Apr 1997 12:40:09 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <18949-216>; Fri, 25 Apr 1997 12:39:57 +0200 Date: Fri, 25 Apr 1997 12:39:45 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt To: zmailer@nic.funet.fi Subject: zmailer-2.99.48 works !!! Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: -----BEGIN PGP SIGNED MESSAGE----- Hello out there ! I just wanted to tell you all: go and grab zmailer-2.99.48 ! This version really works (yes, also with HP-UX !) and even seems to be faster than the previous ones ! (*) Good work, Matti ! The force has been with you :) Greets, Sven (*) there have been around 100 mails on a test machine and before i could stop zmailer these mails have been sent out so some people might wonder why they get so old mail :) - -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: latin1 iQCVAwUBM2CJ9UgFYR9zZnaVAQFb8wP+O632cYpoN1yEHYGsVViTvxSRKVOR0bqx g948F8bv4Cpjbp6SwjFE991kmyubRNiA5S37P1Nn0MWKelEcqxyWE9U69HGu3Ekm EyDegGopxFH/vfklTRt+wV4HG8NZaGhmq2IzlPU52WXuccSHU+nQz4Ccc/NTVLud 02+qNGux9UQ= =zzEk -----END PGP SIGNATURE----- From zmailer-owner@nic.funet.fi Fri Apr 25 15:31:35 1997 Received: from root@cal026031.student.utwente.nl (port 9479 [130.89.228.161]) by nic.funet.fi with ESMTP id <12287-14034>; Fri, 25 Apr 1997 15:29:53 +0300 Received: from lists@cal026031.student.utwente.nl (port 4359 [130.89.228.161]) by cal026031.student.utwente.nl with SMTP id <156817-24581>; Fri, 25 Apr 1997 14:29:25 +0000 Date: Fri, 25 Apr 1997 14:29:11 +0200 (MET DST) From: Mark Visser X-Sender: lists@Cal026031.student.utwente.nl Reply-To: mark@cal026031.student.utwente.nl To: zmailer@nic.funet.fi Subject: still a small error in .48 ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi, While running .44 to .48 i always noted the same: expn majordomo 250-Usage: login2uid login 250-Usage: fullname key 250-Usage: recase [ -u | -l | -p ] string 250-Usage: fullname key 550 no such user: <"|/usr/local/majordomo-1.94.1/wrapper majordomo"> quit 221 2.0.0 cal026031.student.utwente.nl Out Connection closed by foreign host. This happens to pipes. I noted there were some patches, but pipes still don't work well. The following solved it for me (by editing rrouter.cf): change -- /.+ # file # -- well, it could be a slash-notated X.400 address too.. return (((local "file.$origaddr" "$address" $A))) ;; \|.+ # pipe return (((local "pipe.$origaddr" "$address" $A))) ;; -- in -- ### Ranzige hack om procmail te ontwijken ?? (filepipe was local) [/|].+ # file or pipe return (((filepipe "$origaddr" "$address" $A))) ;; -- Then everything works fine (of course you will have to define filepipe als a channel in scheduler.conf). As local channel i'm using procmail delivery. 2.99.48 is already better here..mail to nonexistent adresses now come back to you (with cc to postoffice) where in .46 it was just dumped to /dev/null or so...you never saw an errormail come back...scheduler dumped it. Someone with good ideas ? Is it procmail perhaps ? Mark ------------------------------------------------------------------------- Mark Visser | Student Civiele Technologie & Management Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl Telephone: 053-4895038 | ------------------------------------------------------------------------- Warning: You can get rid of all the bugs by disabling them from the main menu. From zmailer-owner@nic.funet.fi Fri Apr 25 21:24:18 1997 Received: by nic.funet.fi id <591-14034>; Fri, 25 Apr 1997 21:23:06 +0300 Subject: Re: still a small error in .48 ? From: Matti Aarnio To: mark@cal026031.student.utwente.nl Date: Fri, 25 Apr 1997 21:23:00 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Mark Visser" at Apr 25, 97 02:29:11 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 4090 Message-Id: <19970425182306Z591-14034+369@nic.funet.fi> Return-Path: > Hi, > > While running .44 to .48 i always noted the same: > > expn majordomo > 250-Usage: login2uid login > 250-Usage: fullname key > 250-Usage: recase [ -u | -l | -p ] string > 250-Usage: fullname key > 550 no such user: <"|/usr/local/majordomo-1.94.1/wrapper majordomo"> > quit > 221 2.0.0 cal026031.student.utwente.nl Out > Connection closed by foreign host. > > This happens to pipes. I noted there were some patches, but pipes still > don't work well. > > The following solved it for me (by editing rrouter.cf): Nope, wrong location. Problems existed at two locations: - aliases.cf - server.cf Proper patches below. Also appended to the "patch1". (tomorrow starts "patch2", if need be..) Here are some of my tests: EXPN "|pipe" 250 local delivery for <|pipe> EXPN "|pipe space" 250 local delivery for <"|pipe space"> EXPN mea 250 Matti Aarnio EXPN nobodypipe 250 local delivery for <"| cat > /dev/null"> EXPN nobody 250 local delivery for EXPN test-domo 250 local delivery for <"|/usr/local/majordomo-1.94.1/wrapper majordomo"> EXPN no-such-user 550 no such user: Do note that the users can write ANYTHING they want into that parameter, and the system may claim the delivery to succeed, while it very likely will fault it -- because of privilege violation, or some such. The VRFY is even shorter in its flow-thru. It perhaps should be able to tell that some address is actually a list, or some user, but now it won't do that. ... > Then everything works fine (of course you will have to define filepipe als > a channel in scheduler.conf). > > As local channel i'm using procmail delivery. > 2.99.48 is already better here..mail to nonexistent adresses now come back > to you (with cc to postoffice) where in .46 it was just dumped to > /dev/null or so...you never saw an errormail come back...scheduler dumped > it. > > Someone with good ideas ? > Is it procmail perhaps ? You lost me. What is "it" above ? > Mark > ------------------------------------------------------------------------- > Mark Visser | Student Civiele Technologie & Management > Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl > 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl > Telephone: 053-4895038 | /Matti Aarnio --- zmailer-2.99.48/proto/cf/aliases.cf~ Tue Feb 18 10:50:05 1997 +++ zmailer-2.99.48/proto/cf/aliases.cf Fri Apr 25 19:56:47 1997 @@ -359,7 +359,7 @@ if [ -n "$pobox" ]; then db add expansions "$key" pobox - priv=$(login2uid $user) + priv=$(login2uid "$user") return (((local "pob:$user" "$pobox$plustail$domain" $(newattribute $attr privilege $priv)))) fi --- zmailer-2.99.48/proto/cf/server.cf~ Wed Jan 3 10:14:16 1996 +++ zmailer-2.99.48/proto/cf/server.cf Fri Apr 25 20:27:10 1997 @@ -5,24 +5,39 @@ #| This is a prettyprinter for address quads. What it prints is what someone #| doing a VRFY or EXPN query to the SMTP server will see. - local text + local text user user2 + user="$(user $quad)" case $(channel $quad) in - local) sift "$(user $quad)" in + local) ssift "$user" in [|/].* text="local delivery for" + break ;; - .* text=$(login2uid $(user $quad)) # ignore return value - if text="$(fullname $(user $quad))"; then + .* + user2="$user" + ssift "$user2" in + (.*)@([^@]+) # FQDN format address ? + user2="\1" ;; + .* ;; + tfiss + ssift "$user2" in + (.*)\+(.*) # A "+" in the name ? + user2="\1" ;; + .* ;; + tfiss + + text="$(login2uid "$user2")" # ignore return value + if text="$(fullname "$user2")"; then text="$text" else - if text="$(fullname $(recase -l $(user $quad)))"; then + if text="$(fullname "$(recase -l "$user2")")"; then text="$text" else text="550 no such user:" fi fi ;; - tfis + tfiss ;; usenet) text="newsgroup:" ;; *) text="$(channel $quad) delivery" @@ -32,7 +47,7 @@ esac ;; esac - echo "$text <$(user $quad)>" + echo "$text <$user>" } server (key) { From zmailer-owner@nic.funet.fi Fri Apr 25 23:15:28 1997 Received: from root@cal026031.student.utwente.nl (port 1286 [130.89.228.161]) by nic.funet.fi with ESMTP id <460-15564>; Fri, 25 Apr 1997 23:14:25 +0300 Received: from mark@cal026031.student.utwente.nl (port 59653 [130.89.228.161]) by cal026031.student.utwente.nl with SMTP id <156514-26320>; Fri, 25 Apr 1997 22:14:15 +0000 Date: Fri, 25 Apr 1997 22:14:07 +0200 (MET DST) From: Mark Visser To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: still a small error in .48 ? In-Reply-To: <19970425182306Z591-14034+369@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 25 Apr 1997, Matti Aarnio wrote: > > Nope, wrong location. Problems existed at two locations: > - aliases.cf > - server.cf > Proper patches below. Also appended to the "patch1". > (tomorrow starts "patch2", if need be..) Thanks...i wasn't too sure (i only run zmailer for six months now), but i tried and it worked...but i'll try this one. > 250 local delivery for <"|/usr/local/majordomo-1.94.1/wrapper majordomo"> > EXPN no-such-user > 550 no such user: Looks ok here too... > > Do note that the users can write ANYTHING they want into > that parameter, and the system may claim the delivery to > succeed, while it very likely will fault it -- because of > privilege violation, or some such. > Hmm..delivery is not succesfull here. When doing a telnet localhost 25 and expn majordomo gives me a 'correct result'. When mailing something to it, zmailer tells me: (from schedulere.perflog) 861998101 156514-26169 9 6 error local/pipe.majordomo@cal026031.student.utwente.nl (from scheduler) 19970425215516 /usr/local/zmailer/bin/ta/sm: 156514/247//error2 procmail: Unknown user ""|/usr/local/majordomo-1.94.1/wrapper majordomo""[exit status 67 (addressee unknown)] So not succesfull yet. > > > > Someone with good ideas ? > > Is it procmail perhaps ? > > You lost me. What is "it" above ? > Excuse me. I went to dinner and after that i finished the mail. I wanted to type something more here (and probably something different). What i was trying to 'tell' is that i'm using procmail as local channel. And when using: (scheduler.conf) local command="sm -8c $channel procm" it goes wrong (see above) and when using local command="mailbox -8" It's al going ok. Then everything works as it's supposed to be. So there might be a procmail problem here or there's something in zmailer that going wrong here. Mark ------------------------------------------------------------------------- Mark Visser | Student Civiele Technologie & Management Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl Telephone: 053-4895038 | ------------------------------------------------------------------------- Warning: You can get rid of all the bugs by disabling them from the main menu. From zmailer-owner@nic.funet.fi Fri Apr 25 23:52:33 1997 Received: by nic.funet.fi id <130-15564>; Fri, 25 Apr 1997 23:51:01 +0300 Subject: Re: still a small error in .48 ? From: Matti Aarnio To: mark@cal026031.student.utwente.nl (Mark Visser) Date: Fri, 25 Apr 1997 23:50:57 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Mark Visser" at Apr 25, 97 10:14:07 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2126 Message-Id: <19970425205101Z130-15564+349@nic.funet.fi> Return-Path: ... > Hmm..delivery is not succesfull here. > When doing a telnet localhost 25 and expn majordomo gives me a 'correct > result'. > When mailing something to it, zmailer tells me: > > (from schedulere.perflog) > > 861998101 156514-26169 9 6 error local/pipe.majordomo@cal026031.student.utwente.nl > > (from scheduler) > 19970425215516 /usr/local/zmailer/bin/ta/sm: 156514/247//error2 procmail: > Unknown user ""|/usr/local/majordomo-1.94.1/wrapper majordomo""[exit > status 67 (addressee unknown)] > > So not succesfull yet. Oh, I see. You are using "sm" to run local channel. The "sm" MAY need a fix.. Hmm.. Don't know, but must look. In the meantime you can do by defining following channel specs at the scheduler.conf: local/pipe.* ... params like with current local/* command="mailbox -8" local/file.* ... params like with current local/* command="mailbox -8" local/* ... the current "local/*" That way you can have both -- pipes/files and procmail. I have another project to finish by monday, so the sm may have to wait a bit for my time.. > > > Someone with good ideas ? > > > Is it procmail perhaps ? > > > > You lost me. What is "it" above ? > > > > Excuse me. I went to dinner and after that i finished the mail. I wanted > to type something more here (and probably something different). > > What i was trying to 'tell' is that i'm using procmail as local channel. > > And when using: > > (scheduler.conf) > local command="sm -8c $channel procm" > > it goes wrong (see above) > > and when using > > local command="mailbox -8" > > It's al going ok. Then everything works as it's supposed to be. > So there might be a procmail problem here or there's something in zmailer > that going wrong here. I suspect the 'sm' of ZMailer. > Mark > ------------------------------------------------------------------------- > Mark Visser | Student Civiele Technologie & Management > Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl > 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl /Matti Aarnio From zmailer-owner@nic.funet.fi Sat Apr 26 08:16:16 1997 Received: from root@wit393101.student.utwente.nl (port 3077 [130.89.235.11]) by nic.funet.fi with ESMTP id <576-359>; Sat, 26 Apr 1997 08:15:28 +0300 Received: from localhost ([127.0.0.1]) by wit393101.student.utwente.nl with SMTP id <23843-179>; Sat, 26 Apr 1997 07:15:19 +0200 Date: Sat, 26 Apr 1997 07:15:10 +0200 (MET DST) From: Florian Overkamp Reply-To: sim@wit393101.student.utwente.nl To: Mark Visser cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: still a small error in .48 ? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 25 Apr 1997, Mark Visser wrote: > (scheduler.conf) > local command="sm -8c $channel procm" > > it goes wrong (see above) > > and when using > > local command="mailbox -8" > > It's al going ok. Then everything works as it's supposed to be. > So there might be a procmail problem here or there's something in zmailer > that going wrong here. Actually, this is why I defined the filepipe channel (as Mark showed earlier) in the first place. Procmail doesn't know how to handle files or pipes as adressees. Mailbox apparently does. BTW. The modifications were based upon snippets from the zmailer archives, messages about procmail etc. Worked like charm here, (on a .44) so I didn't bother to follow up. /* Florian Overkamp */ /* PGP Key-ID 233FC67D */ /* PGP Fingerprint 00 3F F5 63 DD FB 5D B7 0A 16 0A EC F9 BD 01 97 */ From zmailer-owner@nic.funet.fi Sat Apr 26 13:04:55 1997 Received: from NO-IDENT-SERVICE@[199.4.235.2] (port 60936 [199.4.235.2]) by nic.funet.fi with SMTP id <13228-28097>; Sat, 26 Apr 1997 13:03:53 +0300 Received: (qmail 6965 invoked by uid 103); 26 Apr 1997 06:17:29 -0000 Date: 26 Apr 1997 06:17:29 -0000 Message-ID: <19970426061729.6964.qmail@ssr.com> From: Scott Ballantyne To: zmailer@nic.funet.fi Subject: Thanks... Return-Path: Recent developments have made it clear to me that after several years of digitally wedded bliss, zmailer and I are headed in different directions. I'd just like to thank all of the zmailer community and in particular Matti Aarnio for the tremendously high level of support and the unusually helpful spirit that pervades this list. Matti is still the only person who hacks on a freely distributed package that has offered to call me to help me out, all the way from Finland to the U.S.A. You don't find that kind of thing very often anymore. Even so, I'm moving on to qmail, which has the same design philosophy that the original zmailer did, and then some. Zmailer broke sendmails monolithic design into separate, cooperating processes. Qmail carries this tradition on, but also removes alias/list processing from the guts of the mailer, which is a terrific idea and carries the same power with it that the original idea of 'tools, not solutions' brought to Unix and to the original zmailer. I'm sure that a lot of the features that have been added to zmailer are very worthwhile, and that many of you appreciate them. At the sites where I work, we simply don't use them and they seem to me to be a bit too much about the 'solution', rather than the tools. So, as so often happens, philosophy and life goals have brought another parting of the ways. For those of you who have been feeling as I have about zmailer (and I know there are some, I've seen the comments), I recommend qmail to your attentions. To all of you, thanks again for the many hours you have spent answering questions and helping me and other people out. sdb -- sdb@ssr.com From zmailer-owner@nic.funet.fi Sat Apr 26 13:41:47 1997 Received: from root@iug.org (port 35423 [194.86.187.30]) by nic.funet.fi with ESMTP id <12537-28097>; Sat, 26 Apr 1997 13:41:19 +0300 Received: by the.iug.org id <20532-18057>; Sat, 26 Apr 1997 13:41:08 +0300 Subject: Re: database problem solved. In-Reply-To: <19970423144416Z12783-28309+178@nic.funet.fi> from Matti Aarnio at "Apr 23, 97 05:44:08 pm" From: Niko Lilja To: mea@nic.funet.fi (Matti Aarnio) Date: Sat, 26 Apr 1997 13:40:54 +0300 (EET DST) Cc: zmailer@nic.funet.fi MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 595 Message-Id: <19970426104108Z20532-18057+4@the.iug.org> Return-Path: > > When you're linking the binary of router and makendbm, be sure that > > you heva changed the libs order in Makefile in those sources directory to: > > > > ROUTER_LIB= -lndbm -lgdbm -ldb > > Using order of: > -lndbm -ldb -lgdbm > should work equally well, but apparently it didn't ? > Nope, I noticed that it "must" (must or must anyway) link in that order -lndbm -lgdbm -ldb, maybe the reason is that the databases has defined in that order in the source? And that's why it need to be linked in that order. > /Matti Aarnio -- (Niko Lilja ) From zmailer-owner@nic.funet.fi Mon Apr 28 11:04:28 1997 Received: from NO-IDENT-SERVICE@cesium.clock.org (port 35588 [140.174.97.8]) by nic.funet.fi with ESMTP id <13279-20954>; Mon, 28 Apr 1997 11:04:20 +0300 Received: from [198.68.110.2] ([198.68.110.2]) by cesium.clock.org with ESMTP id <119170-677>; Mon, 28 Apr 1997 01:04:05 -0800 X-Sender: eef-play@pop.mercy.st.clock.org Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 28 Apr 1997 01:03:40 -0700 To: zmailer@nic.funet.fi From: "Erik E. Fair" (Time Keeper) Subject: NetBSD & zmailer; sendmail-style aliases parsing Return-Path: I've been using zmailer for quite some time, but not keeping up with the latest releases. Since I am doing a system upgrade today, I thought I'd grab the latest stuff. To my unhappiness, I found that I'd fallen off the mailing list at the end of february for some reason. Two questions: 1. am I the only one using zmailer on NetBSD-current (sparc)? Some patches from me will be forthcoming. 2. is there an easy way to set zmailer so that it will parse both /etc/aliases and .forward files *exactly* the same way that sendmail does? The gratuitous difference here is a nearly daily exercise in frustration for me, particularly since the parsing rules are not documented in any obvious manual page. thanks, Erik Fair From zmailer-owner@nic.funet.fi Mon Apr 28 13:04:07 1997 Received: by nic.funet.fi id <13071-20954>; Mon, 28 Apr 1997 13:03:47 +0300 Subject: Re: NetBSD & zmailer; sendmail-style aliases parsing From: Matti Aarnio To: fair@cesium.clock.org (Erik E. Fair) Date: Mon, 28 Apr 1997 13:03:42 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Erik E. Fair" at Apr 28, 97 01:03:40 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 602 Message-Id: <19970428100347Z13071-20954+93@nic.funet.fi> Return-Path: ... > Two questions: ... > 2. is there an easy way to set zmailer so that it will parse both > /etc/aliases and .forward files *exactly* the same way that sendmail does? > The gratuitous difference here is a nearly daily exercise in frustration > for me, particularly since the parsing rules are not documented in any > obvious manual page. The ZMailer processes them with RFC-822 rules, which is commented at some of the README files -- but not on manual pages, that is true. Which difference exactly bothers you ? > thanks, > > Erik Fair /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Apr 28 18:58:58 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 37115 [194.67.3.135]) by nic.funet.fi with ESMTP id <13229-14323>; Mon, 28 Apr 1997 18:57:52 +0300 Received: from NO-IDENT-SERVICE@ariel (port 45287 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-27760>; Mon, 28 Apr 1997 19:57:31 +0400 Date: Mon, 28 Apr 1997 19:57:23 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: NetBSD & zmailer; sendmail-style aliases parsing To: Matti Aarnio cc: "Erik E. Fair" , zmailer@nic.funet.fi In-Reply-To: <19970428100347Z13071-20954+93@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > 2. is there an easy way to set zmailer so that it will parse both > > /etc/aliases and .forward files *exactly* the same way that sendmail > > does? The gratuitous difference here is a nearly daily exercise in > > frustration for me, particularly since the parsing rules are not > > documented in any obvious manual page. > > The ZMailer processes them with RFC-822 rules, which > is commented at some of the README files -- but not > on manual pages, that is true. > > Which difference exactly bothers you ? I would pay attention on two things: "special" aliases, as pipes, have different syntax. In Sendmail, you can write: Majordomo: |/blah/blah/wrapper params or Majordomo: |"/blah/blah/wrapper params" In Zmailer, you must write: Majordomo: "|/blah/blah/wrapper params" (the whole thing in quotes). Another difference, is "real-user" mechanizm: networker: \root will deliver to /var/mail/root even if you have "root" aliased to somebody. Actually, this design is brain-dead in Sendmail (IMHO), but in Zmailer it is completely missing. Personally I prefer a Smail3-like "real-user" scheme, a la: networker: real-root and I have a six line patch to aliases.cf to implement this. Matti, do you want it? Eugene From zmailer-owner@nic.funet.fi Mon Apr 28 21:46:56 1997 Received: from NO-IDENT-SERVICE@cesium.clock.org (port 44038 [140.174.97.8]) by nic.funet.fi with ESMTP id <13134-20954>; Mon, 28 Apr 1997 21:45:52 +0300 Received: from [198.68.110.2] ([198.68.110.2]) by cesium.clock.org with ESMTP id <119170-676> convert rfc822-to-8bit; Mon, 28 Apr 1997 11:45:26 -0800 X-Sender: eef-play@pop.mercy.st.clock.org Message-Id: In-Reply-To: References: <19970428100347Z13071-20954+93@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8BIT Date: Mon, 28 Apr 1997 11:44:54 -0700 To: Eugene Crosser From: "Erik E. Fair" (Time Keeper) Subject: Re: NetBSD & zmailer; sendmail-style aliases parsing Cc: Matti Aarnio , "Erik E. Fair" , zmailer@nic.funet.fi Return-Path: At 8:57 -0700 4/28/97, Eugene Crosser wrote: >Another difference, is "real-user" mechanizm: > > networker: \root > >will deliver to /var/mail/root even if you have "root" aliased to somebody. >Actually, this design is brain-dead in Sendmail (IMHO), but in Zmailer it >is completely missing. Personally I prefer a Smail3-like "real-user" >scheme, a la: > > networker: real-root > >and I have a six line patch to aliases.cf to implement this. Matti, do you >want it? > >Eugene This is my principal complaint. This is not an 822 issue; this is a local-part parsing issue, for a syntax that never leaves the host. What you want is a syntax that will never show up in a real E-mail address (i.e. one transmitted over the wire in SMTP), and never show up in a real user name (from /etc/passwd or whatever your user name database is for your OS). Sendmail's choice is unfortunate for its collision with one of RFC822's quoting mechanisms, however I have never seen it cause a problem. I don't like Eugene's (smail) suggestion because that potentially collides with real (ahem!) user names (though I'm glad someone other than me has seen this as a problem and worked on it some!). I really want to be able to support: \fair, fair@foo.com in my .forward files, and in /etc/aliases (for those who do not know, this is sendmail's syntax for saying "leave one copy here in my system mailbox, and send another copy to fair@foo.com" (It's also possible to put a pipe/program delivery, or a direct-file delivery here)). Sendmail's rules are explained in an "aliases(5)" manual page that comes with the distribution. It would be a really good idea for there to be a similar page for zmailer (potentially one for each different aliases.cf file that one could use). Erik Fair From zmailer-owner@nic.funet.fi Tue Apr 29 14:43:14 1997 Received: from TIMEDOUT@falco.its.uu.se (port 16 [130.238.131.113]) by nic.funet.fi with ESMTP id <289-30738>; Tue, 29 Apr 1997 14:42:56 +0300 Received: from alba.its.uu.se ([130.238.131.102]) by falco.its.uu.se with ESMTP id <2059-19782> convert rfc822-to-8bit; Tue, 29 Apr 1997 13:42:38 +0100 X-Mailer: exmh version 2.0gamma 1/27/96 To: Matti Aarnio cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se Subject: Bug in sm truncates messages on AIX (New bug) In-reply-to: Your message of "Tue, 15 Apr 1997 17:56:41 +0300." <19970415145645Z19652-12358+22@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8BIT Date: Tue, 29 Apr 1997 13:42:35 +0200 From: Martin Wendel Message-Id: <97Apr29.134238+0100msz.2059-19782+9@falco.its.uu.se> Return-Path: OS: AIX 4.1.5 zmailer version: 2.99.48 patch1 added. description: A bug in sm cleares some message bodies to size zero. detail: In sm.c appendlet(): /* we are assumed to be positioned properly at start of message body */ This is not always true. In certain cases (I've been able to reproduce but haven't checked the cause) mfd is positioned way outside the file's range. When this happens no body can be read and thus no body written to the delivery agent either. Kludge: Check that mfd file position equals dp->msgbodyoffset, if not make new lseek with dp->msgbodyoffset and SEEK_SET: Example from log (extra logging added): Apr 29 13:27:31 uria sm[42818]: 18603-12326 stat size=2145, Apr 29 13:27:31 uria sm[42818]: 18603-12326 seek offset=536920728, msgbodyoffset=919 Apr 29 13:27:31 uria sm[42818]: 18603-12326 lseek SEEK_SET set to 919 Apr 29 13:27:31 uria sm[42818]: 18603-12326 read i=6, errno=0, bodyoffset=919, sizeestimate=2067 Apr 29 13:27:31 uria sm[42818]: 18603-12326 size=1226, convertmode=1, let_buffer=32768 # grep "seek offset" maillog Apr 29 13:14:05 uria sm[22312]: 18535-8876 seek offset=536921688, msgbodyoffset=1056 Apr 29 13:14:21 uria sm[13614]: 18502-30998 seek offset=536921496, msgbodyoffset=968 Apr 29 13:14:23 uria sm[22312]: 18535-8876 seek offset=536921688, msgbodyoffset=1056 Apr 29 13:14:51 uria sm[22312]: 12515-14868 seek offset=536921688, msgbodyoffset=718 Apr 29 13:15:03 uria sm[22312]: 18535-8876 seek offset=536999648, msgbodyoffset=1056 Apr 29 13:15:21 uria sm[22312]: 16610-10256 seek offset=536997488, msgbodyoffset=852 Apr 29 13:17:21 uria sm[23112]: 18502-14868 seek offset=536920668, msgbodyoffset=1559 -- E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS Tel: 018 18 77 80 Box 887 Int: +46 18 18 77 80 S-751 08 Uppsala Fax: +46 18 18 79 10 Sweden http://www.uu.se/Internservice/ue/personal/MartinWendel.html From zmailer-owner@nic.funet.fi Tue Apr 29 23:07:53 1997 Received: from NO-IDENT-SERVICE@mailrelay.tiac.net (port 21521 [199.0.65.237]) by nic.funet.fi with ESMTP id <12666-2282>; Tue, 29 Apr 1997 23:06:53 +0300 Received: from draco ([206.119.34.99]) by mailrelay.tiac.net (8.8.5/) with SMTP id QAA13435 for ; Tue, 29 Apr 1997 16:07:40 -0400 (EDT) Message-Id: <199704292007.QAA13435@mailrelay.tiac.net> Comments: Authenticated sender is From: "Gerard Hickey" Organization: Nekotech, A Division of Inco, Inc. To: zmailer@nic.funet.fi Date: Tue, 29 Apr 1997 16:06:46 -5 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Rewriting of From: and Reply-To: headers Reply-to: hickey@alphapower.com Priority: normal X-mailer: Pegasus Mail for Win32 (v2.42a) Return-Path: I have finally gotten zmailer-2.99.45beta0 compiled on Linux/AXP! Yes, I have just found out that zmailer is now up to 2.99.48. I have just begun to work on getting it to compile (hopefully with ease considering that I have just gone through the entire process!). Zmailer is working flawlessly except for one minor problem. When I send a message through (i.e. point my PC mail program to the Linux machine) it rewrites the From: and Reply-To: headers. So instead of having headers that look like 'hickey@alphapower.com' I end up getting 'hickey@bucky'. Unfortunately the hickey@bucky address does not work too well outside of my domain. Does anyone have any ideas what is going on? Is it something that I did on the install? Is there a change I can make to a config file (I have been reading through them trying to find where this is happening)? Is this a known problem with 2.99.45 and will be fixed by getting 2.99.48 installed? Thanks.... -- Gerard Hickey hickey@alphapower.com Nekotech Technical Services Voice: +1 603 926 0300 102 Tide Mill Rd. Fax: +1 603 926 0301 Hampton, NH 03842 Nekotech is a provider of high end servers based on the Digital Alpha chipset. Nekotech's home page can be reached at http://www.alphapower.com/ From zmailer-owner@nic.funet.fi Wed Apr 30 00:08:03 1997 Received: from NO-IDENT-SERVICE@mailrelay.tiac.net (port 57610 [199.0.65.237]) by nic.funet.fi with ESMTP id <12404-20111>; Wed, 30 Apr 1997 00:06:32 +0300 Received: from draco ([206.119.34.99]) by mailrelay.tiac.net (8.8.5/) with SMTP id RAA21787; Tue, 29 Apr 1997 17:07:08 -0400 (EDT) Message-Id: <199704292107.RAA21787@mailrelay.tiac.net> Comments: Authenticated sender is From: "Gerard Hickey" Organization: Nekotech, A Division of Inco, Inc. To: Roy C Bixler Date: Tue, 29 Apr 1997 17:06:14 -5 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: Rewriting of From: and Reply-To: headers Reply-to: hickey@alphapower.com CC: zmailer@nic.funet.fi Priority: normal X-mailer: Pegasus Mail for Win32 (v2.42a) Return-Path: > Is your /etc/mail.conf file correct? In our case, it looks like this: I think so. Below is the current mail.conf file. orgdomain=alphapower.com orgdomains=nekoinc.com,incoinc.com,alphapower.com hostname=bucky.$orgdomain mydomain=$orgdomain There was something I was reading in the configuration scripts that made me add the hostname line. I just can not remember it now. Thanks for making me look at this file. I just remembered that I have to correct the orgdomains line (the additional domains are not being used yet). Any thoughts? -- Gerard Hickey hickey@alphapower.com Nekotech Technical Services Voice: +1 603 926 0300 102 Tide Mill Rd. Fax: +1 603 926 0301 Hampton, NH 03842 Nekotech is a provider of high end servers based on the Digital Alpha chipset. Nekotech's home page can be reached at http://www.alphapower.com/ From zmailer-owner@nic.funet.fi Wed Apr 30 10:27:27 1997 Received: by nic.funet.fi id <13086-16035>; Wed, 30 Apr 1997 10:26:48 +0300 Subject: Re: Rewriting of From: and Reply-To: headers From: Matti Aarnio To: hickey@alphapower.com Date: Wed, 30 Apr 1997 10:26:45 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199704292007.QAA13435@mailrelay.tiac.net> from "Gerard Hickey" at Apr 29, 97 04:06:46 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1694 Message-Id: <19970430072648Z13086-16035+547@nic.funet.fi> Return-Path: > I have finally gotten zmailer-2.99.45beta0 compiled on Linux/AXP! > Yes, I have just found out that zmailer is now up to 2.99.48. I have > just begun to work on getting it to compile (hopefully with ease > considering that I have just gone through the entire process!). You did read the INSTALL file, didn't you ? > Zmailer is working flawlessly except for one minor problem. When I > send a message through (i.e. point my PC mail program to the Linux > machine) it rewrites the From: and Reply-To: headers. So instead of > having headers that look like 'hickey@alphapower.com' I end up > getting 'hickey@bucky'. Unfortunately the hickey@bucky address does > not work too well outside of my domain. Check your /etc/hosts I have a feeling it tells something of style: 1.2.3.4 bucky bucky.alphapower.com alphapower.com This however should not matter, as the "localnames" mapping has been around for quite a while. MAILVAR/db/localnames: alphapower.com alphapower.com bucky alphapower.com ... unless you have it coded as: MAILVAR/db/localnames: alphapower.com bucky ... in which case the 'bucky' will not be mapped to the desired presentation name ... > Does anyone have any ideas what is going on? Is it something that I > did on the install? Is there a change I can make to a config file (I > have been reading through them trying to find where this is > happening)? Is this a known problem with 2.99.45 and will be fixed by > getting 2.99.48 installed? Umm.. No, it is not known problem, but you CAN create it with careless configuring... > Thanks.... > -- > Gerard Hickey hickey@alphapower.com /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 2 02:11:50 1997 Received: from NO-IDENT-SERVICE@mariner.interdynamix.com (port 27340 [206.75.211.66]) by nic.funet.fi with ESMTP id <12760-12134>; Fri, 2 May 1997 02:11:35 +0300 Received: from pongo.interdynamix.com ([192.168.1.3]) by mariner.interdynamix.com with SMTP id <495568-224>; Thu, 1 May 1997 17:15:47 -0600 Message-ID: <33692339.1BF@interdynamix.com> Date: Thu, 01 May 1997 17:11:53 -0600 From: Dean Michaels Organization: InterDynamiX Systems Corp X-Mailer: Mozilla 3.01 (Win95; I) MIME-Version: 1.0 To: zmailer@nic.funet.fi Subject: Problems with quoted email addresses. Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: If incoming mail is addressed to <" username"@localdomain.net>, and username (without the leading space) is a valid ID, the message passes through and ends up in a file called '/var/mail/ username'. It ends up that we have about 10-15 files in our /var/mail that all begin with space characters. Any quick ideas on how to prevent this? With the quotes, <" username"@localdomain.net> should not be a valid mailbox. example follows. I'm running a very stock version of .cf files. Also note the errors after the RCPT TO: line in the example. /var/mail> telnet localhost smtp Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. 220 localhost.domain.net ZMailer Server 2.99.48 #1 ESMTP ready at Thu, 1 May 1997 17:01:43 -0600 HELO localhost.domain.net 250 localhost.domain.net Hello localhost.domain.net MAIL FROM: <> 250 2.1.0 Sender syntax Ok RCPT TO: <" dmichael"@domain.net> 250-[: argument expected 250-[: syntax error at: 250 Ok (verified) Ok DATA 354 Start mail input; end with . Subject: test asdf asdf . 250 2.6.0 S.nOG2I20638 message accepted QUIT 221 2.0.0 localhost.domain.net Out Connection closed by foreign host. /var/mail> ls -l *michael -rw------- 1 dmichael staff 807 May 1 17:02 dmichael -rw------- 1 dmichael staff 417 May 1 14:03 dmichael /var/mail> ^^^ note the extra space Anyone have a quick fix? Dean. From zmailer-owner@nic.funet.fi Fri May 2 12:44:05 1997 Received: by nic.funet.fi id <12729-15158>; Fri, 2 May 1997 12:43:55 +0300 Subject: Re: Problems with quoted email addresses. From: Matti Aarnio To: dean@interdynamix.com (Dean Michaels) Date: Fri, 2 May 1997 12:43:51 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <33692339.1BF@interdynamix.com> from "Dean Michaels" at May 1, 97 05:11:53 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2795 Message-Id: <19970502094355Z12729-15158+279@nic.funet.fi> Return-Path: > If incoming mail is addressed to <" username"@localdomain.net>, > and username (without the leading space) is a valid ID, the > message passes through and ends up in a file > called '/var/mail/ username'. > > It ends up that we have about 10-15 files in our /var/mail > that all begin with space characters. Any quick ideas on how > to prevent this? I think this depends of your systems getpwnam() behaviour. I get: User ' mea' unknown Try this small program, and tell what happens. (But do change the USER value at first..) ------- pwtest.c --------- #include #include #define USER " mea" main() { struct passwd *pw = getpwnam(USER); printf ("getpwnam(\"%s\") -> ",USER); if (pw == NULL) printf("USER UNKNOWN\n"); else printf("user='%s', uid=%d\n", pw->pw_name, pw->pw_uid); return 0; } ------------------------- > With the quotes, <" username"@localdomain.net> should not be > a valid mailbox. > > example follows. I'm running a very stock version of .cf files. > Also note the errors after the RCPT TO: line in the example. Oops.. This is more or less cut&paste from the xterm, so TABs are broken, but it should not matter.. Ok, I will collect that, and the SM message truncation, and whatnot else into "patch2" package today. --- ~/src/zmailer-2.99.48/proto/cf/fqdnalias.cf Mon Feb 17 23:09:16 1997 +++ $MAILSHARE/cf/fqdnalias.cf Fri May 2 11:32:54 1997 @@ -94,7 +94,7 @@ a="$MAILVAR/fqlists/$lcaddress" l="$preowner$user$postowner@$host" nattr=() - [ -f $a ] && + [ -f "$a" ] && db add expansions "$lcaddress" fqlist && priv=$(getpriv "644" $priv "$a" maillist) && nattr=$(newattribute $A privilege $priv sender "$l") && > /var/mail> telnet localhost smtp > Trying 127.0.0.1... > Connected to localhost. > Escape character is '^]'. > 220 localhost.domain.net ZMailer Server 2.99.48 #1 ESMTP ready at Thu, 1 > May 1997 17:01:43 -0600 > HELO localhost.domain.net > 250 localhost.domain.net Hello localhost.domain.net > MAIL FROM: <> > 250 2.1.0 Sender syntax Ok > RCPT TO: <" dmichael"@domain.net> > 250-[: argument expected > 250-[: syntax error at: > 250 Ok (verified) Ok > DATA > 354 Start mail input; end with . > Subject: test > > asdf > asdf > . > 250 2.6.0 S.nOG2I20638 message accepted > QUIT > 221 2.0.0 localhost.domain.net Out > Connection closed by foreign host. > /var/mail> ls -l *michael > -rw------- 1 dmichael staff 807 May 1 17:02 dmichael > -rw------- 1 dmichael staff 417 May 1 14:03 dmichael > /var/mail> > ^^^ > note the extra space > Anyone have a quick fix? > > Dean. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 2 14:01:37 1997 Received: from root@apathy.lightweb.net (port 57120 [206.54.187.185]) by nic.funet.fi with ESMTP id <12519-7278>; Fri, 2 May 1997 14:01:12 +0300 Received: by apathy.lightweb.net id <4132-58>; Fri, 2 May 1997 05:56:53 -0500 Message-ID: <19970502055643.03895@apathy.lightweb.net> Date: Fri, 2 May 1997 05:56:43 -0500 Sender: Jeff Harris From: Jeff Harris To: zmailer@nic.funet.fi Subject: Problems (.48) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.64 Return-Path: This may seem like a really, really stupid question, but I've just installed zmailer 2.99.48 on a Solaris 2.5.1 machine (Ultra 1), and it's giving me (at router -i): z# router zmailer@nic.funet.fi : address: zmailer@nic.funet.fi newsgroup: not found (((error norealname zmailer@nic.funet.fi default_attributes))) then root gets the 'real.name' response for some unknown reason. I've been using zmailer on another linux machine and it's worked fine, but yet maybe I've broken something on this Solaris machine when trying to configure it. Does anyone have any idea of where to look? I've poured over all the 'stock' aliases, scheduler.cf, etc, etc, etc... I'm using SMTP+UUCP.cf as my router.cf, all the pretty much standard stuff.. I may be missing a binary or something though as make install broke at one point (but I fixed most of it by hand, or so I think)... BTW, the DNS entries for this host (nettalk.nettalklive.com) aren't set yet but it's a valid IP and hostname (nettalk). I don't think that should have anything to do with it though (/etc/mail.conf is setup).. Matti: do you have a list all of binaries/configuration files that would be produced on, say, a stock zmailer install? I just need to see if I'm missing something important.. :) ie. /usr/bin/zmailer /usr/bin/newaliases /usr/lib/mail/router.cf etc...etc...etc... Thanks :) -- . `$, . `$,$' Jeff Harris - Network Administrator - NetTalk Live! .$' light@lightweb.net -/- http://apathy.lightweb.net .$'$. --------- ------ ----- ---- ---- --- - ' `$. ` From zmailer-owner@nic.funet.fi Fri May 2 15:30:42 1997 Received: by nic.funet.fi id <13011-12134>; Fri, 2 May 1997 15:30:05 +0300 Subject: Re: Bug in sm truncates messages on AIX (New bug) From: Matti Aarnio To: martin.wendel@its.uu.se (Martin Wendel) Date: Fri, 2 May 1997 15:29:59 +0300 (EET DST) Cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se In-Reply-To: <97Apr29.134238+0100msz.2059-19782+9@falco.its.uu.se> from "Martin Wendel" at Apr 29, 97 01:42:35 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 995 Message-Id: <19970502123005Z13011-12134+355@nic.funet.fi> Return-Path: > > OS: AIX 4.1.5 > zmailer version: 2.99.48 patch1 added. > > description: > A bug in sm cleares some message bodies to size zero. > > detail: > In sm.c appendlet(): > > /* we are assumed to be positioned properly at start of message body */ > > This is not always true. In certain cases (I've been able to reproduce but > haven't checked the cause) mfd is positioned way outside the file's range. > When this happens no body can be read and thus no body written to the delivery > agent either. Right. Now I have replaced it with: /* We make sure we are positioned properly ... */ (and doing lseek()) > Kludge: > > Check that mfd file position equals dp->msgbodyoffset, if not make new > lseek with dp->msgbodyoffset and SEEK_SET: With that you do two lseek() calls -- the first to find out where the IO cursor is, and the second to set it... It is easier just to set it. .... > -- > E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 2 18:27:45 1997 Received: from NO-IDENT-SERVICE@mariner.interdynamix.com (port 62418 [206.75.211.66]) by nic.funet.fi with ESMTP id <13204-26536>; Fri, 2 May 1997 18:26:50 +0300 Received: from pongo.interdynamix.com ([192.168.1.3]) by mariner.interdynamix.com with SMTP id <495568-224>; Fri, 2 May 1997 09:31:06 -0600 Message-ID: <336A07CF.4C1F@interdynamix.com> Date: Fri, 02 May 1997 09:27:11 -0600 From: Dean Michaels Organization: InterDynamiX Systems Corp X-Mailer: Mozilla 3.01 (Win95; I) MIME-Version: 1.0 To: Zmailer List Subject: Re: Problems with quoted email addresses. References: <19970502094355Z12729-15158+279@nic.funet.fi> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: > ------- pwtest.c --------- > #include > #include > #define USER " dmichael" > main() > { > struct passwd *pw = getpwnam(USER); > printf ("getpwnam(\"%s\") -> ",USER); > if (pw == NULL) > printf("USER UNKNOWN\n"); > else > printf("user='%s', uid=%d\n", pw->pw_name, pw->pw_uid); > return 0; > } Produces the following under Solaris 2.5.1 /Zmailer> make pwtest cc -o pwtest pwtest.c /Zmailer> ./pwtest getpwnam(" dmichael") -> user='dmichael', uid=1001 /Zmailer> What do you suggest, a wrapper for getpwnam()? Dean. From zmailer-owner@nic.funet.fi Fri May 2 18:56:19 1997 Received: from root@tshimshan.phys.ualberta.ca (port 46097 [129.128.7.76]) by nic.funet.fi with ESMTP id <12953-26536>; Fri, 2 May 1997 18:55:24 +0300 Received: by tshimshan.Phys.UAlberta.Ca id <8047-2>; Fri, 2 May 1997 09:55:09 -0600 Date: Fri, 2 May 1997 09:54:57 -0600 (MDT) Sender: James MacKinnon From: James MacKinnon X-Sender: jmack@tshimshan To: Dean Michaels cc: Zmailer List Subject: Re: Problems with quoted email addresses. In-Reply-To: <336A07CF.4C1F@interdynamix.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 2 May 1997, Dean Michaels wrote: > > ------- pwtest.c --------- > > #include > > #include > > #define USER " dmichael" > > main() > > { > > struct passwd *pw = getpwnam(USER); > > printf ("getpwnam(\"%s\") -> ",USER); > > if (pw == NULL) > > printf("USER UNKNOWN\n"); > > else > > printf("user='%s', uid=%d\n", pw->pw_name, pw->pw_uid); > > return 0; > > } > > Produces the following under Solaris 2.5.1 > > /Zmailer> make pwtest > cc -o pwtest pwtest.c > /Zmailer> ./pwtest > getpwnam(" dmichael") -> user='dmichael', uid=1001 > /Zmailer> > > What do you suggest, a wrapper for getpwnam()? Whoa, this may not generalise to "all" Solaris machines. (Seems not all Solaris's are equal then - Dean's machine is certainly different than mine). I get a proper response from an SS4 and an Ultra-1: SS4: # uname -a SunOS srv 5.5.1 Generic sun4m sparc SUNW,SPARCstation-4 using either gcc or cc (/usr/ucb/cc) I get: # gcc -o pwtest pwtest.c # ./pwtest getpwnam(" jmack") -> USER UNKNOWN # cc -o pwtest pwtest.c # ./pwtest getpwnam(" jmack") -> USER UNKNOWN Ultra-1: # uname -a SunOS tuzo 5.5 Generic sun4u sparc SUNW,Ultra-1 # ./pwtest getpwnam(" jmack") -> USER UNKNOWN (perhaps there is something wonky with libc on Dean's machine?) Cheers, -- James S. MacKinnon Office: P-139 Avahd-Bhatia Physics Lab Team Physics Voice : (403) 492-8226 University of Alberta email : Jim.MacKinnon@Phys.UAlberta.CA Edmonton, Canada T6G 2N5 WWW: http://www.phys.ualberta.ca/~jmack/jmack.html From zmailer-owner@nic.funet.fi Fri May 2 19:56:17 1997 Received: from NO-IDENT-SERVICE@mariner.interdynamix.com (port 29397 [206.75.211.66]) by nic.funet.fi with ESMTP id <12896-12134>; Fri, 2 May 1997 19:55:08 +0300 Received: from pongo.interdynamix.com ([192.168.1.3]) by mariner.interdynamix.com with SMTP id <495568-224>; Fri, 2 May 1997 10:59:08 -0600 Message-ID: <336A1C69.2BAC@interdynamix.com> Date: Fri, 02 May 1997 10:55:05 -0600 From: Dean Michaels Organization: InterDynamiX Systems Corp X-Mailer: Mozilla 3.01 (Win95; I) MIME-Version: 1.0 To: Ken Lalonde CC: zmailer@nic.funet.fi Subject: Re: Problems with quoted email addresses. References: <97May2.115251edt.17111-28702@angel.uunet.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: Ken Lalonde wrote: > > I get "USER UNKNOWN", as expected, on a sparc/solaris 2.5.1 machine, > using "passwd: files" in /etc/nsswitch.conf. > > What does your nsswitch.conf use? We are using nisplus, won't be the first time it's been a pain. Dean. From zmailer-owner@nic.funet.fi Fri May 2 21:21:34 1997 Received: by nic.funet.fi id <13282-7278>; Fri, 2 May 1997 21:20:15 +0300 Subject: Re: Problems with quoted email addresses. From: Matti Aarnio To: dean@interdynamix.com (Dean Michaels) Date: Fri, 2 May 1997 21:20:09 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <336A07CF.4C1F@interdynamix.com> from "Dean Michaels" at May 2, 97 09:27:11 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 327 Message-Id: <19970502182015Z13282-7278+370@nic.funet.fi> Return-Path: > /Zmailer> make pwtest > cc -o pwtest pwtest.c > /Zmailer> ./pwtest > getpwnam(" dmichael") -> user='dmichael', uid=1001 > /Zmailer> > > What do you suggest, a wrapper for getpwnam()? Sort of, look at 2.99.48 patch2. If your NIS+ resolves it, why not to use canonic username pw->pw_name ... > Dean. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 2 21:27:04 1997 Received: from root@mea.tmt.tele.fi (port 8817 [194.251.252.33]) by nic.funet.fi with ESMTP id <12588-15158>; Fri, 2 May 1997 21:25:49 +0300 Received: from mea@localhost by mea.tmt.tele.fi id <231478-16640>; Fri, 2 May 1997 21:25:57 +0000 Sender: Matti Aarnio From: mea@nic.funet.fi To: zmailer@nic.funet.fi Subject: zmailer 2.99.48 patch 2 Message-Id: <19970502182557Z231478-16640+14@mea.tmt.tele.fi> Date: Fri, 2 May 1997 21:25:52 +0000 Return-Path: Here is an extract from the 'patch2'. Some last-moment changes were not logged apparently, but it is not fatal... If you picked 'patch2' before you saw this, you may or may not get the 'final patch2'.. Oh yes, my workstation has some TCP/IP code bug, which caused ftpd to crash out, and I have been too 'lazy' to reboot this thing... (uptime: 45 days) /Matti Aarnio --- zmailer-2.99.48p1/ChangeLog Thu Apr 24 21:11:36 1997 +++ zmailer-2.99.48p2/ChangeLog Fri May 2 21:00:44 1997 @@ -1,3 +1,44 @@ +Fri May 2 16:10:28 1997 Matti Aarnio + + * Makefile.in: + - Version 2.99.48p2 + + * BUGS + Noticed problem at scheduler ETRN -> results in + busy spin... + + * man/aliases.5: + A new man-page... + + * proto/cf/fqdnaliases.c: + Missing quotes from the script + + * proto/scheduler.conf: + Sample scheduler.conf file with tags for three + different channel = local -cases: + - local/file* + - local/pipe* + - local/* + The "local/*" may be run with "sm" to use procmail, + or cyrus (for example) for deliveries to users. + The "sm" (procmail in particular?) is unable to do + pipe-feeds, nor storing to files, it seems. + (Also gave an example of 'punt' on smtp..) + + * smtpserver/smtpcmds.c: + Removed obsolete special case ESMTP token: X-TURNME + + * transports/errormail/errormail.c, transports/sm/sm.c, + transports/mailbox/mailbox.c, transports/hold/hold.c, + transports/fuzzyalias/fuzzyaliases.c, transports/libta/lockaddr.c, + transports/smtp/smtp.c: + "Make sure we are properly positioned at the start + of the message body" (lseek() things) + + * transports/mailbox/mailbox.c: + Save errno before doing library calls -- did yield wrong + error in case the 'dotlock' was busy. + From zmailer-owner@nic.funet.fi Sat May 3 09:08:19 1997 Received: from NO-IDENT-SERVICE@c.realbig.com (port 28166 [206.161.253.66]) by nic.funet.fi with ESMTP id <13180-19478>; Sat, 3 May 1997 09:07:26 +0300 Received: from ws.realbig.com ([206.161.253.65]) by roadrunner.realbig.com with ESMTP id <131106-154>; Sat, 3 May 1997 06:06:53 +0000 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86882-257>; Sat, 3 May 1997 02:06:59 -0500 Date: Sat, 3 May 1997 02:06:49 -0400 (EDT) From: Andy Poling To: zmailer@nic.funet.fi Subject: 2.99.48 smtpserver won't accept()? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to 2.99.48pl2. Now the SMTP server won't accept any connections. Strace shows it sitting in an accept() even after I've attempted to connect to it from a variety of machines. The TCP connection actually never gets established. It looks to me like there's something wrong with the bind() or some ioctl on the socket before the accept(). Before I start tearing into it, I thought I'd see if this sounds familiar to anyone here. For the moment, I've copied back the smtpserver binary from 2.9.38 and it works, but I don't like mixing and matching versions of the various parts of zmailer... Any ideas? -Andy PS - I updated due to a problem between my SMTP client and a Post.Office (gag) SMTP server at metrolink.net. Unfortunately the update didn't affect that problem, but more on that later... Global Auctions From zmailer-owner@nic.funet.fi Sat May 3 13:36:11 1997 Received: by nic.funet.fi id <13193-7409>; Sat, 3 May 1997 13:35:18 +0300 Subject: Re: 2.99.48 smtpserver won't accept()? From: Matti Aarnio To: andy@realbig.com (Andy Poling) Date: Sat, 3 May 1997 13:35:08 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Andy Poling" at May 3, 97 02:06:49 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 851 Message-Id: <19970503103518Z13193-7409+122@nic.funet.fi> Return-Path: > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > 2.99.48pl2. > > Now the SMTP server won't accept any connections. Strace shows it sitting in > an accept() even after I've attempted to connect to it from a variety of > machines. The TCP connection actually never gets established. It looks to > me like there's something wrong with the bind() or some ioctl on the socket > before the accept(). Could you send me "strace -f" logs of both smtpserver binary starts ? Err... Did you compile the ZMailer at that machine ? Which kernel sources the compilation host has ? You may have encountered IPv6 problem at 2.0.20-odd Linux kernels. In 2.0.30 those incomplete vestiges were taken out -- at my suggestion. Possibly you need to add option '-4' to the SMTPOPTIONS. .... > Any ideas? > > -Andy /Matti Aarnio From zmailer-owner@nic.funet.fi Sat May 3 18:03:12 1997 Received: from NO-IDENT-SERVICE@trixie.mse.jhu.edu (port 51721 [128.220.8.14]) by nic.funet.fi with ESMTP id <12282-19478>; Sat, 3 May 1997 18:02:34 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5891>; Sat, 3 May 1997 11:02:22 -0400 Date: Sat, 3 May 1997 11:01:52 -0400 (Eastern Daylight Time) From: Tim DiLauro To: Andy Poling cc: zmailer@nic.funet.fi Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-To: Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sat, 3 May 1997, Andy Poling wrote: > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > 2.99.48pl2. > > Now the SMTP server won't accept any connections. Strace shows it sitting in > an accept() even after I've attempted to connect to it from a variety of > machines. The TCP connection actually never gets established. It looks to > me like there's something wrong with the bind() or some ioctl on the socket > before the accept(). > > Before I start tearing into it, I thought I'd see if this sounds familiar to > anyone here. For the moment, I've copied back the smtpserver binary from > 2.9.38 and it works, but I don't like mixing and matching versions of the > various parts of zmailer... Andy: I, too, just installed 2.99.48pl2 on Linux (2.0.27, Alpha AXP, though). I had very few problems and it is now running well, although not in production yet. Specifically, smtpserver accepts connections and hands mail properly to router. I did need to comment out some IPv6 stuff in the include files, but that was it. Sorry I don't have any real constructive ideas, but I wanted to let you know that it does work on Linux. See ya. -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Sun May 4 20:37:31 1997 Received: from NO-IDENT-SERVICE@trixie.mse.jhu.edu (port 60169 [128.220.8.14]) by nic.funet.fi with ESMTP id <13246-19373>; Sun, 4 May 1997 20:36:26 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5370>; Sun, 4 May 1997 13:13:07 -0400 Date: Sun, 4 May 1997 13:12:42 -0400 (Eastern Daylight Time) From: Tim DiLauro To: zmailer distribution list Subject: router log file Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Zmailer 2.99.48p2/Linux/Alpha I seem to be unable to change the router logfile with the -L option flag. I killed router, made sure it was dead, and tried a command line like this (via zmailer shell script): router -dkn $NROUTERS -L /var/log/mail/router1x An empty /var/log/mail/router1x was created, but all the output went to /var/log/mail/router. I used gdb to attach to each of the router processes to check the value of the "logfn" variable. In each case it was "/var/log/mail/router1x". Also, Matti, it would be nice if a new variable -- say ROUTEROPTIONS -- were added to the /etc/zmailer.conf and to each invocation of "router" in the zmailer shell script. Router is the only major zmailer process that does not allow clear and easy setting of it's options in that file. Thanks. -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Mon May 5 10:48:20 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 34695 [194.67.3.135]) by nic.funet.fi with ESMTP id <13036-23754>; Mon, 5 May 1997 10:47:25 +0300 Received: from NO-IDENT-SERVICE@ariel (port 58042 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-320>; Mon, 5 May 1997 11:47:10 +0400 Date: Mon, 5 May 1997 11:46:56 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: Problems with quoted email addresses. To: Dean Michaels cc: Zmailer List In-Reply-To: <336A07CF.4C1F@interdynamix.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > ------- pwtest.c --------- > > Produces the following under Solaris 2.5.1 > > /Zmailer> make pwtest > cc -o pwtest pwtest.c > /Zmailer> ./pwtest > getpwnam(" dmichael") -> user='dmichael', uid=1001 > /Zmailer> > > What do you suggest, a wrapper for getpwnam()? Here on Solaris 2.5.1 on SPARC it works write (returns "no such user") Eugene From zmailer-owner@nic.funet.fi Tue May 6 01:34:40 1997 Received: from NO-IDENT-SERVICE@c.realbig.com (port 39428 [206.161.253.66]) by nic.funet.fi with ESMTP id <12427-11772>; Tue, 6 May 1997 01:33:17 +0300 Received: from NO-IDENT-SERVICE@b.realbig.com (port 36358 [206.161.253.65]) by roadrunner.realbig.com with ESMTP id <132963-154>; Mon, 5 May 1997 22:14:35 +0000 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86741-256>; Mon, 5 May 1997 18:32:21 -0500 Date: Mon, 5 May 1997 18:32:21 -0400 (EDT) From: Andy Poling To: zmailer@nic.funet.fi Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: In case anyone else runs into this, I'll document the problem: On Sat, 3 May 1997, I wrote: > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > 2.99.48pl2. > > Now the SMTP server won't accept any connections. Strace shows it sitting in > an accept() even after I've attempted to connect to it from a variety of > machines. The TCP connection actually never gets established. It looks to > me like there's something wrong with the bind() or some ioctl on the socket > before the accept(). Matti figured it out. I should have recognized the symptoms. Prior to the 2.0.27 Linux kernel, large arguments (larger than 8 bits - 255) caused select() to hang and never return. The new smtpserver is using select(2000) to combat denial of service attacks. -Andy Global Auctions From zmailer-owner@nic.funet.fi Tue May 6 13:21:53 1997 Received: by nic.funet.fi id <12978-143>; Tue, 6 May 1997 13:21:01 +0300 Subject: Re: 2.99.48 smtpserver won't accept()? From: Matti Aarnio To: andy@realbig.com (Andy Poling) Date: Tue, 6 May 1997 13:20:59 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Andy Poling" at May 5, 97 06:32:21 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1580 Message-Id: <19970506102101Z12978-143+142@nic.funet.fi> Return-Path: > In case anyone else runs into this, I'll document the problem: > > On Sat, 3 May 1997, I wrote: > > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > > 2.99.48pl2. > > > > Now the SMTP server won't accept any connections. Strace shows it > > sitting in an accept() even after I've attempted to connect to it > > from a variety of machines. The TCP connection actually never gets > > established. It looks to me like there's something wrong with the > > bind() or some ioctl on the socket before the accept(). > > Matti figured it out. I should have recognized the symptoms. Prior to the > 2.0.27 Linux kernel, large arguments (larger than 8 bits - 255) caused > select() to hang and never return. The new smtpserver is using > select(2000) to combat denial of service attacks. Sorry, Andy mixed things. It is listen(sock, LIMIT) where the limit value has traditionally been limited to 5, and on those systems that have it fixed, no matter how high the parameter is, it will be 5. (Or lower, minimum is 1, I think.) With newer systems the LIMIT value can be higher, and I did reason that as my Linux 2.0.27 works just fine with 2000 (and is SYN-attack resilient), and Solaris 2.5.1 and DEC/UNIX work too, I should be "home free" with it. It surprises me that limit(sock, 2000) is too high for Linux kernel version 2.0.26. The whole excercise has been for getting the system to be SYN-attack resilient in the SMTP service -- at those systems capable to be such. > -Andy > Global Auctions /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 6 16:21:53 1997 Received: from NO-IDENT-SERVICE@trixie.mse.jhu.edu (port 14858 [128.220.8.14]) by nic.funet.fi with ESMTP id <12827-1375>; Tue, 6 May 1997 16:20:50 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5883>; Tue, 6 May 1997 09:20:37 -0400 Date: Tue, 6 May 1997 09:20:04 -0400 (Eastern Daylight Time) From: Tim DiLauro To: Matti Aarnio cc: Andy Poling , zmailer@nic.funet.fi Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-To: <19970506102101Z12978-143+142@nic.funet.fi> Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 6 May 1997, Matti Aarnio wrote: > > On Sat, 3 May 1997, I wrote: > > > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > > > 2.99.48pl2. > > > > > > Now the SMTP server won't accept any connections. Strace shows it > > > sitting in an accept() even after I've attempted to connect to it > > > from a variety of machines. The TCP connection actually never gets > > > established. It looks to me like there's something wrong with the > > > bind() or some ioctl on the socket before the accept(). > > > > Matti figured it out. I should have recognized the symptoms. Prior to the > > 2.0.27 Linux kernel, large arguments (larger than 8 bits - 255) caused > > select() to hang and never return. The new smtpserver is using > > select(2000) to combat denial of service attacks. > > Sorry, Andy mixed things. It is listen(sock, LIMIT) where > the limit value has traditionally been limited to 5, and on those > systems that have it fixed, no matter how high the parameter is, > it will be 5. (Or lower, minimum is 1, I think.) > > With newer systems the LIMIT value can be higher, and I did reason > that as my Linux 2.0.27 works just fine with 2000 (and is SYN-attack > resilient), and Solaris 2.5.1 and DEC/UNIX work too, I should be > "home free" with it. I guess Configure needs another test or two (and Andy needs a little more sleep :-). -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Tue May 6 19:31:23 1997 Received: from NO-IDENT-SERVICE@c.realbig.com (port 39173 [206.161.253.66]) by nic.funet.fi with ESMTP id <13218-11098>; Tue, 6 May 1997 19:30:55 +0300 Received: from NO-IDENT-SERVICE@b.realbig.com (port 37638 [206.161.253.65]) by roadrunner.realbig.com with ESMTP id <133214-152>; Tue, 6 May 1997 12:19:27 +0000 Received: from ws.realbig.com ([127.0.0.1]) by ws.realbig.com with SMTP id <86741-255>; Tue, 6 May 1997 12:30:29 -0500 Date: Tue, 6 May 1997 12:30:25 -0400 (EDT) From: Andy Poling To: Tim DiLauro cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 6 May 1997, Tim DiLauro wrote: > > > Matti figured it out. I should have recognized the symptoms. Prior to the > > > 2.0.27 Linux kernel, large arguments (larger than 8 bits - 255) caused > > > select() to hang and never return. The new smtpserver is using > > > select(2000) to combat denial of service attacks. > > > > Sorry, Andy mixed things. It is listen(sock, LIMIT) where > > the limit value has traditionally been limited to 5, and on those > > systems that have it fixed, no matter how high the parameter is, > > it will be 5. (Or lower, minimum is 1, I think.) > > > > With newer systems the LIMIT value can be higher, and I did reason > > that as my Linux 2.0.27 works just fine with 2000 (and is SYN-attack > > resilient), and Solaris 2.5.1 and DEC/UNIX work too, I should be > > "home free" with it. > > I guess Configure needs another test or two (and Andy needs a little more > sleep :-). I'll say. If I only had a brain... -Andy Global Auctions From zmailer-owner@nic.funet.fi Tue May 6 19:33:46 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 12040 [204.244.210.193]) by nic.funet.fi with SMTP id <13083-11098>; Tue, 6 May 1997 19:33:33 +0300 Received: from misery.sdf.com [204.244.213.33] by misery.sdf.com with smtp (Exim 1.61 #1) id 0wOn6s-0003Sc-00; Tue, 6 May 1997 09:29:02 -0700 Date: Tue, 6 May 1997 09:29:01 -0700 (PDT) From: Tom Samplonius To: Matti Aarnio cc: Andy Poling , zmailer@nic.funet.fi Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-To: <19970506102101Z12978-143+142@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 6 May 1997, Matti Aarnio wrote: > > In case anyone else runs into this, I'll document the problem: > > > > On Sat, 3 May 1997, I wrote: > > > I just updated one of my Linux (2.0.26) systems from zmailer 2.99.38 to > > > 2.99.48pl2. > > > > > > Now the SMTP server won't accept any connections. Strace shows it > > > sitting in an accept() even after I've attempted to connect to it > > > from a variety of machines. The TCP connection actually never gets > > > established. It looks to me like there's something wrong with the > > > bind() or some ioctl on the socket before the accept(). > > > > Matti figured it out. I should have recognized the symptoms. Prior to the > > 2.0.27 Linux kernel, large arguments (larger than 8 bits - 255) caused > > select() to hang and never return. The new smtpserver is using > > select(2000) to combat denial of service attacks. > > Sorry, Andy mixed things. It is listen(sock, LIMIT) where > the limit value has traditionally been limited to 5, and on those > systems that have it fixed, no matter how high the parameter is, > it will be 5. (Or lower, minimum is 1, I think.) > > With newer systems the LIMIT value can be higher, and I did reason > that as my Linux 2.0.27 works just fine with 2000 (and is SYN-attack > resilient), and Solaris 2.5.1 and DEC/UNIX work too, I should be > "home free" with it. > > It surprises me that limit(sock, 2000) is too high for Linux > kernel version 2.0.26. > > The whole excercise has been for getting the system to be SYN-attack > resilient in the SMTP service -- at those systems capable to be such. > > > -Andy > > Global Auctions > > /Matti Aarnio I was wondering why smtpserver would be calling select() on incoming connections, now I know :) BTW, BSD4.4 systems have a define in sys/socket.h called SOMAXCONN that defines the maximum value for listen. If the supplied param is bigger than SOMAXCONN, it is set to SOMAXCONN. I'm pretty sure that Solaris does this too. I really doubt that you are actually get a "real" 2000 entry listen queue :) Tom From zmailer-owner@nic.funet.fi Tue May 6 22:54:26 1997 Received: by nic.funet.fi id <13138-26828>; Tue, 6 May 1997 22:53:42 +0300 Subject: Re: 2.99.48 smtpserver won't accept()? From: Matti Aarnio To: tom@sdf.com (Tom Samplonius) Date: Tue, 6 May 1997 22:53:35 +0300 (EET DST) Cc: andy@realbig.com, zmailer@nic.funet.fi In-Reply-To: from "Tom Samplonius" at May 6, 97 09:29:01 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 971 Message-Id: <19970506195342Z13138-26828+3@nic.funet.fi> Return-Path: Tom Samplonius writes: > > I was wondering why smtpserver would be calling select() on incoming > connections, now I know :) > > BTW, BSD4.4 systems have a define in sys/socket.h called SOMAXCONN that > defines the maximum value for listen. If the supplied param is bigger > than SOMAXCONN, it is set to SOMAXCONN. Yes, things used to be like that. I don't think they are anymore, though. At Solaris, and DEC UNIX there are variables in kernel (not #defines anymore), and at DEC UNIX 4.0 there are even ways to enforce minimum value to be higher than the default 5 for use on system running binaries with low values in the listen() limit. On Solaris there seems to be runtime config parameter: tcp_conn_req_max though what value it has now, I haven't tried to find out. > I'm pretty sure that Solaris does this too. I really doubt that you are > actually get a "real" 2000 entry listen queue :) Yes, you can get it :-) > Tom /Matti Aarnio From zmailer-owner@nic.funet.fi Wed May 7 00:10:30 1997 Received: from NO-IDENT-SERVICE@crocus.gamma.ru (port 56847 [193.124.255.1]) by nic.funet.fi with ESMTP id <12787-30121>; Wed, 7 May 1997 00:09:45 +0300 Received: from srcc.UUCP (uucp@localhost) by crocus.gamma.ru (8.8.5/8.7.3) with UUCP id BAA00717 for zmailer@nic.funet.fi; Wed, 7 May 1997 01:09:19 +0400 (MSD) Received: by gamma.srcc.msu.su; Wed, 7 May 1997 01:08:50 +0400 Received: from localhost(really [127.0.0.1]) by pccross.average.org via smtpd with esmtp id for ; Wed, 7 May 1997 00:08:39 +0400 (MSD) (Smail-3.2 1996-Jul-4 #12 built 1996-Jul-20) Date: Wed, 7 May 1997 00:08:39 +0400 (MSD) From: Eugene Crosser Sender: crosser@pccross.average.org Reply-To: Eugene Crosser Subject: Anti-spam filtering To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="1136789001-1254073978-862949319=:2398" Return-Path: --1136789001-1254073978-862949319=:2398 Content-Type: TEXT/PLAIN; LANGUAGE=ru I thought that I should repost this one from the "smail-users" mailing list... Eugene --1136789001-1254073978-862949319=:2398 Content-Type: MESSAGE/RFC822 Content-Description: Forwarded: [Re: spam filters for smail? ] Received: from srcc by pccross.average.org via rmail with uucp id for crosser@pccross.msk.su; Tue, 6 May 1997 23:39:22 +0400 (MSD) (Smail-3.2 1996-Jul-4 #12 built 1996-Jul-20) Received: by gamma.srcc.msu.su; Tue, 6 May 1997 19:52:37 +0400 Received: from aupair.cs.athabascau.ca (root@aupair.cs.athabascau.ca [131.232.10.8]) by gw-srcc.gamma.ru (8.8.5/8.8.5) with ESMTP id TAA28266 for ; Tue, 6 May 1997 19:51:05 +0400 (MSD) Resent-Date: Tue, 6 May 1997 19:51:05 +0400 (MSD) Received: by aupair.cs.athabascau.ca via sendmail with stdio id for smail3-wizards-dist@cs.athabascau.ca; Tue, 6 May 1997 08:58:43 -0600 (MDT) (Smail-3.2.0.91 1997-Jan-14 #1 built 1997-May-2) Old-Return-Path: Message-Id: <199705061458.IAA20393@lautrec.esys.ca> X-Authentication-Warning: lautrec.esys.ca: lyndon@localhost didn't use HELO protocol To: Larry Auton cc: bdb@hotlips.gts.net (Bruce Becker), smail3-wizards@cs.athabascau.ca Subject: Re: spam filters for smail? In-reply-to: Your message of "Mon, 05 May 1997 14:14:28 EDT." Mime-Version: 1.0 Content-Type: multipart/mixed ; boundary="==_Exmh_6106623340" Date: Tue, 06 May 1997 08:58:32 -0600 From: Lyndon Nerenberg Resent-Message-ID: <"moctT.A.t-D.jc0bz"@aupair.cs.athabascau.ca> Resent-From: smail3-wizards@cs.athabascau.ca X-Mailing-List: archive/latest/152 X-Loop: smail3-wizards@cs.athabascau.ca Precedence: list Resent-Sender: smail3-wizards-request@cs.athabascau.ca This is a multipart MIME message. --==_Exmh_6106623340 Content-Type: text/plain; charset=us-ascii Folks, there is work ongoing within the IETF to define a general purpose filtering scheme for email. Rather than inventing some adhoc system we should be following their work. If you're interested in this you want to get onto the MTA-FILTERS mailing list: echo subscribe ietf-mta-filters | mail majordomo@imc.org. I've appended a copy of the current filtering language draft. --lyndon --==_Exmh_6106623340 Content-Type: text/plain ; name="draft-showalter-sieve-00.txt"; charset=us-ascii Content-Description: draft-showalter-sieve-00.txt Content-Disposition: attachment; filename="draft-showalter-sieve-00.txt" Network Working Group T. Showalter Internet Draft Carnegie Mellon Document: draft-showalter-sieve-00.txt April 1997 Expire sometime after it's released SIEVE: A Mail Filtering Language Status of this Memo This draft has no standing and is for limited distribution. Please do not implement anything within without contacting the author, and even then, use extreme caution; everything in this draft is carved in warm butter. Please use discretion in distribution. The following is here for when we actually publish this: This document is an Internet-Draft. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as ``work in progress.'' To learn the current status of any Internet-Draft, please check the ``1id-abstracts.txt'' listing contained in the Internet- Drafts Shadow Directories on ftp.is.co.za (Africa), ftp.nordu.net (Europe), munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), or ftp.isi.edu (US West Coast). The protocol discussed in this document is experimental and subject to change. Persons planning on either implementing or using this protocol are STRONGLY URGED to get in touch with the author before embarking on such a project. Abstract This document describes a mail filtering language for filtering messages at time of final delivery. It is designed to be independent of protocol, and implementable on either a mail client or mail server which uses multiple folders. It is meant to be extensible, simple, and independent of access protocol, mail architecture, and operating systems used to implement it. Mail filtering systems are widely used for a variety of reasons, including organization of messages (filtering out mailing list Showalter [Page 1] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 messages for separate reading). They are becoming increasingly useful in avoiding unsolicited mail. Existing languages are not consistant across client, server, or operating system, and are frequently difficult for users to use. This language is not tied to any particular system or mail architecture and is suitable for running on a mail server where users may not be allowed to execute arbitrary programs, such as on black box IMAP servers. Showalter [Page 2] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 Table of Contents This document is content-free. Showalter [Page 3] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 0. Unfinished 0.1. Known Weaknesses The following suggestions have been made, and will probably be addressed by extensions. An extension for regular expressions will be written. While regular expressions are of questionable usefulness for users, the programmers writing implementations desperately want regular expressions. Envelope-matching commands are not readily supported by all mail systems, and putting them in the draft will result in a system that cannot be implemented by a mail architecture that does not adequately store envelopes. "Detailed" addressing or "subaddressing" (i.e., the "fmh" in an address "tjs+fmh@andrew.cmu.edu") is not handled, and will be moved to an extension for those systems that offer it. The newline problem is relatively, but not completely, solved. We'll be arguing this until the end of time. A previous version included a "valid" test. I have tentatively removed it because Randy had noted it was too fuzzy to implement, and that's probably true. The formal grammar is not complete, and needs to be revised to make the best use of ABNF, whatever its final state is. My knowledge of email is not comprehensive, and as a result, there might be a better way to express some of the concepts in here. An "include" command is not included, but has been suggested for an extension. I need to run a spelling checker on this document. I hate nroff. 0.2. Open Issues Do "fileinto" commands need to be moved to a separate document? 1. Introduction There are a number of reasons to use a filtering system. Mail traffic for most users has been increasing due both to increased Showalter [Page 4] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 usage of e-mail, the emergence of unsolicited email as a form of advertising, and increased usage of mailing lists. This language is offered in order to try and provide a standard language that can be used to create filters for e-mail. It is not tied to any particular operating system or mail architecture. It requires the use of [IMAIL]-compliant messages and support of multiple folders, but should work with a wide variety of systems that support these criteria. The language is powerful enough to be useful, but limited in power in order to allow for a reasonably bulletproof server-side filtering system. The language is not Turing-complete, and provides no way to write a loop or a function, nor are variables are provided. The intention is to make it impossible for users to do anything more complex than write simple mail filters. Implementations of the language are expected to take place at time of final delivery. In systems where the MTA does final delivery -- IMAP4 and traditional UNIX mail, for instance, it is reasonable to sort when the MTA deposits mail into the user's mailbox. If the MTA does not do final delivery, or lacks the power to sort into separate mailboxes (as is the case under POP3), the MUA must do filtering into local filters. Experience at Carnegie Mellon has shown that if a filtering system is made available to users, many will make use of it in order to file messages from specific users or mailing lists. However, many users did not make use of the Andrew system's FLAMES filtering language due to difficulty in programming it. Due to this expectation, this language has been made simple enough to allow many users to make use of it. 1.1. Conventions used in this document Line breaks have been inserted for readability. In the sections of this document that discuss the requirements of various keywords and operators, the following conventions have been adopted. Each section on an test, action, or conditional has a line labeled "Syntax:". This line describes the arguments each command requires. Required arguments are listed inside angle brackets ("<" and ">"). Optional arguments are listed inside square brackets ("[" and "]"). The formal grammar for these commands is described in section 10 and is the authoritative reference on how to construct these commands. Showalter [Page 5] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 The key words "MUST", "MUST NOT", "SHOULD", "SHOULD NOT", "CAN", and "MAY" in this document are to be interpreted as defined in [KEYWORDS]. 1.2. Example mail messages The following mail messages will be used throughout this document in examples. Message A ----------------------------------------------------------- Date: Tue, 1 Apr 1997 09:06:31 -0800 (PST) From: coyote@anvil.dementia.org To: roadrunner@birdseed.thekeep.org Subject: I have a present for you Look, I'm sorry about the whole anvil thing, but I can make it up to you. I've got some great birdseed over here at my place -- top of the line stuff -- and if you come by, I'll have it all wrapped up for you. I'm really sorry for all the problems I've caused for you over the years, and I know we can work this out. ----------------------------------------------------------- Message B ----------------------------------------------------------- From: youcouldberich!@reply-by-postal-mail Sender: b1ff@znic.net To: rube@znic.net Date: Mon, 31 Mar 1997 18:26:10 -0800 (PST) Subject: $$$ YOU, TOO, CAN BE A MILLIONAIRE! $$$ YOU MAY HAVE ALREADY WON TEN MILLION DOLLARS, BUT I DOUBT IT! SO JUST POST THIS TO SIX HUNDRED NEWSGROUPS! IT WILL GUARANTEE THAT YOU GET AT LEAST FIVE RESPONSES WITH MONEY! MONEY! MONEY! COLD HARD CASH! YOU WILL RECEIVE OVER $20,000 IN LESS THAN TWO MONTHS! AND IT'S LEGAL!!!!!!!!! !!!!!!!!!!!!!!!!!!111111111!!!!!!!11111111111!!1 JUST SEND $5 IN SMALL, UNMARKED BILLS TO THE ADDRESSES BELOW! ----------------------------------------------------------- 2. Design 2.1. Form of the language This language is made up as a set of commands. Each command is either an action or a conditional. Each conditional contains a test; depending on the results of the test, one set of commands in a Showalter [Page 6] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 control structure is taken. 2.2. Whitespace Whitespace is used to separate commands. Whitespace is made up of tabs, newlines (which can be CR, LF, or both), and the space character. The amount of whitespace used is not significant. 2.3. Comments Comments begin with a "#" character that is not contained within a string and continue until the next newline. Example: if size over 100K then # this is a comment toss endif 2.4. Numbers Numbers are normally given as ordinary decimal numbers. However, those numbers that have a tendency to be fairly large, such as message sizes, such as message sizes, may have a "K", "M", or "G" appended to indicate a multiple of a base-two number. To be comparable with the power- of-two-based versions of SI units that computers fre- quently use, K specifies kilo, or 1,024 (2^10) times the value of the number; M specifies mega, or 1,048,576 (2^20) times the value of the number; and G specifies giga, or 1,073,741,824 (2^30) times the value of the number. Numbers are limited to 32 bits by this specification. [OPEN: If numbers are limited to 32 bits, gigabit-sized numbers probably aren't very useful. Should I remove them?] 2.5. Strings Scripts involve large numbers of strings. Typically, short quoted strings suffice for most uses, but a more convenient form is provided for longer strings. A quoted string starts and ends with a single double quote (the " character). A backslash ("\") inside of a quoted string is followed by either another backslash or a double quote. This two-character sequence represents a Showalter [Page 7] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 single backslash or double-quote within the string. [If there are missing words in the paragraph, I had problems with nroff; please point it out to me.] For entering larger amounts of text, such as an email message, a longer form is allowed, known as a "user- message". It starts with the keyword "message" and ends with the sequence of a newline, a single period, and another newline. Any line that begins with ".." is con- sidered to begin with ".". XXX this example needs formatting work Example: if any-of (header ("from") contains ("bart" "homer" "smithers" "burns" "lisa") then header ("subject") contains ("URGENT")) then fileinto "INBOX" else reply message You are not one of the people I regularly correspond with. I have deleted your message due to the large volume of email I regularly receive. If you feel that you need to speak with me directly, and cannot find your answer in my web pages, please send mail with the word "URGENT" in the subject line. Thank you for your time. [ONE SINGLE DOT, which I can't get nroff to produce ...] endif 2.5.1. Headers [OPEN ISSUE: Is this section necessary or useful?] Headers are a subset of strings. In the Internet Message Specification [IMAIL], each header line is allowed to have whitespace nearly anywhere in the line, including after the field name and before the subsequent colon. Thus, the lines "From: acm@andrew.cmu.edu" is equivalent to "From : acm@andrew.cmu.edu". Within a SIEVE script, header names are never considered to have spaces. Only the "From" in the above headers is considered to be there. While a header can be listed as "From " within a header list (say, for the "header" command) such usage is absurd. The following colon is never specified; a header "From:", as well as a header ":From", is guaranteed never to happen within a valid header. 2.5.2. Addresses Showalter [Page 8] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 [OPEN ISSUE: Is this section necessary or useful?] A number of commands call for email addresses, which are also a subset of strings. These addresses must be compliant with [IMAIL]. Implemen- tations MUST insure the addresses are syntactically valid, and need not insure that they are actually deliverable. 2.7. Evaluation If evaluation of the script fails to file the message into any mailbox, as in the following script, the message is filed into INBOX. With any of the short messages offered above, the following script produces no actions. Example: if size over 500K then toss endif then the "normal" action is taken. The "normal" action is defined to be the action that is taken normally, such as in a situation where the user does no filtering. Under most situations, the normal action is to file into the user's main mailbox (such as "INBOX" under IMAP). Implementations define the specific meanings of actions. Implementa- tions may impose restrictions on the actions taken, such as only honor- ing one "reply", "bounce", or "forward" per message. Precedence is not important in any of the commands in this base specifi- cation. However, as an extension might make it more important, all rules MUST be evaluated in left-to-right order. Those operations that may implement short-circuit evaluation (such as the "all-of" and "any- of" operators, which preform logical "and" and "or" operations, respec- tively) SHOULD do so. 3. Conditionals and Control Structures In order for a script to do more than one set of actions, control struc- tures are needed. 3.1. If Syntax: if then [elsif then [elsif ...]] [else ] endif The "if" control structure is borrowed from any number of programming languages. It is evaluated in the usual way, as follows: if is true, then are evaluated. If an elsif keyword exists, and is true, then are evaluated. Any number of Showalter [Page 9] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 elsif cases may be included, and are evaluated serially. If is false and the s are false as well, then the are evaluated. The "if" block is terminated with an "endif" keyword, which is required. In the following example, both Message A and B are dropped. Example: if header ("from") contains ("coyote") then toss elsif header ("subject") contains ("$$$") then toss else fileinto "INBOX" endif Only one set of commands in an if ... elsif ... elsif ... else ... endif block is executed. In the script below, when run over message A, forwards the message to acm@andrew.cmu.edu; message B, to service@andrew.cmu.edu; any other mes- sage is forwarded to postman@andrew.cmu.edu. Example: if header ("") contains ("") then forward "acm@andrew.cmu.edu"; elsif header ("Subject") contains ("$$$") then forward "service@andrew.cmu.edu"; else forward "postman@andrew.cmu.edu"; endif 3.2. Require Syntax: require Require SHOULD be declared in a user script before an extension is used. It instructs the evaluator that the extension named extension-name, sup- plied as a string, MUST be present in order to allow further processing. If the string specifies an extension that the evaluating mechanism sup- ports, then processing continues. Otherwise, an error has been encoun- tered, and the script should not be evaluated. Require is intended to demand the use of an extension not present in this document. The following example will fail on any server that does not implement the extension known as DWIM. Showalter [Page 10] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 Example: require "dwim"; if header ("subject") contains-nocase ("the secret message") then dwim blurdybloop body; endif stop 4. Actions This document supplies six actions that may be taken on a message: normal, fileinto, forward, bounce, toss, and stop. 4.1. Action bounce Syntax: bounce The "bounce" action resends the message to the sender, wrapping it in a "bounce" form, noting that it was rejected by the recipient. In the following script, message A is bounced to the sender. Example: if header ("from") contains ("coyote@anvil.dementia.org") then bounce "I am not taking mail from you, you killer!" endif 4.2. Action fileinto Syntax: fileinto [OPEN ISSUE: I could be talked into making fileinto optional for POP3 server agents that wanted to simply throw mail out and then do user filtering on the client.] The "fileinto" action drops the message into a named folder. In the following script, message A is filed into folder "INBOX.harassment". Example: if header ("to") contains 4.3. Action forward Syntax: forward
The "forward" action is used to forward the message to another user at the supplied address, as a mail forwarding feature does. The "forward" action makes no changes to the message body or headers, and only modi- fies the envelope. A simple script can be used for forwarding: Example: forward "tjs@andrew.cmu.edu" Showalter [Page 11] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 4.4. Action normal Syntax: normal The "normal" action is whatever action is taken in lieu of all other actions; generally, this simply means to drop the message into the user's normal mailbox. This command provides a way to execute this action without naming it explicitly, providing a way to use it indepen- dent of system. Syntax: if size under 1M then normal else toss 4.5. Action reply Syntax: reply The "reply" action is used to generate a form letter reply to the origi- nal sender. Message is a string to be sent as a reply message. The multi-line production described in the Formal Grammar section is intended for this purpose. In the following example, any message over 500K (or 512,000 octets) would be thrown out; otherwise, the message would be filed into INBOX. Example: if size over 500K reply message Your message was unnecessarily large. I reject all large messages; you will need to contact me directly. toss endif OPEN: Specify headers transmitted? OPEN: Specify way to do vacation? A previous version of this had a -days switch to specify number of days to do a new reply. 4.6. Action stop The "stop" action ends all processing. If no actions have been executed, then the normal action is taken. In the following script, if the mail is from the address "wall@andrew.cmu.edu" it is forwarded to "tjs@xanadu.wv.us"; otherwise the mail receives a reply, and is thrown out. Example: if (header ("from") matches ("wall@andrew.cmu.edu")) then forward "tjs@xanadu.wv.us"; stop endif reply "I'm on vacation and not taking any messages; try after Sunday. I have thrown your message out. Please resend it later." ; toss Showalter [Page 12] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 4.7. Action toss Toss drops the message. In the following script, any mail from "wall@andrew.cmu.edu" is thrown out. Example: if header ("from") contains ("wall@andrew.cmu.edu") then toss endif 5. Tests Tests are used in conditionals to decide which part(s) of the condi- tional to execute. 5.1. all-of Syntax: all-of ( [,] [,] ... ) The all-of test preforms a logical AND on the tests supplied to it. The comma in between tests is optional. Example: all-of (false false) => false all-of (false true) => false all-of (true, true) => true 5.2. any-of Syntax: all-of ( [,] [,] ... ) The any-of test preforms a logical OR on the tests supplied to it. The comma in between tests is optional. Example: all-of (false false) => false all-of (false true) => true all-of (true, true) => true 5.3. exists Syntax: exists The "exists" test is true if the headers listed in the argument exist within the message. All of the headers must exist or the test is false. Showalter [Page 13] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 Example: if not exists ("From" "Date" "Subject" "To" "Message-ID") then toss endif 5.4. false Syntax: false The "false" test always evaluates to false. 5.5. header Syntax: header <"contains"/"is"/"matches"/"contains-nocase" / "is- nocase"/"matches-nocase"> The "header" test evaluates to true if the header name matches key. How the match is done is described by the second argument. The basic match- ing forms are case sensitive. Each matching form has a corresponding form ending in "-nocase"; these are not case sensitive. All matchings on header field names MUST be done in a case insensitive manner. The "is" argument demands that one of the fields of the headers listed in header-name-list can be found in the key-list. It requires an abso- lute match. It is true if there are repeated arguments in the header- name-list or the key-list. The "contains" argument demands that one of the values of the headers named in header-name-list partially matches one of the values in key- list. It is true if there are repeated arguments in the header-name- list or the key-list. The string "" is contained in any header that exists. The "matches" argument demands that one of the fields of the headers listed in header-name-list matches a "glob" pattern described by one of the members of key-list. A glob pattern is a UNIX-style filename glob, which has the following special characters: * Match zero or more characters ? Match any single character Escape next character The string "" matches all strings that exist. That is, if a message Showalter [Page 14] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 contains the line [XXX formatted improperly] X-Blurdybloop: death to the heathens the tests on that header evaluate as follows: header ("X-Blurdyblop") is ("") => false header ("X-Blurdyblop") matches ("") => true header ("X-Blurdyblop") contains ("") => true 5.6. not Syntax: not The "not" test takes some other test, and returns the opposite result. 5.7. size Syntax: size <"over" / "under"> The "size" test deals with the size of a message. The test is true only if the second argument is "over" and the size of the message is strictly greater than the number of octets specified as limit. If the second argument is "under", then the test is true only if the message size is strictly less than the number of octets specified as limit. In either case, if the message size is exactly the limit, the test is false. The size of a message is defined to be the number of octets from the initial header until the last character in the message body. 5.8. support Syntax: support The "support" test evaluates to true if the extension named by is supported. In the following script, all mail is filed into INBOX unless the "black-magic" extension is supported. Oth- erwise, behavior is defined by the black-magic extension. Syntax: if support "black-magic" then black-magic ("kgb@andrew.cmu.edu") endif Showalter [Page 15] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 5.9. true Syntax: true The "true" test is always true. 6. Errors in Processing a Script In any sort of programming language, even a very simple one, errors are inevitable. In this case, users are expected to make errors -- even if the actual script is machine-generated, mailbox rights might change to disallow users from writing to a mailbox, a mailbox may no longer exist, or a variety of other problems. It is imperative that mail be allowed to get through in any case. If an error is found in a script, an implementation MUST make an attempt to resolve the condition. Implementations SHOULD check a script before it is run in order to insure that it is valid. Implementations SHOULD NOT try and recover from a script with errors, and should file mail into the user's primary mailbox. Users MUST be notified of errors in processing a script. The method by which users are notified is implementation defined, but a mail message describing the error is suggested if a preferable alternative cannot be found Implementations that allow for the script to be checked for syntax errors in advance of mail receipt (i.e., client-based filtering, or server-based filtering with a submission protocol aware of this language) SHOULD notify the user of the error and refuse to accept a syntactically invalid script, or one that makes use of extensions that the server does not report. Implementations that allow server-based filtering (i.e., as part of an IMAP server) MUST allow mail to be filed normally (i.e., for IMAP, into the user's INBOX) in case of a syntax error in the script, and MUST notify the user of an error in some form (such as sending the user a mail message notifying them of the error). Implementations SHOULD avoid over-sending error messages to the user's mailbox. Implementations are REQUIRED to notify users of errors in filtering scripts. If there are errors in the script being used, mail SHOULD be filed into INBOX. Implementations MUST NOT discard mail. 7. Extensibility New control structures, actions, and tests can be added to the language. Showalter [Page 16] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 Sites must make these features known to their users; an extension nego- tiation mechanism is not defined by this document. For the formal grammar, an extension SHOULD define one of the symbols beginning with "extension-". Any extensions to this language MUST define a unique string that describes that extension. Such strings SHOULD include a version number. The purpose of such a string is for the "require" and "support" condi- tionals, which mandates that script requires the use of that extension. Additionally, in a situation where there is a submission protocol and an extension advertisement mechanism, so that scripts submitted can be checked against the mail server for valid extensions. 7.1. Capability Mechanism [A brief description of the capability string will be included here.] 7.2. Registry [A registration mechanism will be included here.] 7.3. Capability Transport [A brief description of what is required for a capability transport will be defined here. Transports will be defined in separate documents.] 8. Transmission This document does not define a method for accessing stored scripts at run-time or as they are written, nor does it define a character set encoding for scripts. If the method of handing a script off to the server allows for MIME- typing of data as described in [MIME] and the data is encoded in UTF-8 as described in [UTF-8], the MIME type for a SIEVE script is XXX/XXX. Implementations SHOULD check a script before it is run in order to insure that it is valid. Implementations SHOULD NOT try and recover from a script with errors, and should file mail into the user's primary mailbox. 9. Acknowledgments 10. Formal Grammar The grammar used in this section is the same as the ABNF described in [ABNF] with one exception: the delimiter used with "#" is any amount of Showalter [Page 17] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 whitespace (that is, CR, LF, spaces, and tabs), as described by the "WSP" terminal, followed by a single comma, and additional whitespace. Two commas without something in between them is a protocol error, and is prohibited. In the case of alternative or optional rules in which a later rule over- laps an earlier rule, the rule which is listed earlier MUST take prior- ity. action = toss / fileinto / forward / bounce / reply / stop / extension-action address = string ;; any legal IMAIL address any-of = "any-of" WSP "(" [WSP] #(condition) [WSP] ")" all-of = "all-of" WSP "(" [WSP] #(condition) [WSP] ")" big-number = number [ UNIT ] bounce = "bounce" WSP string ;; string is a text message to be sent with the bounce as the ;; reason control-structure = if / extension-control-structure command = action [WSP] ";" [WSP] / control-structure commands = #(command) comment = "#" *CHAR newline fileinto = "fileinto" WSP string forward = "forward" WSP address if = "if" WSP test WSP "then" WSP commands WSP #("elsif" WSP test WSP "then" WSP commands) [ "else" WSP commands WSP ] "endif" ;; if then ;; [elsif then [elsif ...]] ;; [else ] endif header = "header" WSP string-list WSP match-keyword WSP string-list match-keyword = "contains" / "matches" / "is" / "contains-nocase" / "contains-nocase" / "is-nocase" Showalter [Page 18] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 newline = CRLF / CR / LF ;; A CRLF is ALWAYS one newline. number = 1*DIGIT or = condition WSP "or" WSP condition quoted-string = " ;; ;; \ inside a string maps to ;; Note that newlines and other weird characters ;; are all strings. size = "size" WSP ( "over" / "under" ) WSP big-number stop = "stop" string = quoted-string / user-message string-list = "(" [WSP] #(string) [WSP] ")" test = [WSP] any-of / all-of / exists / false / header / not / size / extension-test [WSP] UNIT = "K" / "M" / "G" ;; kilobytes, megabytes, or gigabytes user-message = "message" [WSP] newline "." newline ;; note when used, ;; a CR that is not followed by an LF becomes a CRLF; ;; an LF that is not followed by a CR becomes a CRLF. ;; a leading .. on a line is mapped to . WSP = " " / CR / LF / tab ;; just whitespace 10. Security Considerations Users must get their mail. It is impera- tive that whatever method implementations use to store the user-defined filtering scripts be reasonably secure. It is equally important that implementations sanity-check the user's scripts, and not allow users to create on-demand mailbombs. For instance, an implementation that allows a user to bounce, forward, or reply multiple times to a single message might also allow a user to create a mailbomb triggered by mail from a specific user. 11. Author's Address Showalter [Page 19] draft-showalter-sieve-PRE.txt SIEVE 8 Apr 1997 Tim Showalter Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 E-Mail: tjs@andrew.cmu.edu Appendix A. References [ABNF] Crocker, D., "Augmented BNF for Syntax Specifications: ABNF", Internet Mail Consortium, Work in Progress. [KEYWORDS] Bradner, S., "Key words for use in RFCs to Indicate Require- ment Levels", RFC 2119, Harvard University, March 1997. [IMAP] Crispin, M., "Internet Mail Access Protocol - version 4rev1", RFC 2060, University of Washington, December 1996. [IMAIL] Crocker, D., "Standard for the Format of ARPA Internet Text Mes- sages", STD 11, RFC 822, University of Delaware, August 1982. [SMTP] Postel, J., "Simple Mail Transfer Protocol", STD 10, RFC 821, USC/Information Sciences Institute, August 1982. [UTF-8] Yergeau, F. "UTF-8, a transformation format of Unicode and ISO 10646", RFC 2044, Alis Technologies, October 1996. Showalter [Page 20] --==_Exmh_6106623340-- --1136789001-1254073978-862949319=:2398-- From zmailer-owner@nic.funet.fi Fri May 9 23:56:06 1997 Received: from root@bluestem.prairienet.org (port 57478 [192.17.3.4]) by nic.funet.fi with ESMTP id <12758-3256>; Fri, 9 May 1997 23:55:02 +0300 Received: from bozo (slip218.ice.net [206.102.146.218]) by bluestem.prairienet.org (8.8.5/8.8.5) with SMTP id PAA08789 for ; Fri, 9 May 1997 15:50:44 -0500 (CDT) Message-Id: <199705092050.PAA08789@bluestem.prairienet.org> Comments: Authenticated sender is From: kunglao@prairienet.org To: zmailer@nic.funet.fi Date: Sat, 10 May 1997 15:53:38 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: FAQ, docs, books? Priority: normal X-mailer: Pegasus Mail for Win32 (v2.53/R1) Return-Path: before I ask too many stupid questions, is there a web or ftp site or something with plenty of documentation on zmailer? Or is it fully sendmail compatible, that I can just pick up sendmail docs/books and manage through it? Thanks --- "Oh, what sad times are these when passing ruffians can say 'Ni!' at will to old ladies. There is a pestulance upon this land, nothing is sacred. Even those who arrange and design shrubberies are under considerable economic stress of this period of history." -- Roger the Shrubber From zmailer-owner@nic.funet.fi Sat May 10 00:54:02 1997 Received: from IDENT-NONSENSE@babbage.mingpaoxpress.com (port 9211 [205.150.120.3]) by nic.funet.fi with ESMTP id <563-12547>; Sat, 10 May 1997 00:53:13 +0300 Received: by www.mingpaoxpress.com id <1930-287>; Fri, 9 May 1997 17:52:28 -0400 Received: by www.mingpaoxpress.com id <1929-287>; Fri, 9 May 1997 17:52:13 -0400 Path: acli Subject: mailbox locking failures Organization: Ming Pao Daily News (Canada) Message-ID: Date: Fri, 9 May 1997 21:52:09 GMT From: Ambrose Li To: zmailer@nic.funet.fi Reply-To: Ambrose Li Return-Path: It has been some time since I upgraded my zmailers. Yesterday I came across an unexpected error in my 2.99.43. The error was a failure to dotlock the mailbox; I was just starting up pine at the moment, which lasted a few seconds. I was wondering whether newer versions of zmailer will treat failure to dotlock as a transient condition and retry later, or still treat it as a permanent error and bounce the mail. -- Ambrose Li. acli@mingpaoxpress.com. Ming Pao Newspapers (Canada) Ltd., EDP department. 1355 Huntingwood Drive, Scarborough, Ontario, M1S 3J1 Canada. Voice +1 416 321 0088 Fax +1 416 321 9663. From zmailer-owner@nic.funet.fi Sat May 10 01:09:54 1997 Received: by nic.funet.fi id <654-27840>; Sat, 10 May 1997 01:08:47 +0300 Subject: Re: mailbox locking failures From: Matti Aarnio To: acli@mingpaoxpress.com Date: Sat, 10 May 1997 01:08:40 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Ambrose Li" at May 9, 97 09:52:09 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 935 Message-Id: <19970509220847Z654-27840+9@nic.funet.fi> Return-Path: > It has been some time since I upgraded my zmailers. Yesterday > I came across an unexpected error in my 2.99.43. > > The error was a failure to dotlock the mailbox; I was just > starting up pine at the moment, which lasted a few seconds. > I was wondering whether newer versions of zmailer will treat > failure to dotlock as a transient condition and retry later, > or still treat it as a permanent error and bounce the mail. Version "2.99.48 patch 2" fixes that. I did finally begun to wonder the phenomena enough to check it; ChangeLog tells: * transports/mailbox/mailbox.c: Save errno before doing library calls -- did yield wrong error in case the 'dotlock' was busy. > -- > Ambrose Li. acli@mingpaoxpress.com. Ming Pao Newspapers (Canada) Ltd., > EDP department. 1355 Huntingwood Drive, Scarborough, Ontario, M1S 3J1 > Canada. Voice +1 416 321 0088 Fax +1 416 321 9663. /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 13 05:28:15 1997 Received: from TIMEDOUT@rainier.nds.com (port 3091 [199.74.149.12]) by nic.funet.fi with ESMTP id <13097-32299>; Tue, 13 May 1997 05:27:01 +0300 Received: by stg.sel.sony.com id <170369-13991>; Mon, 12 May 1997 19:26:30 -0700 X-Mailer: Mail User's Shell (7.2.4 2/2/92) From: Brian Dockter To: zmailer@nic.funet.fi Subject: Fullnames on mail addresses. Message-Id: <97May12.192630pdt.170369-13991+1@stg.sel.sony.com> Date: Mon, 12 May 1997 19:26:23 -0700 Return-Path: I have need to change incoming/outgoing email address from being just the user name to the user's fullname (of the form First_Last). I have found the hook for the fullnames database used for incoming mail addresses (although I haven't yet figured out the magic to change it from first.last to first_last), but I haven't found the hooks for rewriting outgoing email addresses. Can someone give me some hints about what I should do? Thanks, Brian -- Brian Dockter (KC7JZL) | Email: Brian_Dockter@stg.sel.sony.com Sr. Software Engineer | Voice: 206-524-0014 Sony Electronics, Seattle | FAX: 206-524-3440 From zmailer-owner@nic.funet.fi Tue May 13 08:28:16 1997 Received: from root@acli.interlog.com (port 60944 [198.53.146.23]) by nic.funet.fi with ESMTP id <689-20446>; Tue, 13 May 1997 08:27:59 +0300 Received: from acli@localhost (port 49424 [127.0.0.1]) by acli.interlog.com with SMTP id <16191-12151>; Tue, 13 May 1997 01:28:12 +0000 Date: Tue, 13 May 1997 01:28:09 -0400 (EDT) From: Ambrose Li Sender: acli@acli.interlog.com To: zmailer@nic.funet.fi Subject: zmailer 2.99.48p2 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I just compiled 2.99.48p2 on my home machine, and noticed an oddity: smtpserver/policytest.c included some ndbm-specific stuff even though configure detected an absence of ndbm: policytest.c: In function `policydefine': policytest.c:76: union has no member named `_ndbm' policytest.c:96: union has no member named `_ndbm' make[1]: *** [policytest.o] Error 1 make[1]: Leaving directory `/mnt/source/zmailer-2.99.48/smtpserver' make: *** [all] Error 1 [/mnt/source/zmailer-2.99.48]2>acli@ada[191]> which is fixed by: --- smtpserver/policytest.c.dist Tue May 13 00:45:20 1997 +++ smtpserver/policytest.c Tue May 13 00:45:49 1997 @@ -73,7 +73,9 @@ return; } +#ifdef HAVE_NDBM_H rel->ndbm = NULL; /* Zero it -- well it is zeroed already.. */ +#endif switch (rel->dbt) { #ifdef HAVE_NDBM_H case _dbt_ndbm: @@ -93,10 +95,12 @@ default: break; } +#ifdef HAVE_NDBM_H if (rel->ndbm == NULL) { /* XX: ERROR! Could not open the database! */ return; } +#endif #ifdef HAVE_NDBM_H #endif -- Ambrose Li ~ acli@acli%interlog.com (Hello, world!\n) ~ http://www.interlog.com/~acli/~acli/ ~ print flush quit From zmailer-owner@nic.funet.fi Tue May 13 18:58:02 1997 Received: from NO-IDENT-SERVICE@aecgate.aec.ca (port 37645 [142.56.1.1]) by nic.funet.fi with ESMTP id <278-20446>; Tue, 13 May 1997 18:57:17 +0300 Received: from soltoo ([142.56.132.34]) by aecgate.aec.ca with ESMTP id <47086-10>; Tue, 13 May 1997 09:57:07 -0600 Received: by aec.ca id <5706-27467>; Tue, 13 May 1997 09:56:49 -0600 Sender: Trevor Paquette From: "Trevor Paquette" Message-Id: <970513095640.ZM340@sol132.aec.ca> Date: Tue, 13 May 1997 09:56:40 -0600 X-Mailer: Z-Mail (4.0.1 13Jan97) To: zmailer@nic.funet.fi Subject: URGENT: smtp session aborted.. 2.99.46p2 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-Path: HEre is the situation: clients using a mailhost running sendmail can send mail clients using a mailhost running zmailer 2.97 can send mail clients using a mailhost running zmailer 2.99.10/13 can send mail however when connecting to a server running zmailer 2.99.46p2 some hosts get: (What happens is that the client never send the HELO greeting. It looks like it never sees the SMTP greeting banner.. Has anyone seen this before?? 103# connection from IDENT-FD2-FAULT-1-32@karmrust.aec.ca (port 2035) 103w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Tue, 13 May 1997 09:19:26 -0600 103# remote from [142.56.113.79] 103# aborted: session terminated 100# connection from IDENT-FD2-FAULT-1-32@mtodesco2.aec.ca (port 2283) 100w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Tue, 13 May 1997 09:18:04 -0600 100# remote from [142.56.118.58] 100# aborted: session terminated 2071# connection from IDENT-FD2-FAULT-1-32@rkokotai.aec.ca (port 1327) 2071w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Mon, 12 May 1997 10:28:27 -0600 2071# remote from [142.56.118.101] 2071# aborted: session terminated -- Trevor Paquette | MetroNet Solutions |Work:(403)543-2355 TrevorPaquette@mcc.net | 2700, 421 7th Ave SW | Fax:(403)543-2854 http://www.mcc.net | Calgary, Ab, Canada |ICBM:51'05"N/114'01"W Senior Unix Network Architect| T2P 4K9 |Mind:In the Rockies From zmailer-owner@nic.funet.fi Tue May 13 20:53:17 1997 Received: by nic.funet.fi id <12953-23668>; Tue, 13 May 1997 20:52:00 +0300 Subject: Re: zmailer 2.99.48p2 From: Matti Aarnio To: acli@acli.interlog.com (Ambrose Li) Date: Tue, 13 May 1997 20:51:48 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Ambrose Li" at May 13, 97 01:28:09 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1334 Message-Id: <19970513175200Z12953-23668+255@nic.funet.fi> Return-Path: > I just compiled 2.99.48p2 on my home machine, and noticed an > oddity: > > smtpserver/policytest.c included some ndbm-specific stuff even > though configure detected an absence of ndbm: Oops. Right. My systems have always had NDBM, so I never noticed this one... The ZMailer is undergoing tortureous Ansification at my home machine (tortureous to myself...) and changes are likely so extensive that it will be called 2.99.49, though I did find a couple of call parametrization errors with it already, nice side-effect at that.. I have decided to let several things to do writes into "const char *" objects, because it is far easier to find those few places where it is ok, instead of allowing writes everywhere.. Are there compilers that will blow at this ? That is, are there compilers that will not allow lvalues to be const ? (Like in: "*(const char *foo) = 1;" ) > policytest.c: In function `policydefine': > policytest.c:76: union has no member named `_ndbm' > policytest.c:96: union has no member named `_ndbm' > make[1]: *** [policytest.o] Error 1 > make[1]: Leaving directory `/mnt/source/zmailer-2.99.48/smtpserver' > make: *** [all] Error 1 .... > -- > Ambrose Li ~ acli@acli%interlog.com (Hello, world!\n) > ~ http://www.interlog.com/~acli/~acli/ ~ print flush quit /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 13 21:39:41 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 51974 [13.1.64.93]) by nic.funet.fi with SMTP id <13261-13404>; Tue, 13 May 1997 21:38:18 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <19383(13)>; Tue, 13 May 1997 11:37:25 PDT X-NS-Transport-ID: 0800200AF21F3722219F Date: Tue, 13 May 1997 11:36:40 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: zmailer 2.99.48p2 In-Reply-to: "mea@nic.funet:fi:Xerox's message of Tue, 13 May 1997 10:51:48 PDT" To: mea@nic.funet.fi cc: acli@acli.interlog.com, zmailer@nic.funet.fi Message-Id: <97May13.113725pdt."19383(13)"@alpha.xerox.com> Return-Path: Yes! There are compilers that will not compile code that writes into const locations. This is what Sun's compiler (sunpro version 4.0) has to say: void xxx (char *foo) { *(const char *)foo = 1; } cc -o foo foo.c "foo.c", line 2: left operand must be modifiable lvalue: op "=" cc: acomp failed for foo.c *** Error code 2 gcc 2.7 issues a warning. From zmailer-owner@nic.funet.fi Tue May 13 23:47:11 1997 Received: from NO-IDENT-SERVICE@mercury.uniserve.com (port 24077 [204.191.197.248]) by nic.funet.fi with ESMTP id <13266-12056>; Tue, 13 May 1997 23:46:29 +0300 Received: from kst ([204.244.242.135]) by mercury.uniserve.com with ESMTP id NAA14034 for ; Tue, 13 May 1997 13:46:09 -0700 (PDT) Message-Id: <199705132046.NAA14034@mercury.uniserve.com> From: "Kevin Traas" To: "zmailer" Subject: Config Question Date: Tue, 13 May 1997 13:45:59 -0700 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: I'm new to ZMailer and have a question. If it's something covered in docs somewhere, just kindly point me in the right direction. Thanks. I've successfully compiled and installed zmailer-2.99.48p2 on my Debian GNU/Linux 1.2.x system. Mail is entering and leaving the system via SMTP normally. My question is: I've just reconfigured my LAN and DNS. The system that used to receive e-mail to my domain is being replaced by this Linux system running ZMailer. The problem I've run into is that for any mail coming to username@DN, errors occur if username doesn't exist on the Linux system. For ease, I'll call the Linux system A and the old system B. I now have DNS configured so that any mail addressed to username@DN is delivered to system A. I'd like to configure ZMailer to automatically forward any mail for which it can't resolve the username to system B. The reason for this is that I want to phase out system B. In reading somewhere, I came across "smarthost". Is this what I need to configure? If so, how? If not, then what? TIA for your help, Kevin Traas Systems Analyst Edmondson Roper CA http://www.eroper.bc.ca From zmailer-owner@nic.funet.fi Wed May 14 02:24:45 1997 Received: from root@bluestem.prairienet.org (port 34552 [192.17.3.4]) by nic.funet.fi with ESMTP id <12620-23513>; Wed, 14 May 1997 02:23:24 +0300 Received: from bozo (m65bmi.Dave-World.net [206.26.2.65]) by bluestem.prairienet.org (8.8.5/8.8.5) with SMTP id SAA08057 for ; Tue, 13 May 1997 18:19:00 -0500 (CDT) Message-Id: <199705132319.SAA08057@bluestem.prairienet.org> Comments: Authenticated sender is From: kunglao@prairienet.org To: zmailer@nic.funet.fi Date: Wed, 14 May 1997 06:22:37 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: Config Question Priority: normal In-reply-to: <199705132046.NAA14034@mercury.uniserve.com> X-mailer: Pegasus Mail for Win32 (v2.53/R1) Return-Path: > > I'm new to ZMailer and have a question. If it's something covered in docs > somewhere, just kindly point me in the right direction. Thanks. Ditto. > My question is: I've just reconfigured my LAN and DNS. The system that > used to receive e-mail to my domain is being replaced by this Linux system > running ZMailer. The problem I've run into is that for any mail coming to > username@DN, errors occur if username doesn't exist on the Linux system. > > For ease, I'll call the Linux system A and the old system B. > > I now have DNS configured so that any mail addressed to username@DN is > delivered to system A. I'd like to configure ZMailer to automatically > forward any mail for which it can't resolve the username to system B. The > reason for this is that I want to phase out system B. I have a similar situation, and I haven't gotten too far with the docs in the tar. I will have two hostnames defined (one's an alias) in my dns to point to my Linux box (with zmailer of course) connected to the Internet, say host1.some.thing.com and host2.some.thing.com. How I'm wanting to set up the mail routing is for zmailer to pick up anything addressed to host1.some.thing.com and some.thing.com, and deliver it to users on the Linux box (giving "user doesn't exist as appropriate). But then forward any mail addressed to host2.some.thing.com to a second SMTP server on my LAN, not dirrectly connected to the 'net (using IP masq'ing). Can I? Might there be the errors mentioned above if a user on host2 doesn't exist on host1? Many thanks. --- "Oh, what sad times are these when passing ruffians can say 'Ni!' at will to old ladies. There is a pestulance upon this land, nothing is sacred. Even those who arrange and design shrubberies are under considerable economic stress of this period of history." -- Roger the Shrubber From zmailer-owner@nic.funet.fi Wed May 14 14:18:59 1997 Received: by nic.funet.fi id <203-23513>; Wed, 14 May 1997 14:18:08 +0300 Subject: Re: zmailer 2.99.48p2 From: Matti Aarnio To: Nicholas_Briggs.PARC@xerox.com Date: Wed, 14 May 1997 14:18:06 +0300 (EET DST) Cc: acli@acli.interlog.com, zmailer@nic.funet.fi In-Reply-To: <97May13.113725pdt."19383(13)"@alpha.xerox.com> from "Nicholas_Briggs.PARC@xerox.com" at May 13, 97 11:36:40 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 759 Message-Id: <19970514111808Z203-23513+189@nic.funet.fi> Return-Path: > > Yes! There are compilers that will not compile code that writes into > const locations. > > This is what Sun's compiler (sunpro version 4.0) has to say: ... How about doing following (either of the two ways): const char *foocon; void xxxtest () { char *foowrite; /* 1 */ foowrite = (char *)foocon; /* 1 */ *foowrite = 1; /* 1 */ *(char *)foocon = 1; /* 2 */ } If either works, we can survive without massive rewrite. GCC warns about both of these cases. My test a moment ago at the university sparc gave a license error with Sun acomp ... Weird. > cc -o foo foo.c > "foo.c", line 2: left operand must be modifiable lvalue: op "=" > cc: acomp failed for foo.c > *** Error code 2 > > gcc 2.7 issues a warning. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed May 14 16:03:13 1997 Received: from NO-IDENT-SERVICE@mailrelay.tiac.net (port 17924 [199.0.65.237]) by nic.funet.fi with ESMTP id <13254-23513>; Wed, 14 May 1997 16:02:10 +0300 Received: from draco (draco.alphapower.com [206.119.34.99]) by mailrelay.tiac.net (8.8.5/) with SMTP id JAA19671 for ; Wed, 14 May 1997 09:02:00 -0400 (EDT) Message-Id: <199705141302.JAA19671@mailrelay.tiac.net> Comments: Authenticated sender is From: "Gerard Hickey" Organization: Nekotech, A Division of Inco, Inc. To: zmailer@nic.funet.fi Date: Wed, 14 May 1997 09:01:49 -5 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Re: Config Question Reply-to: hickey@alphapower.com Priority: normal X-mailer: Pegasus Mail for Win32 (v2.42a) Return-Path: > > My question is: I've just reconfigured my LAN and DNS. The system that > > used to receive e-mail to my domain is being replaced by this Linux system > > running ZMailer. The problem I've run into is that for any mail coming to > > username@DN, errors occur if username doesn't exist on the Linux system. > > > > For ease, I'll call the Linux system A and the old system B. > > > > I now have DNS configured so that any mail addressed to username@DN is > > delivered to system A. I'd like to configure ZMailer to automatically > > forward any mail for which it can't resolve the username to system B. The > > reason for this is that I want to phase out system B. There is a very easy answer for this scenerio. Use the aliases file. For all the users that are on system B, you will need to add an alias to route their mail over to system B. You would just add a line like the following for each user on B. nothere: nothere@B Once all the users are there, you need to do a "zmailer newaliases" to activate the new aliases map. -- Gerard Hickey hickey@alphapower.com Nekotech Technical Services Voice: +1 603 926 0300 102 Tide Mill Rd. Fax: +1 603 926 0301 Hampton, NH 03842 Nekotech is a provider of high end servers based on the Digital Alpha chipset. Nekotech's home page can be reached at http://www.alphapower.com/ From zmailer-owner@nic.funet.fi Wed May 14 18:51:33 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 14525 [194.67.3.135]) by nic.funet.fi with ESMTP id <13029-21309>; Wed, 14 May 1997 18:50:50 +0300 Received: from NO-IDENT-SERVICE@ariel (port 44984 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68045-316>; Wed, 14 May 1997 19:50:41 +0400 Date: Wed, 14 May 1997 19:50:38 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: zmailer 2.99.48p2 To: Matti Aarnio cc: Nicholas_Briggs.PARC@xerox.com, acli@acli.interlog.com, zmailer@nic.funet.fi In-Reply-To: <19970514111808Z203-23513+189@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > ... How about doing following (either of the two ways): > > > const char *foocon; > void xxxtest () { > char *foowrite; /* 1 */ > foowrite = (char *)foocon; /* 1 */ > *foowrite = 1; /* 1 */ > *(char *)foocon = 1; /* 2 */ > } Funny but neither Sparkworks nor gcc gives even any warning: crosser@chronos:/tmp/$ gcc --version 2.7.2 crosser@chronos:/tmp/$ cc -V cc: SC3.0.1 13 Jul 1994 usage: cc [ options] files. Use 'cc -flags' for details crosser@chronos:/tmp/$ gcc -Wall -c x.c crosser@chronos:/tmp/$ cc -c x.c crosser@chronos:/tmp/$ cat x.c const char *foocon; void xxxtest () { char *foowrite; /* 1 */ foowrite = (char *)foocon; /* 1 */ *foowrite = 1; /* 1 */ *(char *)foocon = 1; /* 2 */ } crosser@chronos:/tmp/$ From zmailer-owner@nic.funet.fi Wed May 14 19:52:36 1997 Received: by nic.funet.fi id <13065-21309>; Wed, 14 May 1997 19:51:39 +0300 Subject: Re: zmailer 2.99.48p2 From: Matti Aarnio To: crosser@online.ru Date: Wed, 14 May 1997 19:51:37 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 14, 97 07:50:38 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 936 Message-Id: <19970514165139Z13065-21309+237@nic.funet.fi> Return-Path: > Funny but neither Sparkworks nor gcc gives even any warning: I used invocations: sol:~$ gcc ..(about all -W switches that 'man gcc' lists) .. -O -g -c try.c sol:~$ cc -O -g -c try.c "try.c", line 3: warning: assignment type mismatch: pointer to char "=" pointer to const char sol:~$ cc -V cc: SC3.0.1 13 Jul 1994 Back to home to make sure I get warnings about these potentially bad things, but not rejections... > crosser@chronos:/tmp/$ gcc --version > 2.7.2 > crosser@chronos:/tmp/$ cc -V > cc: SC3.0.1 13 Jul 1994 > usage: cc [ options] files. Use 'cc -flags' for details > crosser@chronos:/tmp/$ gcc -Wall -c x.c > crosser@chronos:/tmp/$ cc -c x.c > crosser@chronos:/tmp/$ cat x.c > const char *foocon; > void xxxtest () { > char *foowrite; /* 1 */ > foowrite = (char *)foocon; /* 1 */ > *foowrite = 1; /* 1 */ > *(char *)foocon = 1; /* 2 */ > } /Matti Aarnio From zmailer-owner@nic.funet.fi Thu May 15 21:14:26 1997 Received: by nic.funet.fi id <985-6279>; Thu, 15 May 1997 21:13:08 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49 beta 1 at my workstation.. Message-Id: <19970515181308Z985-6279+354@nic.funet.fi> Date: Thu, 15 May 1997 21:12:59 +0300 Return-Path: Ok, an extremely raw "pre-release" of 2.99.49 is now loadable from ftp://mea.tmt.tele.fi/ This one contains massive ansification, which should make several compilers more happy about. Also there are a lot of places where (u_char *) has been changed to (char *) -- all over in fact, and lots of "const"s have been added also all over. It does work properly with my weak router regression test, though. /Matti Aarnio From zmailer-owner@nic.funet.fi Thu May 15 22:42:58 1997 Received: from NO-IDENT-SERVICE@press-gopher.uchicago.edu (port 8621 [128.135.234.36]) by nic.funet.fi with ESMTP id <410-6279>; Thu, 15 May 1997 22:41:50 +0300 Received: from rcb@localhost (fake: uid#100@localhost) by press-gopher.uchicago.edu with SMTP id <26584-202>; Thu, 15 May 1997 14:41:36 -0500 Date: Thu, 15 May 1997 14:41:29 -0500 (CDT) Sender: Roy C Bixler From: Roy C Bixler To: zmailer@nic.funet.fi Subject: duplicate email from Apple/Eudora Internet servers In-Reply-To: <19970515181308Z985-6279+354@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi, I am wondering if there is anyone else here who has gotten massive duplicated email from a site running Apple Internet server or Eudora Internet server. The offending sites connect once every 15 to 30 minutes and submit the same message as on previous connections. Right now, I am getting around this problem by outfitting victim users with a simple 'procmail' filter to quietly sink the duplicates. But I am wondering if there is any other solution available besides this, complaining the postmasters of the remote site and complaining to Eudora and Apple. Here is a sample SMTP conversation below: 16312w 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT r eady at Thu, 15 May 1997 14:32:45 -0500 16312# remote from [206.126.246.3] 16312r EHLO mail.ness.org 16312w 250-press-gopher.uchicago.edu expected "EHLO woz.ness.org" 16312w 250-SIZE 0 16312w 250-8BITMIME 16312w 250-PIPELINING 16312w 250-CHUNKING 16312w 250-ENHANCEDSTATUSCODES 16312w 250-EXPN 16312w 250-VRFY 16312w 250-DSN 16312w 250-X-RCPTLIMIT 10000 16312w 250-ETRN 16312w 250 HELP 16312r MAIL FROM: SIZE=2522 16312w 250 2.1.0 Sender syntax Ok 16312# -- pipeline input exists 34 bytes 16312r RCPT TO: 16312w 250 2.1.5 Ok; can accomodate 2522 byte message 16312r BDAT 2522 LAST 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool Any ideas will be much appreciated! Thanks, Roy Bixler The University of Chicago Press postmaster rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Fri May 16 04:41:52 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 31763 [13.1.64.93]) by nic.funet.fi with SMTP id <326-25621>; Fri, 16 May 1997 04:41:16 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <18044(1)>; Thu, 15 May 1997 18:40:41 PDT X-NS-Transport-ID: 0800200AF21F37222338 Date: Thu, 15 May 1997 18:40:00 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-to: "mea@nic.funet:fi:Xerox's message of Tue, 6 May 1997 03:20:59 PDT" To: mea@nic.funet.fi cc: andy@realbig.com, zmailer@nic.funet.fi Message-Id: <97May15.184041pdt."18044(1)"@alpha.xerox.com> Return-Path: Now oddly enough, I saw this same symptom (sitting in an accept, but no connections get established) for the first time today, with zmailer 2.2e10, running on a SPARC 5 with SunOS 4.1.3_U1 -- the smtpserver binary hasn't changed, the OS hasn't changed, and everything had been working fine for 30+ days before this. Killing and restarting the smtpserver process made things work for about 15-20m, and then it happened again. Rebooting the system, it hasn't happened in the 3 hours it has been up. I'm a little worried about this... \nick From zmailer-owner@nic.funet.fi Fri May 16 12:40:51 1997 Received: by nic.funet.fi id <847-10549>; Fri, 16 May 1997 12:39:58 +0300 Subject: Re: 2.99.48 smtpserver won't accept()? From: Matti Aarnio To: Nicholas_Briggs.PARC@xerox.com Date: Fri, 16 May 1997 12:39:46 +0300 (EET DST) Cc: andy@realbig.com, zmailer@nic.funet.fi In-Reply-To: <97May15.184041pdt."18044(1)"@alpha.xerox.com> from "Nicholas_Briggs.PARC@xerox.com" at May 15, 97 06:40:00 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2068 Message-Id: <19970516093958Z847-10549+141@nic.funet.fi> Return-Path: > Now oddly enough, I saw this same symptom (sitting in an accept, but no > connections get established) for the first time today, with zmailer 2.2e10, > running on a SPARC 5 with SunOS 4.1.3_U1 -- the smtpserver binary hasn't > changed, the OS hasn't changed, and everything had been working fine for 30+ > days before this. > > Killing and restarting the smtpserver process made things work for about > 15-20m, and then it happened again. > > Rebooting the system, it hasn't happened in the 3 hours it has been up. > > I'm a little worried about this... Doing 'netstat' would show you many sockets for the local smtp port in SYN_RCVD state. There is NOTHING we can do to it (in the user-space) at the older operating system releases. This situation is sometimes called "SYN-Attack" (when maliciously caused), or it can be caused by routing problem when our ACKs to the incoming SYNs don't reach the connect()or... Killing the smtpserver throws away those nascend sockets, and thus the new queue at the start of the new smtpserver will be clean.. At SunOS 4.1.x the network code is deeply BSD, and I think you can find the routine responsible for SOMAXCON macro comparison, and change that fixed value in the binary... (See G.R.Wright,W.R.Stevens: TCP/IP Illustrated, Volume2, "The Implementation", page 455, "solisten() function") I did code listen() with HIGH limit, which can help you in similar case, provided the system does not limit the SO_MAXCON lower... At lattest Linuxes (2.0.30+, 2.1.30+) the system is extremely SYN resilient, so appear to be with Solaris 2.5.1 (with patch), HP-UX 9.x/10.x (with patch), DEC UNIX 4.0, new FreeBSD/ NetBSD codes, etc. Above I say 'resilient', I don't believe any of the methods is 100% SYN-attack-proof, but presuming certain lowish maximum on how quickly the SYNs are coming, system can be made to tolerate it. ( If any of them is proofed against 100Mbps ether full of fake SYNs, then it is SYN-attack-proof, but not before ;-) ) > \nick /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 16 14:13:27 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 21986 [194.67.3.135]) by nic.funet.fi with ESMTP id <840-26884>; Fri, 16 May 1997 14:12:32 +0300 Received: from NO-IDENT-SERVICE@ariel (port 49148 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-316>; Fri, 16 May 1997 15:12:22 +0400 Date: Fri, 16 May 1997 15:12:20 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49 beta 1 at my workstation.. To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970515181308Z985-6279+354@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > Ok, an extremely raw "pre-release" of 2.99.49 is now loadable > from ftp://mea.tmt.tele.fi/ Solaris 2.5.1: gcc -g -O2 -DHAVE_CONFIG_H -I../.. -I../../include -I../../libsh -c bind.c bind.c: In function `getrrtypec': bind.c:714: `AF_INET' undeclared (first use this function) bind.c:714: (Each undeclared identifier is reported only once bind.c:714: for each function it appears in.) *** Error code 1 looks like sys/socket.h was not #included? Eugene From zmailer-owner@nic.funet.fi Fri May 16 16:39:02 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 48655 [13.1.64.93]) by nic.funet.fi with SMTP id <75-26884>; Fri, 16 May 1997 16:38:42 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <18511(8)>; Fri, 16 May 1997 06:38:21 PDT X-NS-Transport-ID: 0800200AF21F37222382 Date: Fri, 16 May 1997 06:38:08 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: 2.99.48 smtpserver won't accept()? In-Reply-to: "mea@nic.funet:fi:Xerox's message of Fri, 16 May 1997 02:39:46 PDT" To: mea@nic.funet.fi cc: Nicholas_Briggs.PARC@xerox.com, andy@realbig.com, zmailer@nic.funet.fi Message-Id: <97May16.063821pdt."18511(8)"@alpha.xerox.com> Return-Path: There was no SYN-attack going on at the time. I am going to up the listen queue size, however. From zmailer-owner@nic.funet.fi Fri May 16 20:44:05 1997 Received: from NO-IDENT-SERVICE@mercury.uniserve.com (port 27653 [204.191.197.248]) by nic.funet.fi with ESMTP id <422-16416>; Fri, 16 May 1997 20:43:27 +0300 Received: from kst ([204.244.242.135]) by mercury.uniserve.com with ESMTP id KAA02400 for ; Fri, 16 May 1997 10:42:44 -0700 (PDT) Message-Id: <199705161742.KAA02400@mercury.uniserve.com> From: "Kevin Traas" To: "zmailer" Subject: Looking for a quick answer Date: Thu, 1 May 1997 10:42:22 -0700 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: I've just installed zmailer (2.99.48p2) a few days ago and, for the first bit, everything was working fine. SMTP transfers were working just fine. Incoming and outgoing mail was working fine.... But I "broke" it.... I'm now getting the following error message from zmailer for every message received: (Note, domain name is replaced with $DN to "protect the innocent"... ....) : Trying to talk with myself! Of course, with hindsight, I should have logged what changes I was making to the config; but, of course, I haven't, so I haven't been able to backtrack very effectively to figure out what I changed to mess things up. I should also mention that I'm new to ZMailer and this is my first go at getting it set up on a system. If you have any quick ideas on what I might have done to cause this, please let me know. I'd appreciate any suggestions I can get. (One thing I've noted from the message above is that it's indicating an smtp route; however, the recipient address is local.... ??? ... ) Have I got routing screwed up somehow? If so, where do I go to fix it? Thanks, Kevin Traas Systems Analyst Edmondson Roper CA http://www.eroper.bc.ca From zmailer-owner@nic.funet.fi Sun May 18 00:06:48 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 25673 [128.214.46.64]) by nic.funet.fi with ESMTP id <12968-10485>; Sun, 18 May 1997 00:06:21 +0300 Received: from jethro.fuller.edu (mail@jethro.fuller.edu [206.1.27.7]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id AAA19752 for ; Sun, 18 May 1997 00:04:33 +0300 (EET DST) Received: from waterf.org [206.1.27.36] (mail) by jethro.fuller.edu with smtp (Exim 1.62 #1) id 0wSpIs-0002k1-00 (Debian); Sat, 17 May 1997 12:38:07 -0700 Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wSpIG-0002co-00 (Debian); Sat, 17 May 1997 12:37:28 -0700 Date: Sat, 17 May 1997 12:37:27 -0700 (PDT) From: Christoph Lameter To: zmailer Subject: 2.99.48.2 Bounces sent to postoffice instead of List In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: The bounces from Linux Mailing Lists at vger should go back to vger.rutgers.edu. I seem to be getting them(?). Can I switch those off? >From MAILER-DAEMON Sat May 17 12:28:54 1997 Received: by server.i-m-f.org id <387651-6943>; Sat, 17 May 1997 12:28:21 +0000 To: owner-linux-admin-outgoing@vger.rutgers.edu From: The Post Office Sender: mailer-daemon Subject: Delivery reports about your email Cc: The Post Office MIME-Version: 1.0 Content-Type: multipart/report; report-type=delivery-status; boundary="A100B2B45.863897315=_/server.i-m-f.org" Message-Id: <19970517192821Z387651-6943+26@server.i-m-f.org> Date: Sat, 17 May 1997 12:28:15 +0000 Return-Path: <> --A100B2B45.863897315=_/server.i-m-f.org Content-Type: text/plain This is a collection of reports about email delivery process concerning a message you originated: : ...\ <<- RCPT To: ORCPT=rfc822;dvanecek@third-wave.com ->> 550 ... User unknown --A100B2B45.863897315=_/server.i-m-f.org Content-Type: message/delivery-status Reporting-MTA: dns; server.i-m-f.org --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Sun May 18 04:56:39 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 45391 [128.214.46.64]) by nic.funet.fi with ESMTP id <625-29587>; Sun, 18 May 1997 04:55:49 +0300 Received: from jethro.fuller.edu (mail@jethro.fuller.edu [206.1.27.7]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id EAA22004 for ; Sun, 18 May 1997 04:47:53 +0300 (EET DST) Received: from waterf.org [206.1.27.36] (mail) by jethro.fuller.edu with smtp (Exim 1.62 #1) id 0wSpJB-0002k6-00 (Debian); Sat, 17 May 1997 12:38:25 -0700 Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wSpDV-0002cG-00 (Debian); Sat, 17 May 1997 12:32:33 -0700 Date: Sat, 17 May 1997 12:32:32 -0700 (PDT) From: Christoph Lameter cc: zmailer Subject: 2.99.48.2 piping, Qs (New to Zmail..), Debian Package In-Reply-To: <199705161742.KAA02400@mercury.uniserve.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: unlisted-recipients:; (no To-header on input) Return-Path: I have started using Zmailer to run an exploder for the Linux Mailing lists with a daily throughput of 30.000 - 100.000 messages (.com domain only right now) on a Cyrix 686 166, 16M Ram, Debian 1.3, Linux 2.0.30. Used to run exim but exim got flaky on days with more than 50.000 messages and got fits managing huge message queues. -- Pipes -- The following entries in /etc/aliases does not work (this is BeroList a simple List Server used for some local Linux unrelated Lists). #-- berolist begin news:"|/usr/sbin/list news" talk:"|/usr/sbin/list talk" admin:"|/usr/sbin/list admin" #-- berolist end I have gotten Signal 11 and returncode 139 in the Queue Logs. I can run the list program manually just fine and feed the message that would be piped to it into stdin. One issue might be that berolist does produce output to show the progress of message delivery on stdout. Is that a problem for Zmailer? Why does Zmailer not log the output or sent it back to the sender? -- Queue Display -- Where can I find documentation what thes HQ QA etc abbreviations in the Queue display mean? I really like the output of "mailq -Q" but it would be better if I knew the details. -- Forcing message delivery --- How can I force the delivery of a message immediately and see whats going on? -- Load limitation ? --- Is there some way to tell Zmailer to stop beginning new deliveries at a certain load average? I notice that the load is usually around 0.5 but sometimes there are spikes up to load of 8 or more when vger.rutgers.edu starts transmitting after being down for awhile. Would be nice if the load on the system is stable so that the responsetimes are reasonable and consistent even when overflooded. -- Daily Reports ? --- Exim had a nice daily status report. Does something similar exist for Zmailer? -- Binary Package for Linux -- I packaged Zmailer as a debian package. Its available from ftp://lalug.org/debian_local (Hope that is ok?) --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Sun May 18 05:02:00 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 49743 [128.214.46.64]) by nic.funet.fi with ESMTP id <904-10485>; Sun, 18 May 1997 05:01:34 +0300 Received: from jethro.fuller.edu (mail@jethro.fuller.edu [206.1.27.7]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id EAA22074 for ; Sun, 18 May 1997 04:56:47 +0300 (EET DST) Received: from waterf.org [206.1.27.36] (mail) by jethro.fuller.edu with smtp (Exim 1.62 #1) id 0wSoZa-0002dL-00 (Debian); Sat, 17 May 1997 11:51:19 -0700 Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wSoQX-0002Wq-00 (Debian); Sat, 17 May 1997 11:41:57 -0700 Date: Sat, 17 May 1997 11:41:56 -0700 (PDT) From: Christoph Lameter To: Remco van de Meent cc: zmailer@nic.funet.fi Subject: Re: zmailer on Debian systems In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I have Zmailer debianized. You get get a .deb file from ftp://lalug.org/debian_local. On Tue, 22 Apr 1997, Remco van de Meent wrote: >Hey > > >I'm trying to install ZMailer 2.99.46 on my Debian 1.2.9 box. > > >David Trueman posted a message about the router segfaulting on his Debian >box at Thu, 6 Feb 1997, using .45. There was a reply on that message which stated >that the problem should be resolved in .46. But I got those segfauls even >using .46 :( > >I think the cause of the problem is in the 'strange' *db*.h files and >libraries, which come with Debian. > >So, my question is: is anyone using ZMailer on a Debian machine >successfully? Or maybe anyone has a solution, or some ideas about how to >solve it? > > >Thanks! > > >bye -R. > >// Remco van de Meent >// email: remco@oloon.student.utwente.nl >// www: http://oloon.student.utwente.nl >// " Never make any mistaeks. " > --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Sun May 18 05:20:20 1997 Received: from NO-IDENT-SERVICE@trixie.mse.jhu.edu (port 47883 [128.220.8.14]) by nic.funet.fi with ESMTP id <12927-18129>; Sun, 18 May 1997 05:19:21 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5370>; Sat, 17 May 1997 22:18:55 -0400 Date: Sat, 17 May 1997 22:18:26 -0400 (Eastern Daylight Time) From: Tim DiLauro To: Kevin Traas cc: zmailer Subject: Re: Looking for a quick answer In-Reply-To: <199705161742.KAA02400@mercury.uniserve.com> Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 1 May 1997, Kevin Traas wrote: > But I "broke" it.... I'm now getting the following error message from > zmailer for every message received: (Note, domain name is replaced with > $DN to "protect the innocent"... ....) > > : Trying to talk with myself! Kevin: Did you make changes to the db/localnames file. If so, it needs to be sorted in ascending order to work properly. -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Mon May 19 19:04:41 1997 Received: by nic.funet.fi id <1030-29324>; Mon, 19 May 1997 19:03:28 +0300 Subject: Re: 2.99.48.2 Bounces sent to postoffice instead of List From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Mon, 19 May 1997 19:03:25 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 17, 97 12:37:27 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 831 Message-Id: <19970519160328Z1030-29324+281@nic.funet.fi> Return-Path: > The bounces from Linux Mailing Lists at vger should go back to > vger.rutgers.edu. I seem to be getting them(?). Can I switch those off? You are getting "Cc: postoffice" messages. Edit $MAILSHARE/forms/ -files > >From MAILER-DAEMON Sat May 17 12:28:54 1997 > Received: by server.i-m-f.org id <387651-6943>; Sat, 17 May 1997 12:28:21 > +0000 > To: owner-linux-admin-outgoing@vger.rutgers.edu > From: The Post Office > Sender: mailer-daemon > Subject: Delivery reports about your email > Cc: The Post Office > MIME-Version: 1.0 > Content-Type: multipart/report; report-type=delivery-status; > boundary="A100B2B45.863897315=_/server.i-m-f.org" > Message-Id: <19970517192821Z387651-6943+26@server.i-m-f.org> > Date: Sat, 17 May 1997 12:28:15 +0000 > Return-Path: <> /Matti Aarnio From zmailer-owner@nic.funet.fi Mon May 19 19:40:15 1997 Received: by nic.funet.fi id <1604-30669>; Mon, 19 May 1997 19:38:57 +0300 Subject: Re: 2.99.48.2 piping, Qs (New to Zmail..), Debian Package From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Mon, 19 May 1997 19:38:53 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 17, 97 12:32:32 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3965 Message-Id: <19970519163857Z1604-30669+256@nic.funet.fi> Return-Path: > I have started using Zmailer to run an exploder for the Linux Mailing > lists with a daily throughput of 30.000 - 100.000 messages (.com domain > only right now) on a Cyrix 686 166, 16M Ram, Debian 1.3, Linux 2.0.30. > > Used to run exim but exim got flaky on days with more than 50.000 > messages and got fits managing huge message queues. A task that may become laborous for ZMailers too, but not THAT difficult as to most other MTAs... :-) > -- Pipes -- > > The following entries in /etc/aliases does not work (this is BeroList a > simple List Server used for some local Linux unrelated Lists). > > #-- berolist begin > news:"|/usr/sbin/list news" > talk:"|/usr/sbin/list talk" > admin:"|/usr/sbin/list admin" > #-- berolist end > > I have gotten Signal 11 and returncode 139 in the Queue Logs. > I can run the list program manually just fine and feed the message that > would be piped to it into stdin. > > One issue might be that berolist does produce output to show the progress > of message delivery on stdout. Is that a problem for Zmailer? Why does > Zmailer not log the output or sent it back to the sender? ZMailer reads ONE line of that output, and does something with it, but it does not handle multiple lines of the piped process output.. Picking more lines of the output, and sending it to the error address (which is what ?) is thinkable. > -- Queue Display -- > > Where can I find documentation what thes HQ QA etc abbreviations in the > Queue display mean? I really like the output of "mailq -Q" but it would be > better if I knew the details. smtp/aol.com/0 R=1 A=147 P=19598 HA=571s FA=571s OF=1 QA=1d18h smtp/some.com/0 R=1 A=58 W=1860s QA=11h11m28s "From the Source" is not acceptable answer ? Ok, 'man mailq' should have it, but I haven't gotten around, so here is the short version: R = Recipients for this channel/host thread A = Attempts on delivering this P = TransportAgent PID HA = '#hungry age' -- how long since saw '#hungry' state from TA FA = Feed-Age -- how long since last time did feed the TA OF = OverFeed count -- how many jobs were fed, and are unprocessed W = Wait time until next attempt QA = QueueAge -- age of the oldest message in the thread > -- Forcing message delivery --- > > How can I force the delivery of a message immediately and see whats going > on? With SMTP by issueing command: ETRN some.com (for above examples..) > -- Load limitation ? --- > > Is there some way to tell Zmailer to stop beginning new deliveries at a > certain load average? I notice that the load is usually around 0.5 but > sometimes there are spikes up to load of 8 or more when vger.rutgers.edu > starts transmitting after being down for awhile. > Would be nice if the load on the system is stable so that the > responsetimes are reasonable and consistent even when overflooded. The scheduler does not have load-average testing function, but the 'mytime()' daemon could do the load-average extraction from the system too.. (That is, it should not be done 10 000 times because each new recipient assimilation is potential place to start a transporter, when there are that many recipients (or more!) in the queue...) I don't like it, as there are very few things that are more non-standard than load-average extraction in various systems.. > -- Daily Reports ? --- > > Exim had a nice daily status report. Does something similar exist for > Zmailer? I know several scripts in existence, and now ZMailer produces pretty much similar syslogs to that of the sendmail. Thus sendmail's scripts are fairly easily modifiable.. Could you send example of EXIMs report ? Maybe something similar can be created. > -- Binary Package for Linux -- > > I packaged Zmailer as a debian package. Its available from > ftp://lalug.org/debian_local > (Hope that is ok?) Yes. > --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- /Matti Aarnio From zmailer-owner@nic.funet.fi Mon May 19 20:18:41 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 57724 [128.214.46.64]) by nic.funet.fi with ESMTP id <2016-11613>; Mon, 19 May 1997 20:17:45 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id UAA09547; Mon, 19 May 1997 20:17:34 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTW42-0004Wq-00 (Debian); Mon, 19 May 1997 10:17:38 -0700 Date: Mon, 19 May 1997 10:17:36 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Debian/Linux Diffs In-Reply-To: <19970519163857Z1604-30669+256@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: >> Used to run exim but exim got flaky on days with more than 50.000 >> messages and got fits managing huge message queues. > > A task that may become laborous for ZMailers too, but not THAT > difficult as to most other MTAs... :-) I have gotten around 60.000 messages per day in the last days and Zmail handled it fine. >> Where can I find documentation what thes HQ QA etc abbreviations in the >> Queue display mean? I really like the output of "mailq -Q" but it would be >> better if I knew the details. > "From the Source" is not acceptable answer ? > Ok, 'man mailq' should have it, but I haven't gotten around, so > here is the short version: I will sent diffs to you that change the mailq manpage. >> How can I force the delivery of a message immediately and see whats going >> on? > > With SMTP by issueing command: > > ETRN some.com > > (for above examples..) By telnetting to port 25??? Danger danger... >> Exim had a nice daily status report. Does something similar exist for >> Zmailer? > > I know several scripts in existence, and now ZMailer produces pretty > much similar syslogs to that of the sendmail. Thus sendmail's scripts > are fairly easily modifiable.. The diffs include a "zmailstats" program which is the reworked exim status program. >> I packaged Zmailer as a debian package. Its available from >> ftp://lalug.org/debian_local >> (Hope that is ok?) > > Yes. Ok. Zmail will be released into the Debian Experimental distribution in an hour or so. Also available now from ftp.fuller.edu:/Linux Example zmailstats output: Zmailer statistics from 1997-05-18 06:51:57 to 1997-05-19 07:14:20 Grand total summary ------------------- At least one address TOTAL Volume Messages Hosts Delayed Failed Received 6033KB 686 3 10 1.5% 10 1.5% Delivered 1032MB 59612 1380 Deliveries by transport ----------------------- Volume Messages smtp 1032MB 59612 Messages received per hour (each dot is 5 messages) --------------------------------------------------- 00-01 6 . 01-02 1 02-03 15 ... 03-04 54 .......... 04-05 53 .......... 05-06 47 ......... 06-07 231 .............................................. 07-08 25 ..... 08-09 16 ... 09-10 27 ..... 10-11 18 ... 11-12 13 .. 12-13 23 .... 13-14 22 .... 14-15 7 . 15-16 18 ... 16-17 21 .... 17-18 22 .... 18-19 11 .. 19-20 0 20-21 39 ....... 21-22 7 . 22-23 7 . 23-24 3 Deliveries per hour (each dot is 130 deliveries) ------------------------------------------------ 00-01 1470 ........... 01-02 464 ... 02-03 1747 ............. 03-04 6484 ................................................. 04-05 6004 .............................................. 05-06 5438 ......................................... 06-07 3824 ............................. 07-08 3545 ........................... 08-09 774 ..... 09-10 3154 ........................ 10-11 3084 ....................... 11-12 2056 ............... 12-13 2641 .................... 13-14 2139 ................ 14-15 739 ..... 15-16 4323 ................................. 16-17 3908 .............................. 17-18 1588 ............ 18-19 804 ...... 19-20 104 20-21 1575 ............ 21-22 2428 .................. 22-23 1167 ........ 23-24 152 . Time spent on the queue: all messages ------------------------------------- Under 1m 2 20.0% 20.0% 5m 3 30.0% 50.0% 15m 3 30.0% 80.0% 30m 2 20.0% 100.0% Time spent on the queue: messages with at least one remote delivery ------------------------------------------------------------------- Under 1m 2 20.0% 20.0% 5m 3 30.0% 50.0% 15m 3 30.0% 80.0% 30m 2 20.0% 100.0% Relayed messages ---------------- 3435 vger.rutgers.edu => mail.hotmail.com 2970 vger.rutgers.edu => future.futsoft.com 1358 vger.rutgers.edu => mx1.pobox.com ... Total: 59566 (plus 0 unshown) Top 50 sending hosts by message count ------------------------------------- 485 5491664 vger.rutgers.edu 199 684033 local 2 1597 hur-s0.fuller.edu Top 50 sending hosts by volume ------------------------------ 485 5491664 vger.rutgers.edu 199 684033 local 2 1597 hur-s0.fuller.edu Top 50 local senders by message count ------------------------------------- 150 553512 root 48 129697 daemon 1 824 list Top 50 local senders by volume ------------------------------ 150 553512 root 48 129697 daemon 1 824 list Top 50 destinations by message count ------------------------------------ 3435 44768589 mail.hotmail.com 2970 30983663 future.futsoft.com 1358 24588644 mx1.pobox.com 825 12540415 mail.geocities.com ... Top 50 destinations by volume ----------------------------- 3435 44768589 mail.hotmail.com 2970 30983663 future.futsoft.com 1358 24588644 mx1.pobox.com ... Top 50 local destinations by message count ------------------------------------------ Top 50 local destinations by volume ----------------------------------- List of errors -------------- 180 ^M<<- .^M->> 554 [ Arg list too long ] Input line to submit longer than 254 character 33 ^M<<- .^M->> 554 [ No such file or directory ] Input line to submit longer than 254 character 16 ^M<<- RCPT To: ORCPT=rfc822;TCHJR@JUNO.COM^M->> 552 ... Mail quota exceede 193 ^M<<- RCPT To: ORCPT=rfc822;jcc2@juno.com^M->> 552 ... Mail quota exceede 34 smtp; 500 (DNS: no such domain: jo.hackery.com ... Errors encountered: 681 ----------------------- --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Mon May 19 20:25:41 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 3197 [128.214.46.64]) by nic.funet.fi with ESMTP id <1322-29324>; Mon, 19 May 1997 20:24:45 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id UAA09612; Mon, 19 May 1997 20:24:38 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTWB9-0004ep-00 (Debian); Mon, 19 May 1997 10:24:59 -0700 Date: Mon, 19 May 1997 10:24:59 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Things to be done before a stable relase In-Reply-To: <19970519163857Z1604-30669+256@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Having packaged zmailer for Debian I saw the following issue that I think need dealing with before a release: 1. Make gnu autoconf fully functional. Installing in a different subtree does not work. 2. Consistent cleanup in all directories. make distclean does not remove all junk. 3. Get rid of obsolete stuff and streamline the sourcepackage. Most of the scripts supplied are obsolete for example. 4. Complete the documentation. 5. The zmailer.conf needs to be organized to either contain compile time values or runtime configuration values not both. 6. A good generic example setup is missing. Had to generate some on my own for the Debian distribution. --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Mon May 19 20:31:17 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 11645 [128.214.46.64]) by nic.funet.fi with ESMTP id <1543-11613>; Mon, 19 May 1997 20:30:29 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id UAA09656; Mon, 19 May 1997 20:30:25 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTWGc-0004wQ-00 (Debian); Mon, 19 May 1997 10:30:38 -0700 Date: Mon, 19 May 1997 10:30:36 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Questions re Zmail In-Reply-To: <19970519163857Z1604-30669+256@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: 1. What is this "Timeserver" and why does it garble the ps line? 11831 ? S 0:00 /usr/lib/zmailer/router -dkn 4 11836 ? S 0:04 /usr/lib/zmailer/scheduler -l /var/log/mail/scheduler.perflog -S -H 11837 ? S 0:00 TimeServer ailer/scheduler -l /var/log/mail/scheduler.perflog -S -H 11838 ? S 0:00 /usr/lib/zmailer/router -dkn 4 11839 ? S 0:00 /usr/lib/zmailer/router -dkn 4 2. Why does mailq need the -Q option when querying a scheduler running with -Q? This is quirk. Make mailq discover what protocol is used by the scheduler and act appropriately. 3. Is there any mailing list manager available that uses the native mailing list features of Zmailer? Also how big can the lists be that Zmailer can reasonably handle natively 100.000? 10.000? 1.000? Or is it better to have bulkmail or smartlist handle the mailing lists? 4. Logging via syslogd It seesm that the syslog daemon consumes quite a bit of cpu time and is often waiting for disk i/o to take place. Is there some way to switch logging off or do logging in a different way? I also noticed that the scheduler/router log quite a bit of information to files that is not necessary. Any way to switch those off? --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Mon May 19 21:35:37 1997 Received: from NO-IDENT-SERVICE@[206.161.253.66] (port 47876 [206.161.253.66]) by nic.funet.fi with ESMTP id <1425-29324>; Mon, 19 May 1997 21:34:22 +0300 Received: from NO-IDENT-SERVICE@localhost (port 40708 [127.0.0.1]) by roadrunner.realbig.com with SMTP id <133020-173>; Mon, 19 May 1997 14:22:59 +0000 Date: Mon, 19 May 1997 14:22:55 -0400 (EDT) From: Andy Poling To: Christoph Lameter cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: Questions re Zmail In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 19 May 1997, Christoph Lameter wrote: > 1. What is this "Timeserver" and why does it garble the ps line? > > 11831 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > 11836 ? S 0:04 /usr/lib/zmailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > 11837 ? S 0:00 TimeServer ailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > 11838 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > 11839 ? S 0:00 /usr/lib/zmailer/router -dkn 4 It is responsible for serving the current date/time. I guess the idea is that, at maximum, one gettimeofday() per second will occur due to the scheduler. On a busy mail system it could be a significant win. I too noticed that, but it's not really harmful - it's just not over-writing the entire previous argv. > Also how big can the lists be that Zmailer can reasonably handle natively > > 100.000? 10.000? 1.000? [...] > Or is it better to have bulkmail or smartlist handle the mailing lists? It's all a matter of how much memory the router process will use when processing a message to the list. A longer list = more memory needed. The actual list expansion isn't a problem - just the size of the router while processing a large number of addresses in one message. Thus an external list expander (especially one written in an interpreted language) is not likely to help things. > I also noticed that the scheduler/router log quite a bit of information to > files that is not necessary. Any way to switch those off? The router logging can be controlled by the following line in /etc/zmailer.conf: LOGLEVEL="address: deferred: file: header_defer: info: recipient:" Just remove the things you don't want logged. The above is pretty much everything. Incidentally, my (Slackware) Linux mail system running Zmailer is delivering messages to an average (over the last 3 days, which have been relatively quiet) of 60-70k recipients per day (mostly the Miata mailing list) without a hitch. Until very recently, I delivered this volume with a P90 with 80MB of memory. Your 150MHz Cyrix CPU might be sufficient, but 16MB of RAM is probably too skimpy for serious work like this with any MTA. -Andy Global Auctions From zmailer-owner@nic.funet.fi Mon May 19 22:34:34 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 38783 [128.214.46.64]) by nic.funet.fi with ESMTP id <1509-29324>; Mon, 19 May 1997 22:33:21 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id WAA10559; Mon, 19 May 1997 22:32:53 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTYB3-0005DA-00 (Debian); Mon, 19 May 1997 12:33:01 -0700 Date: Mon, 19 May 1997 12:33:00 -0700 (PDT) From: Christoph Lameter To: Andy Poling cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: Questions re Zmail In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 19 May 1997, Andy Poling wrote: >On Mon, 19 May 1997, Christoph Lameter wrote: >> 1. What is this "Timeserver" and why does it garble the ps line? >> >> 11831 ? S 0:00 /usr/lib/zmailer/router -dkn 4 >> 11836 ? S 0:04 /usr/lib/zmailer/scheduler -l /var/log/mail/scheduler.perflog -S -H >> 11837 ? S 0:00 TimeServer ailer/scheduler -l /var/log/mail/scheduler.perflog -S -H >> 11838 ? S 0:00 /usr/lib/zmailer/router -dkn 4 >> 11839 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > >It is responsible for serving the current date/time. I guess the idea is >that, at maximum, one gettimeofday() per second will occur due to the >scheduler. On a busy mail system it could be a significant win. Why would a daemon be more efficient than the kernel code? Also I am running xntp on that host. The kernel time is really accurate. The "TimeServer" is useless AFAIK. >It's all a matter of how much memory the router process will use when >processing a message to the list. A longer list = more memory needed. The >actual list expansion isn't a problem - just the size of the router while >processing a large number of addresses in one message. Thus an external >list expander (especially one written in an interpreted language) is not >likely to help things. How much memory does it need for 10.000 adresses? >Incidentally, my (Slackware) Linux mail system running Zmailer is delivering >messages to an average (over the last 3 days, which have been relatively >quiet) of 60-70k recipients per day (mostly the Miata mailing list) without >a hitch. Until very recently, I delivered this volume with a P90 with 80MB >of memory. Your 150MHz Cyrix CPU might be sufficient, but 16MB of RAM is >probably too skimpy for serious work like this with any MTA. 16M is just fine for an exploder with 60k messages. There is already a problem with the syslogd getting to be a bottleneck which will not be affected by adding memory. I am not planning to run those huge mailing lists off the exploder. Thanks for the tips. We better stuff 128M or so into the mailing list host then. --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Mon May 19 22:42:26 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 10500 [204.244.210.193]) by nic.funet.fi with ESMTP id <1394-30400>; Mon, 19 May 1997 22:40:59 +0300 Received: from localhost (tom@localhost) by misery.sdf.com (8.8.5/8.8.5) with SMTP id MAA00649; Mon, 19 May 1997 12:39:31 -0700 (PDT) Date: Mon, 19 May 1997 12:39:29 -0700 (PDT) From: Tom Samplonius To: Christoph Lameter cc: Andy Poling , Matti Aarnio , zmailer@nic.funet.fi Subject: Re: Questions re Zmail In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Mon, 19 May 1997, Christoph Lameter wrote: > On Mon, 19 May 1997, Andy Poling wrote: > > >On Mon, 19 May 1997, Christoph Lameter wrote: > >> 1. What is this "Timeserver" and why does it garble the ps line? > >> > >> 11831 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > >> 11836 ? S 0:04 /usr/lib/zmailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > >> 11837 ? S 0:00 TimeServer ailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > >> 11838 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > >> 11839 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > > > >It is responsible for serving the current date/time. I guess the idea is > >that, at maximum, one gettimeofday() per second will occur due to the > >scheduler. On a busy mail system it could be a significant win. > > Why would a daemon be more efficient than the kernel code? > > Also I am running xntp on that host. The kernel time is really accurate. > The "TimeServer" is useless AFAIK. You don't get it. The TimeServer reduces the number of syscalls. Rather doing a gettimeofday() for every message, it only does it once a second. The time is written into shared memory. You will see that it is used all over the place in the scheduler code. xntpd is actually quite useless. The scheduler does not need accurate time of day, but just the time between events. Tom From zmailer-owner@nic.funet.fi Tue May 20 11:50:30 1997 Received: by nic.funet.fi id <16048-29157>; Tue, 20 May 1997 11:49:15 +0300 Subject: Re: Questions re Zmail From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Tue, 20 May 1997 11:49:09 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 19, 97 10:30:36 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 4449 Message-Id: <19970520084915Z16048-29157+193@nic.funet.fi> Return-Path: Most questions have gotten answers by other people, thanks :-) > 1. What is this "Timeserver" and why does it garble the ps line? > > 11831 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > 11836 ? S 0:04 /usr/lib/zmailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > 11837 ? S 0:00 TimeServer ailer/scheduler -l /var/log/mail/scheduler.perflog -S -H > 11838 ? S 0:00 /usr/lib/zmailer/router -dkn 4 > 11839 ? S 0:00 /usr/lib/zmailer/router -dkn 4 A couple years back I did syscall traceing on the scheduler, and guess what syscall it did most frequently ? gettimeofday() :-) ( or time(), varied by the systems ) Now on Linux that is "cheap" operation, but on SunOS 4.1 it does cost (if not arm and leg) LOTS of time, nearly a millisecond! (Linux: circa 20 microseconds) In certain scheduling events it may be called several thousand times just for going thru the timing lists... Also yesterday here was a question about having knowledge about system load average, and IMO if the scheduler would need to check it every time it is about to fork() a new transport agent, it would soon be doing that generally very expensive operation over and over again... (On Linux it is cheap, but other systems are not so nice..) If the "TimeServer" would present the LoadAvg in its shared memory block, then it would be cheaper... For that I would suggest alternate solution: SCHEDULEROPTIONS=" ... -R 2 ..." That 'R' option is 'maxforkRate', that is how many forks the scheduler does during any given second. Indirectly it affects on how fast the scheduler can induce load to the system.. > 2. Why does mailq need the -Q option when querying a scheduler running > with -Q? This is quirk. Make mailq discover what protocol is used by the > scheduler and act appropriately. Actually the 'mailq' can't. The output IS empty for the case of having nothing in the queue, but it LOOKS empty, when the -Q option is used at the scheduler. > 3. Is there any mailing list manager available that uses the native > mailing list features of Zmailer? Yes, Majordomo does pretty well :-) (At vger.rutgers.edu) Or actually, most MLMs eventually run some external list expansion routine, which in sendmail case has traditionally been an alias: listname-outgoing: ":include:/path/to/file" which in ZMailer's case is simply placeing that file into directory MAILSHARE/lists/ with name listname-outgoing. > Also how big can the lists be that Zmailer can reasonably handle natively > > 100.000? 10.000? 1.000? > > Or is it better to have bulkmail or smartlist handle the mailing lists? With VERY BIG lists it is better to use external program to split the number of recipients down to more manageable size pieces -- like 1000 - 1500 recipients each. Also preferrably sort them by the domains so that the message has more changes to be delivered to more than one recipient at the time over the SMTP connections... It would be nice if the list expander would use zmailer(3) library (and data format!) to do the message submissions, then it would work neatly, and with minimal process creation. ( = No 'sendmail' running at all..) The ZMailer router process stays much leaner when it need not do the whole list expansion in core.. Years back, router memory consumption in list expansion grew in O(N**2) fashion, and it did need 150 MB memory to handle 600 addresses. Now it handles 2500 addresses with 20 MB. Much more linearly, though not as lean as I would like it to be... > 4. Logging via syslogd > > It seesm that the syslog daemon consumes quite a bit of cpu time and is > often waiting for disk i/o to take place. Is there some way to switch > logging off or do logging in a different way? syslogd's problem is that it does fsync() after each file write event, which is hurting the system... I am considering adding global method to turn off the syslog() logging in the 2.99.49. In the mean time, you would better to find a version of syslogd that is more sensible -- ask davem@caip.rutgers.edu, I asked him to do such for vger.rutgers.edu. (Before I put the newest ZMailer there...) > I also noticed that the scheduler/router log quite a bit of information to > files that is not necessary. Any way to switch those off? There were good replies for this. > --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 20 12:04:14 1997 Received: from NO-IDENT-SERVICE@atomic.clock.org (port 12036 [198.68.110.24]) by nic.funet.fi with ESMTP id <16209-29157> convert rfc822-to-8bit; Tue, 20 May 1997 12:02:44 +0300 Received: from [198.68.110.2] (mechanical.clock.org [198.68.110.2]) by atomic.clock.org (8.8.5/8.8.5) with ESMTP id CAA29413; Tue, 20 May 1997 02:02:39 -0700 (PDT) X-Sender: eef-play@pop.mercy.st.clock.org Message-Id: In-Reply-To: <19970520084915Z16048-29157+193@nic.funet.fi> References: from "Christoph Lameter" at May 19, 97 10:30:36 am Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8BIT Date: Tue, 20 May 1997 02:02:30 -0700 To: Matti Aarnio From: "Erik E. Fair" (Time Keeper) Subject: Re: Questions re Zmail Cc: zmailer@nic.funet.fi Return-Path: The scheduler/mailq thing could be solved by dropping state someplace (e.g. a file) that scheduler could create when it has been started with "-Q" and mailq could look for. Erik From zmailer-owner@nic.funet.fi Tue May 20 17:24:42 1997 Received: from mail@hur-s0.fuller.edu (port 46360 [206.1.27.36]) by nic.funet.fi with SMTP id <15913-8475>; Tue, 20 May 1997 17:23:43 +0300 Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTppP-0007wV-00 (Debian); Tue, 20 May 1997 07:23:51 -0700 Date: Tue, 20 May 1997 07:23:51 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: Questions re Zmail In-Reply-To: <19970520084915Z16048-29157+193@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 20 May 1997, Matti Aarnio wrote: > Also yesterday here was a question about having knowledge about > system load average, and IMO if the scheduler would need to check > it every time it is about to fork() a new transport agent, it > would soon be doing that generally very expensive operation > over and over again... (On Linux it is cheap, but other systems > are not so nice..) If the "TimeServer" would present the LoadAvg > in its shared memory block, then it would be cheaper... That is a good idea. >> 2. Why does mailq need the -Q option when querying a scheduler running >> with -Q? This is quirk. Make mailq discover what protocol is used by the >> scheduler and act appropriately. > > Actually the 'mailq' can't. The output IS empty for the case > of having nothing in the queue, but it LOOKS empty, when the -Q > option is used at the scheduler. Then make the output be some characteristic string. It would be nice if mailq would display without the -Q option even if the scheduler uses -Q. > With VERY BIG lists it is better to use external program to split > the number of recipients down to more manageable size pieces -- like > 1000 - 1500 recipients each. Also preferrably sort them by the > domains so that the message has more changes to be delivered to > more than one recipient at the time over the SMTP connections... > > It would be nice if the list expander would use zmailer(3) library > (and data format!) to do the message submissions, then it would > work neatly, and with minimal process creation. ( = No 'sendmail' > running at all..) Maybe one could use the buildin script language to write a small listserver? Also does the subscription mechanism to the list work? I have followed instructions but not received any response to my subscribe request. From zmailer-owner@nic.funet.fi Tue May 20 17:48:26 1997 Received: by nic.funet.fi id <13258-22364>; Tue, 20 May 1997 17:47:23 +0300 Subject: Re: Questions re Zmail From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Tue, 20 May 1997 17:47:17 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 20, 97 07:23:51 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2438 Message-Id: <19970520144723Z13258-22364+223@nic.funet.fi> Return-Path: > On Tue, 20 May 1997, Matti Aarnio wrote: > > > Also yesterday here was a question about having knowledge about > > system load average, and IMO if the scheduler would need to check > > it every time it is about to fork() a new transport agent, it > > would soon be doing that generally very expensive operation > > over and over again... (On Linux it is cheap, but other systems > > are not so nice..) If the "TimeServer" would present the LoadAvg > > in its shared memory block, then it would be cheaper... > > That is a good idea. Except that 'loadavg' code is non-trivially system dependent, sometimes even version dependent. Does anybody out there have autoconfigured loadavg code ? > >> 2. Why does mailq need the -Q option when querying a scheduler running > >> with -Q? This is quirk. Make mailq discover what protocol is used by the > >> scheduler and act appropriately. > > > > Actually the 'mailq' can't. The output IS empty for the case > > of having nothing in the queue, but it LOOKS empty, when the -Q > > option is used at the scheduler. > > Then make the output be some characteristic string. It would be nice if > mailq would display without the -Q option even if the scheduler > uses -Q. Ah... So the 'classical' display could be with some '-D' option (for 'detailed delivery states'), and the default would be this 'Q' ? > > With VERY BIG lists it is better to use external program to split > > the number of recipients down to more manageable size pieces -- like > > 1000 - 1500 recipients each. Also preferrably sort them by the > > domains so that the message has more changes to be delivered to > > more than one recipient at the time over the SMTP connections... > > > > It would be nice if the list expander would use zmailer(3) library > > (and data format!) to do the message submissions, then it would > > work neatly, and with minimal process creation. ( = No 'sendmail' > > running at all..) > > Maybe one could use the buildin script language to write a small > listserver? Not at the moment. Current script language is very memory intensive :-( > Also does the subscription mechanism to the list work? I have followed > instructions but not received any response to my subscribe request. It should... is the canonic location, and I think I have seen successfull uses of the alias: /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 20 17:55:13 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 32811 [128.214.46.64]) by nic.funet.fi with ESMTP id <16243-7084>; Tue, 20 May 1997 17:54:00 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-2.59) with SMTP id RAA27564; Tue, 20 May 1997 17:53:38 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTqIT-00080K-00 (Debian); Tue, 20 May 1997 07:53:53 -0700 Date: Tue, 20 May 1997 07:53:53 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: Questions re Zmail In-Reply-To: <19970520144723Z13258-22364+223@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 20 May 1997, Matti Aarnio wrote: >> > Actually the 'mailq' can't. The output IS empty for the case >> > of having nothing in the queue, but it LOOKS empty, when the -Q >> > option is used at the scheduler. >> >> Then make the output be some characteristic string. It would be nice if >> mailq would display without the -Q option even if the scheduler >> uses -Q. > > Ah... So the 'classical' display could be with some '-D' option > (for 'detailed delivery states'), and the default would be this 'Q' ? No. Put enough intelligence into mailq to recognize what the other end supports. I have had forgotten -Q a number of times and at first was very irritated at it. >> Also does the subscription mechanism to the list work? I have followed >> instructions but not received any response to my subscribe request. > > It should... is the canonic location, > and I think I have seen successfull uses of the alias: > Ok. I have sent another subscribe request. --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Tue May 20 18:26:29 1997 Received: by nic.funet.fi id <15428-8475>; Tue, 20 May 1997 18:25:44 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49beta2 (diff) Message-Id: <19970520152544Z15428-8475+199@nic.funet.fi> Date: Tue, 20 May 1997 18:25:34 +0300 Return-Path: Hello, The 2.99.49beta2 diff (no full dump) is available at my workstation ( ftp://mea.tmt.tele.fi/ ) for those with keen interest at what I have done last night (besides of watching Conan O'Brian show at NBC... .. because all european stations stop at around midnight, except the NBC..) - DSN bugfix with scheduler/smtp getting different ideas of what the status is - man/mailq.1* -manpage update - router/libdb/bind.c -- included missed - made the smtp channel to use getaddrinfo() API instead of the older gethostbyname() et.al. API. It compiles, but tests are very partial.. /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 20 19:24:34 1997 Received: from NO-IDENT-SERVICE@pobox.csc.fi (port 25696 [128.214.46.62]) by nic.funet.fi with ESMTP id <12682-29157>; Tue, 20 May 1997 19:23:16 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox2.csc.fi (8.8.5/8.8.5/CSC/SGI-2.59) with SMTP id TAA05282 for ; Tue, 20 May 1997 19:23:10 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wTrgh-0001Qa-00 (Debian); Tue, 20 May 1997 09:22:59 -0700 Date: Tue, 20 May 1997 09:22:59 -0700 (PDT) From: Christoph Lameter cc: zmailer@nic.funet.fi Subject: Logging of router In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: unlisted-recipients:; (no To-header on input) Return-Path: I have set the LOGLEVEL="file:" Why do I still get huge amounts of <9705201552.AA41624@stud2.tuwien.ac.at>: address: nesta@rhythm.com <9705201552.AA41624@stud2.tuwien.ac.at>: address: plr@isgtec.com <9705201552.AA41624@stud2.tuwien.ac.at>: address: riffraff@nastything.com in the routerlogs? Is there any other place to control logging of the router? --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Tue May 20 19:33:05 1997 Received: by nic.funet.fi id <846-8475>; Tue, 20 May 1997 19:31:47 +0300 Subject: Re: Logging of router From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Tue, 20 May 1997 19:31:46 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 20, 97 09:22:59 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 578 Message-Id: <19970520163147Z846-8475+232@nic.funet.fi> Return-Path: > I have set the LOGLEVEL="file:" > > Why do I still get huge amounts of > <9705201552.AA41624@stud2.tuwien.ac.at>: address: nesta@rhythm.com > <9705201552.AA41624@stud2.tuwien.ac.at>: address: plr@isgtec.com > <9705201552.AA41624@stud2.tuwien.ac.at>: address: riffraff@nastything.com > > in the routerlogs? > > Is there any other place to control logging of the router? See the cf/*.cf -files, there are some log statements, which propably are the culprit -- look for "address:" string > --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- /Matti Aarnio From zmailer-owner@nic.funet.fi Wed May 21 15:29:46 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 64721 [194.67.3.135]) by nic.funet.fi with ESMTP id <15463-15140>; Wed, 21 May 1997 15:28:40 +0300 Received: from NO-IDENT-SERVICE@ariel (port 61865 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-316>; Wed, 21 May 1997 16:28:25 +0400 Date: Wed, 21 May 1997 16:28:13 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49beta2 (diff) To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970520152544Z15428-8475+199@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > - router/libdb/bind.c -- included missed still missing in router/libdb/header.c (and it should be before libz.h but after someone #includes sys/types.h). Eugene From zmailer-owner@nic.funet.fi Wed May 21 16:01:21 1997 Received: from NO-IDENT-SERVICE@aecgate.aec.ca (port 57094 [142.56.1.1]) by nic.funet.fi with ESMTP id <12736-18206>; Wed, 21 May 1997 16:00:38 +0300 Received: from soltoo ([142.56.132.34]) by aecgate.aec.ca with ESMTP id <46475-8>; Wed, 21 May 1997 07:00:36 -0600 Received: by unix.aec.ca id <5701-473>; Wed, 21 May 1997 07:00:15 -0600 Sender: Trevor Paquette From: "Trevor Paquette" Message-Id: <970521070004.ZM10732@soltoo> Date: Wed, 21 May 1997 07:00:04 -0600 In-Reply-To: "Trevor Paquette" "URGENT: smtp session aborted.. 2.99.46p2" (May 13, 9:56) References: <970513095640.ZM340@sol132.aec.ca> X-Mailer: Z-Mail (4.0.1 13Jan97) To: zmailer@nic.funet.fi Subject: FIXED: URGENT: smtp session aborted.. 2.99.46p2 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-Path: WE have resolved this issue by upgrading everyone to the latest version of the client side TCPIP stack. Matti, (who really surprised me by phoning me!! Thanks again!) said that he has seen this before with the following combination: Windows client + FTP OnNet TCP version 2.1 and lower + Netscape + Zmailer 2.99.38 and higher + Ethernet (Matti it is still there with Zmailer 2.99.48p2) By changing the client stack to OnNet 2.5 the problem goes away. Hope this helps someone else. On May 13, 9:56, Trevor Paquette wrote: > Subject: URGENT: smtp session aborted.. 2.99.46p2 > HEre is the situation: > > clients using a mailhost running sendmail can send mail > clients using a mailhost running zmailer 2.97 can send mail > clients using a mailhost running zmailer 2.99.10/13 can send mail > > however when connecting to a server running zmailer 2.99.46p2 some hosts get: > > (What happens is that the client never send the HELO greeting. It looks like it > never sees the SMTP greeting banner.. > Has anyone seen this before?? > > 103# connection from IDENT-FD2-FAULT-1-32@karmrust.aec.ca (port 2035) > 103w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Tue, > 13 May 1997 09:19:26 -0600 > 103# remote from [142.56.113.79] > 103# aborted: session terminated > > 100# connection from IDENT-FD2-FAULT-1-32@mtodesco2.aec.ca (port 2283) > 100w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Tue, > 13 May 1997 09:18:04 -0600 > 100# remote from [142.56.118.58] > 100# aborted: session terminated > > 2071# connection from IDENT-FD2-FAULT-1-32@rkokotai.aec.ca (port 1327) > 2071w 220 sol132.aec.ca ZMailer Server 2.99.46 #1 ESMTP+IDENT ready at Mon, > 12 May 1997 10:28:27 -0600 > 2071# remote from [142.56.118.101] > 2071# aborted: session terminated > >-- End of excerpt from Trevor Paquette -- Trevor Paquette | MetroNet Solutions |Work:(403)543-2355 TrevorPaquette@mcc.net | 2700, 421 7th Ave SW | Fax:(403)543-2854 http://www.mcc.net | Calgary, Ab, Canada |ICBM:51'05"N/114'01"W Senior Unix Network Architect| T2P 4K9 |Mind:In the Rockies From zmailer-owner@nic.funet.fi Wed May 21 16:48:54 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 50643 [194.67.3.135]) by nic.funet.fi with ESMTP id <12402-19841>; Wed, 21 May 1997 16:48:09 +0300 Received: from NO-IDENT-SERVICE@ariel (port 61986 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-319>; Wed, 21 May 1997 17:47:47 +0400 Date: Wed, 21 May 1997 17:47:32 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49beta2 (diff) To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970520152544Z15428-8475+199@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: extern char **environ; - declaration missing in libsh/variables.c (I wonder how did it compile for you?) There are a few places where libz.h produces a warning because sys/socket.h is not #included. Maybe better #include it right in libz.h? In the file transport/libta/diagnostic.c: a lot of conflicting types for functions defined in both libc.h and mail.h. The same functions are defined differently. libc/getaddrinfo.c (and maybe others?): there are no int16_t and u_int16_t in Solaris; *int32* seems to be defined in a wrong way. The worst thing: identifier "sun" is #defined to '1' in the compiler and cannot be used! (around line 346). The same problem with *int*_t in getnameinfo.c. in smtpserver/smtpserver.c line 1053 is an operator that goes before data declaration in the line 1055. Should be moved down. Also, there is a lot of warnings "assignment discards 'const' from pointer target type" but I guess that was your intention :) Compiled on Solaris 2.5.1 with gcc 2.7.2 Eugene From zmailer-owner@nic.funet.fi Wed May 21 19:31:45 1997 Received: by nic.funet.fi id <15426-19841>; Wed, 21 May 1997 19:31:14 +0300 Subject: Re: 2.99.49beta2 (diff) From: Matti Aarnio To: crosser@online.ru Date: Wed, 21 May 1997 19:31:10 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 21, 97 05:47:32 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 304 Message-Id: <19970521163114Z15426-19841+190@nic.funet.fi> Return-Path: .... > Compiled on Solaris 2.5.1 with gcc 2.7.2 > > Eugene Indeed... SunSoft cc 3.0.1 barfs too. More tomorrow, once I get everything compiled, and fixed. Oh yes, 'beta3' does not help here. Diffs will appear to be 'beta4', but tonight I have to run elsewere.. /Matti Aarnio From zmailer-owner@nic.funet.fi Wed May 21 20:24:39 1997 Received: from NO-IDENT-SERVICE@ttn.ru (port 41732 [194.190.29.11]) by nic.funet.fi with ESMTP id <12840-15140>; Wed, 21 May 1997 20:23:41 +0300 Received: (from root@localhost) by ttn.ru (8.8.5/8.8.5) id VAA19825; Wed, 21 May 1997 21:23:21 +0400 (MSD) Date: Wed, 21 May 1997 21:23:21 +0400 (MSD) Message-Id: <199705211723.VAA19825@ttn.ru> From: "Oleg V. Zaitsev" To: zmailer@nic.funet.fi Organization: Taganrog town Telecommunication Node X-Mailer: Emacs 19.29 X-Class: Fast Precedence: special-delivery MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Reply-to: olz@ttn.ru Return-Path: unsubscribe all From zmailer-owner@nic.funet.fi Thu May 22 18:30:18 1997 Received: by nic.funet.fi id <1488-3756>; Thu, 22 May 1997 18:29:52 +0300 Subject: Re: zmailer -- policy based MXing From: Matti Aarnio To: jwr@icm.edu.pl (Jan Rychter) Date: Thu, 22 May 1997 18:29:46 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Jan Rychter" at May 22, 97 03:40:02 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 348 Message-Id: <19970522152952Z1488-3756+17@nic.funet.fi> Return-Path: > Are you planning to do some work on those in the near future, or are > they low-priority things ? > > (I started configuring smtpserver.cf and wondered why most of the > options don't work...) I have been working at it at home. Priority is not as high as you might wish, but various SPAMs are upping it... > thanks, > --J. /Matti Aarnio From zmailer-owner@nic.funet.fi Thu May 22 18:40:21 1997 Received: by nic.funet.fi id <1762-5919>; Thu, 22 May 1997 18:39:38 +0300 Subject: Re: FIXED: URGENT: smtp session aborted.. 2.99.46p2 From: Matti Aarnio To: tpaquett@sol132.aec.ca (Trevor Paquette) Date: Thu, 22 May 1997 18:39:35 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <970521070004.ZM10732@soltoo> from "Trevor Paquette" at May 21, 97 07:00:04 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1855 Message-Id: <19970522153938Z1762-5919+12@nic.funet.fi> Return-Path: > WE have resolved this issue by upgrading everyone to the latest version > of the client side TCPIP stack. > > Matti, (who really surprised me by phoning me!! Thanks again!) said that he > has seen this before with the following combination: > > Windows client + FTP OnNet TCP version 2.1 and lower + Netscape + Zmailer > 2.99.38 and higher + Ethernet > > (Matti it is still there with Zmailer 2.99.48p2) Yes, and changeing SMTPOPTIONS="-asve -M 10000000 -l /var/log/mail/smtpserver" to SMTPOPTIONS="-sve -M 10000000 -l /var/log/mail/smtpserver" should help too. At Telecom Finland we run without the ident lookup (-a) active at the server. Since then we have had no problems. > By changing the client stack to OnNet 2.5 the problem goes away. Hope this > helps someone else. > > On May 13, 9:56, Trevor Paquette wrote: > > Subject: URGENT: smtp session aborted.. 2.99.46p2 > > HEre is the situation: > > > > clients using a mailhost running sendmail can send mail > > clients using a mailhost running zmailer 2.97 can send mail > > clients using a mailhost running zmailer 2.99.10/13 can send mail > > > > however when connecting to a server running zmailer 2.99.46p2 some hosts... I don't remember, when I added the ident lookup... ChangeLog leads me to believe it was before 2.90.x ... Interesting, I should compare 2.99.13 and 2.99.36 smtpservers (or libidents) to see what they do differently. Of course the question may relate on the "-a" option at the runtime parameters too.. > -- > Trevor Paquette | MetroNet Solutions |Work:(403)543-2355 > TrevorPaquette@mcc.net | 2700, 421 7th Ave SW | Fax:(403)543-2854 > http://www.mcc.net | Calgary, Ab, Canada |ICBM:51'05"N/114'01"W > Senior Unix Network Architect| T2P 4K9 |Mind:In the Rockies /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 01:22:06 1997 Received: by nic.funet.fi id <140-12387>; Fri, 23 May 1997 01:21:52 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49beta4 - Solaris compilation tested Message-Id: <19970522222152Z140-12387+34@nic.funet.fi> Date: Fri, 23 May 1997 01:21:52 +0300 Return-Path: Good -- morning, at ftp://mea.tmt.tele.fi/ there is now beta4 which compiles at the Solaris 2.5.1 with SC4.0 compiler. More things follow latter. Good night. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 01:29:44 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 48150 [128.214.46.64]) by nic.funet.fi with ESMTP id <48-6054>; Fri, 23 May 1997 01:28:59 +0300 Received: from waterf.org (mail@hur-s0.fuller.edu [206.1.27.36]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-3.01) with SMTP id BAA08478; Fri, 23 May 1997 01:28:51 +0300 (EET DST) Received: from localhost [127.0.0.1] (clameter) by waterf.org with smtp (Exim 1.62 #1) id 0wUgLa-0002pM-00 (Debian); Thu, 22 May 1997 15:28:34 -0700 Date: Thu, 22 May 1997 15:28:34 -0700 (PDT) From: Christoph Lameter To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: 2.99.49beta4 - Solaris compilation tested In-Reply-To: <19970522222152Z140-12387+34@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Well pretty late for a morning. Afternoon is almost through in Los Angeles. What does the patch4 do for other platforms like Linux? On Fri, 23 May 1997, Matti Aarnio wrote: >Good -- morning, > > at ftp://mea.tmt.tele.fi/ there is now beta4 which compiles > at the Solaris 2.5.1 with SC4.0 compiler. > > More things follow latter. Good night. > >/Matti Aarnio > > --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- +++ --- From zmailer-owner@nic.funet.fi Fri May 23 09:10:38 1997 Received: from NO-IDENT-SERVICE@[206.161.253.66] (port 20552 [206.161.253.66]) by nic.funet.fi with ESMTP id <12381-13715>; Fri, 23 May 1997 09:10:31 +0300 Received: from NO-IDENT-SERVICE@localhost (port 64 [127.0.0.1]) by roadrunner.realbig.com with SMTP id <131089-127>; Fri, 23 May 1997 00:24:31 +0000 Date: Fri, 23 May 1997 00:24:26 -0400 (EDT) From: Andy Poling To: Matti Aarnio cc: Trevor Paquette , zmailer@nic.funet.fi Subject: Re: FIXED: URGENT: smtp session aborted.. 2.99.46p2 In-Reply-To: <19970522153938Z1762-5919+12@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 22 May 1997, Matti Aarnio wrote: > Of course the question may relate on the "-a" option at the > runtime parameters too.. Here I think you've hit upon it. I don't think it used to be enabled by default... -Andy Global Auctions From zmailer-owner@nic.funet.fi Fri May 23 15:32:05 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 10493 [194.67.3.135]) by nic.funet.fi with ESMTP id <1659-13715>; Fri, 23 May 1997 15:30:48 +0300 Received: from NO-IDENT-SERVICE@ariel (port 33457 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-318>; Fri, 23 May 1997 16:30:34 +0400 Date: Fri, 23 May 1997 16:30:21 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49beta4 - Solaris compilation tested To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970522222152Z140-12387+34@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > at ftp://mea.tmt.tele.fi/ there is now beta4 which compiles > at the Solaris 2.5.1 with SC4.0 compiler. It does not with gcc: gcc does not allow invalid C constructs in error messages in libc/get{name,addr}info.c (those if SIZEOF... not match anything), lines 207 and 153. I changed blah-blah to #error "balh-blah" and it compiled cleanly. Eugene From zmailer-owner@nic.funet.fi Fri May 23 15:53:01 1997 Received: by nic.funet.fi id <1286-6054>; Fri, 23 May 1997 15:51:41 +0300 Subject: Re: 2.99.49beta4 - Solaris compilation tested From: Matti Aarnio To: clameter@waterf.org (Christoph Lameter) Date: Fri, 23 May 1997 15:51:37 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Christoph Lameter" at May 22, 97 03:28:34 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 661 Message-Id: <19970523125141Z1286-6054+227@nic.funet.fi> Return-Path: > Well pretty late for a morning. Afternoon is almost through in Los > Angeles. I refer to my local time -- see my original 'Date:' header :) I had to get out of the office before the last commuter train of the night went past ... > What does the patch4 do for other platforms like Linux? Nothing adverse apparently, as seen at my home RedHat 4.0. It does merge in (and clean a bit) your Debian things. > On Fri, 23 May 1997, Matti Aarnio wrote: > >Good -- morning, > > > > at ftp://mea.tmt.tele.fi/ there is now beta4 which compiles > > at the Solaris 2.5.1 with SC4.0 compiler. > > > > More things follow latter. Good night. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 16:15:59 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 6142 [194.67.3.135]) by nic.funet.fi with ESMTP id <1366-13715>; Fri, 23 May 1997 16:14:49 +0300 Received: from SOCKFAULT1@ariel (port 33559 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68051-18982>; Fri, 23 May 1997 17:14:37 +0400 Date: Fri, 23 May 1997 17:14:35 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49beta4 - Solaris compilation tested To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970522222152Z140-12387+34@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: When you run it, it delivers locally OK, but it would not accept from outside for non-local targets, and would not deliver by smtp. Smtp transport reports: "No MX and no data" about perfectly legal domains that where correctly processed by router. Smtpserver dies terrible death if target is not local: Starting program: /v1/src/zmailer/zmailer-2.99.49b3/smtpserver/smtpserver -i 220 chronos ZMailer Server 2.99.49b4 #1 ESMTP ready at Fri, 23 May 1997 17:12:47 +0400 helo x 250 chronos Hello x mail from:<> 250 2.1.0 Sender syntax Ok rcpt to: 000 Ok Program received signal SIGSEGV, Segmentation fault. free (cp=0xefff908b) at malloc.c:328 328 if (TAG(p0) == FREE) { (gdb) bt #0 free (cp=0xefff908b) at malloc.c:328 #1 0x1bda0 in smtp_rcpt (SS=0xefffb330, buf=0xefff9080 "rcpt to:", cp=0xefff9089 "crosser@online.ru>") at smtpcmds.c:746 #2 0x16650 in smtpserver (SS=0xefffb330, insecure=0) at smtpserver.c:1307 #3 0x14ad0 in main (argc=2, argv=0xeffffc6c) at smtpserver.c:410 Eugene From zmailer-owner@nic.funet.fi Fri May 23 17:17:23 1997 Received: by nic.funet.fi id <1517-6054>; Fri, 23 May 1997 17:16:30 +0300 Subject: Re: 2.99.49beta4 - Solaris compilation tested From: Matti Aarnio To: crosser@online.ru Date: Fri, 23 May 1997 17:16:28 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 23, 97 04:30:21 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1711 Message-Id: <19970523141630Z1517-6054+235@nic.funet.fi> Return-Path: > > at ftp://mea.tmt.tele.fi/ there is now beta4 which compiles > > at the Solaris 2.5.1 with SC4.0 compiler. > > It does not with gcc: gcc does not allow invalid C constructs in > error messages in libc/get{name,addr}info.c (those if SIZEOF... > not match anything), lines 207 and 153. I changed ... You need to rerun the configure to get them fixed. I did notice these two items too latter (of night) at home. I hacked (at night) also a bit of the policy filtering things, you can get: zmailer-2.99.49b5.diff with the first thoughts about the code... Because I have a busy weekend coming (I have two HAM-Radio meetings..), here are some bits what some of you could do for the policy filtering code: The intention if to have some (arbitary) format of keyed database which contains binary representation of the "policy.dat" source. The parsers are written, but the main program is not yet done for that compiler. Also the policytest.c need routines that: - Build proper keyed queries (for address and domain), and do the lookups - If something is found, the resulting attributes need to be iterated, while keeping in mind that due to "="-alias constructor one can create loops, which must be broken by some means... This iterator should get a set of attributes to look for, and if it can find them all before the attributes run out, it will stop, and return the result, else it shall run until the end of the attributes (or loop-break..) Only the first instance of the attribute is registered. The "="-alias can be present multiple times for some keyed item, thus joining two sets of attribute lists. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 17:20:42 1997 Received: from TIMEDOUT@falco.its.uu.se (port 4873 [130.238.131.113]) by nic.funet.fi with ESMTP id <1276-6054>; Fri, 23 May 1997 17:19:38 +0300 Received: from alba.its.uu.se ([130.238.131.102]) by falco.its.uu.se with ESMTP id <2056-22150>; Fri, 23 May 1997 16:19:22 +0100 X-Mailer: exmh version 2.0gamma 1/27/96 To: Matti Aarnio cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se Subject: Bug in sm still truncates messages on AIX (New bug) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 23 May 1997 16:19:20 +0200 From: Martin Wendel Message-Id: <97May23.161922+0100msz.2056-22150+4@falco.its.uu.se> Return-Path: zmailer version: 2.99.48.p2 AIX version 4.1.5 I've seen this: May 23 15:16:10 uria sm[15970]: 24601-26122 readalready=32768, bodyoffset=641, sizeestimate=54091 When logging in sm.c appendlet(): #ifndef HAVE_MMAP /* can we use cache of message body data */ if (convertmode == _CONVERT_NONE && readalready != 0) { syslog(LOG_DEBUG, "%s readalready=%d, bodyoffset=%d, sizeestimate=%d", dp->msgfile, readalready, dp->msgbodyoffset, dp->msgsizeestimate); lastch = let_buffer[readalready-1]; if (writebuf(mp, fp, let_buffer, readalready) != readalready) return EX_IOERR; if (lastch != '\n') if (writebuf(mp, fp, "\n", 1) != 1) return EX_IOERR; return EX_OK; } Result is (of course) only about 32k of the message being delivered. I'm trying setting readalready=0 first in appendlet now and will try it over the weekend. Seems something weird is happening with the read()'s... Could perhaps check for readalready==32k and reset then but... -- E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS Tel: 018 18 77 80 Box 887 Int: +46 18 18 77 80 S-751 08 Uppsala Fax: +46 18 18 79 10 Sweden http://www.uu.se/Internservice/ue/personal/MartinWendel.html From zmailer-owner@nic.funet.fi Fri May 23 22:34:16 1997 Received: by nic.funet.fi id <1316-13715>; Fri, 23 May 1997 22:33:23 +0300 Subject: Re: Bug in sm still truncates messages on AIX (New bug) From: Matti Aarnio To: martin.wendel@its.uu.se (Martin Wendel) Date: Fri, 23 May 1997 22:33:19 +0300 (EET DST) Cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se In-Reply-To: <97May23.161922+0100msz.2056-22150+4@falco.its.uu.se> from "Martin Wendel" at May 23, 97 04:19:20 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 844 Message-Id: <19970523193323Z1316-13715+253@nic.funet.fi> Return-Path: > zmailer version: 2.99.48.p2 > AIX version 4.1.5 > > I've seen this: > May 23 15:16:10 uria sm[15970]: 24601-26122 readalready=32768, bodyoffset=641, sizeestimate=54091 > When logging in sm.c appendlet(): Weird... Very weird indeed... ... Umm.. Ah.. check_7bit_cleanness() may be the culprit here. It leaves the loop sometimes before it can read until the EOF, and I presume that your emails are with 8-bit chars, and/or you have "Content-Conversion: prohibited" headers... > -- > E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS > Tel: 018 18 77 80 Box 887 > Int: +46 18 18 77 80 S-751 08 Uppsala > Fax: +46 18 18 79 10 Sweden > http://www.uu.se/Internservice/ue/personal/MartinWendel.html /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 22:38:15 1997 Received: by nic.funet.fi id <1583-29679>; Fri, 23 May 1997 22:37:26 +0300 Subject: Re: 2.99.49beta4 - Solaris compilation tested From: Matti Aarnio To: crosser@online.ru Date: Fri, 23 May 1997 22:37:14 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 23, 97 05:14:35 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 517 Message-Id: <19970523193726Z1583-29679+248@nic.funet.fi> Return-Path: > When you run it, it delivers locally OK, but it would not accept from > outside for non-local targets, and would not deliver by smtp. Smtp > transport reports: "No MX and no data" about perfectly legal domains > that where correctly processed by router. Smtpserver dies terrible > death if target is not local: The smtpserver died terrible death always, I think. I found that bug. I was reusing a variable that should not have been touched. Testing continues. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 23:34:15 1997 Received: by nic.funet.fi id <1486-6054>; Fri, 23 May 1997 23:33:14 +0300 Subject: Re: duplicate email from Apple/Eudora Internet servers From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Fri, 23 May 1997 23:33:10 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at May 15, 97 02:41:29 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1952 Message-Id: <19970523203314Z1486-6054+274@nic.funet.fi> Return-Path: > Hi, > > I am wondering if there is anyone else here who has gotten massive > duplicated email from a site running Apple Internet server or Eudora > Internet server. The offending sites connect once every 15 to 30 minutes > and submit the same message as on previous connections. Right now, I am > getting around this problem by outfitting victim users with a simple > 'procmail' filter to quietly sink the duplicates. But I am wondering if > there is any other solution available besides this, complaining the > postmasters of the remote site and complaining to Eudora and Apple. I have pondered this question, and can't figure out it being ZMailer's fault. The RFC 1830 examples are buggy, and if the implementor isn't extremely carefull, they may make a mistake.. Still, the BDAT session looks just fine, so I have no idea what the trouble is at the sending side. > Here is a sample SMTP conversation below: > > 16312w 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT r > eady at Thu, 15 May 1997 14:32:45 -0500 > 16312# remote from [206.126.246.3] > 16312r EHLO mail.ness.org > 16312w 250-press-gopher.uchicago.edu expected "EHLO woz.ness.org" > 16312w 250-SIZE 0 > 16312w 250-8BITMIME > 16312w 250-PIPELINING > 16312w 250-CHUNKING > 16312w 250-ENHANCEDSTATUSCODES > 16312w 250-EXPN > 16312w 250-VRFY > 16312w 250-DSN > 16312w 250-X-RCPTLIMIT 10000 > 16312w 250-ETRN > 16312w 250 HELP > 16312r MAIL FROM: SIZE=2522 > 16312w 250 2.1.0 Sender syntax Ok > 16312# -- pipeline input exists 34 bytes > 16312r RCPT TO: > 16312w 250 2.1.5 Ok; can accomodate 2522 byte message > 16312r BDAT 2522 LAST > 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool > > Any ideas will be much appreciated! > > Thanks, > Roy Bixler > The University of Chicago Press postmaster > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 23 23:41:04 1997 Received: from mail@hur-s0.fuller.edu (port 15742 [206.1.27.36]) by nic.funet.fi with SMTP id <1636-6054>; Fri, 23 May 1997 23:40:01 +0300 Received: from dx100.waterf.org [192.168.0.3] by waterf.org with esmtp (Exim 1.62 #1) id 0wV17p-0007Zh-00 (Debian); Fri, 23 May 1997 13:39:45 -0700 From: "Christoph Lameter" To: "Matti Aarnio" , "Roy C Bixler" Cc: Subject: Re: duplicate email from Apple/Eudora Internet servers Date: Fri, 23 May 1997 20:40:16 -0700 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1161 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Message-Id: Return-Path: Eudora is known to duplicate mail in some situations. We have had a problem with the standard pop3 server on our Unix system with Eudora. We switched to use cucipop instead of the standard pop3 server and the problem was gone. ---------- > From: Matti Aarnio > To: Roy C Bixler > Cc: zmailer@nic.funet.fi > Subject: Re: duplicate email from Apple/Eudora Internet servers > Date: Friday, May 23, 1997 1:33 PM > > > Hi, > > > > I am wondering if there is anyone else here who has gotten massive > > duplicated email from a site running Apple Internet server or Eudora > > Internet server. The offending sites connect once every 15 to 30 minutes > > and submit the same message as on previous connections. Right now, I am > > getting around this problem by outfitting victim users with a simple > > 'procmail' filter to quietly sink the duplicates. But I am wondering if > > there is any other solution available besides this, complaining the > > postmasters of the remote site and complaining to Eudora and Apple. > > I have pondered this question, and can't figure out it being > ZMailer's fault. The RFC 1830 examples are buggy, and if the > implementor isn't extremely carefull, they may make a mistake.. > > Still, the BDAT session looks just fine, so I have no idea > what the trouble is at the sending side. > > > Here is a sample SMTP conversation below: > > > > 16312w 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT r > > eady at Thu, 15 May 1997 14:32:45 -0500 > > 16312# remote from [206.126.246.3] > > 16312r EHLO mail.ness.org > > 16312w 250-press-gopher.uchicago.edu expected "EHLO woz.ness.org" > > 16312w 250-SIZE 0 > > 16312w 250-8BITMIME > > 16312w 250-PIPELINING > > 16312w 250-CHUNKING > > 16312w 250-ENHANCEDSTATUSCODES > > 16312w 250-EXPN > > 16312w 250-VRFY > > 16312w 250-DSN > > 16312w 250-X-RCPTLIMIT 10000 > > 16312w 250-ETRN > > 16312w 250 HELP > > 16312r MAIL FROM: SIZE=2522 > > 16312w 250 2.1.0 Sender syntax Ok > > 16312# -- pipeline input exists 34 bytes > > 16312r RCPT TO: > > 16312w 250 2.1.5 Ok; can accomodate 2522 byte message > > 16312r BDAT 2522 LAST > > 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool > > > > Any ideas will be much appreciated! > > > > Thanks, > > Roy Bixler > > The University of Chicago Press postmaster > > rcb@press-gopher.uchicago.edu > > /Matti Aarnio > > From zmailer-owner@nic.funet.fi Fri May 23 23:49:45 1997 Received: by nic.funet.fi id <1723-13715>; Fri, 23 May 1997 23:48:48 +0300 Subject: 2.99.49beta5 - running at mea.tmt.tele.fi ... From: Matti Aarnio To: crosser@online.ru Date: Fri, 23 May 1997 23:48:42 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 23, 97 05:14:35 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 770 Message-Id: <19970523204848Z1723-13715+263@nic.funet.fi> Return-Path: > When you run it, it delivers locally OK, but it would not accept from > outside for non-local targets, and would not deliver by smtp. Smtp > transport reports: "No MX and no data" about perfectly legal domains > that where correctly processed by router. Smtpserver dies terrible > death if target is not local: Ok, now I have 2.99.49beta5 running at my workstation. And the diffs are available too. Those of you who picked the "b5.diff" earlier today, this is not the same... I have not noticed (yet) problems at the getaddrinfo() mechanism, and I have sent several messages with explicite testing in mind. My workstation is i686 Linux, but the Solaris tests are partial -- and DEC Alpha not done at all. ... > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Sat May 24 02:14:35 1997 Received: from NO-IDENT-SERVICE@press-gopher.uchicago.edu (port 42956 [128.135.234.36]) by nic.funet.fi with ESMTP id <1645-26434>; Sat, 24 May 1997 02:13:49 +0300 Received: from TIMEDOUT@dhcp-849726459.qualcomm.com (port 2367 [129.46.137.21]) by press-gopher.uchicago.edu with ESMTP id <26529-202>; Fri, 23 May 1997 18:13:31 -0500 Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 16:11:25 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 16:11:01 -0700 To: Roy C Bixler From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: Matti Aarnio , zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message >> 16312r BDAT 2522 LAST >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool While ZMail may think it is sending a response, AIMS/EIMS is never getting it. I got this from a captured session: Fri, 23 May 1997 15:30:37 -0700 128.135.234.36 Connecting to press-gopher.uchicago.edu Fri, 23 May 1997 15:30:38 -0700 128.135.234.36 SMTPC connection opened 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT ready at Fri, 23 May 1997 17:30:44 -0500 EHLO dhcp-849726459.qualcomm.com 250-press-gopher.uchicago.edu Hello dhcp-849726459.qualcomm.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANCEDSTATUSCODES 250-EXPN 250-VRFY 250-DSN 250-X-RCPTLIMIT 10000 250-ETRN 250 HELP MAIL FROM: SIZE=790 RCPT TO: 250 2.1.0 Sender syntax Ok 250 2.1.5 Ok; can accomodate 790 byte message BDAT 790 LAST Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 15:26:30 -0700 To: glenn@cybertech.apple.com From: Glenn Anderson Subject: test testing. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed EIMS timed out waiting for a response to the data it sent. As a result, EIMS can only assume that the message as not been received so retries it again later. I also tried doing a manual send using Telnet and press-gopher.uchicago.edu just closed the connection after receiving the BDAT data, it did not return a response. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Sat May 24 02:16:55 1997 Received: from NO-IDENT-SERVICE@press-gopher.uchicago.edu (port 42956 [128.135.234.36]) by nic.funet.fi with ESMTP id <1051-11207>; Sat, 24 May 1997 02:15:51 +0300 Received: from TIMEDOUT@dhcp-849726459.qualcomm.com (port 2369 [129.46.137.21]) by press-gopher.uchicago.edu with ESMTP id <26529-202>; Fri, 23 May 1997 18:15:32 -0500 Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 16:11:25 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 16:11:01 -0700 To: Roy C Bixler From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: Matti Aarnio , zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message >> 16312r BDAT 2522 LAST >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool While ZMail may think it is sending a response, AIMS/EIMS is never getting it. I got this from a captured session: Fri, 23 May 1997 15:30:37 -0700 128.135.234.36 Connecting to press-gopher.uchicago.edu Fri, 23 May 1997 15:30:38 -0700 128.135.234.36 SMTPC connection opened 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT ready at Fri, 23 May 1997 17:30:44 -0500 EHLO dhcp-849726459.qualcomm.com 250-press-gopher.uchicago.edu Hello dhcp-849726459.qualcomm.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANCEDSTATUSCODES 250-EXPN 250-VRFY 250-DSN 250-X-RCPTLIMIT 10000 250-ETRN 250 HELP MAIL FROM: SIZE=790 RCPT TO: 250 2.1.0 Sender syntax Ok 250 2.1.5 Ok; can accomodate 790 byte message BDAT 790 LAST Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 15:26:30 -0700 To: glenn@cybertech.apple.com From: Glenn Anderson Subject: test testing. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed EIMS timed out waiting for a response to the data it sent. As a result, EIMS can only assume that the message as not been received so retries it again later. I also tried doing a manual send using Telnet and press-gopher.uchicago.edu just closed the connection after receiving the BDAT data, it did not return a response. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Sat May 24 02:19:04 1997 Received: from TIMEDOUT@dhcp-849726459.qualcomm.com (port 17673 [129.46.137.21]) by nic.funet.fi with ESMTP id <1112-16120>; Sat, 24 May 1997 02:18:10 +0300 Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 16:11:25 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 16:11:01 -0700 To: Roy C Bixler From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: Matti Aarnio , zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message >> 16312r BDAT 2522 LAST >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool While ZMail may think it is sending a response, AIMS/EIMS is never getting it. I got this from a captured session: Fri, 23 May 1997 15:30:37 -0700 128.135.234.36 Connecting to press-gopher.uchicago.edu Fri, 23 May 1997 15:30:38 -0700 128.135.234.36 SMTPC connection opened 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT ready at Fri, 23 May 1997 17:30:44 -0500 EHLO dhcp-849726459.qualcomm.com 250-press-gopher.uchicago.edu Hello dhcp-849726459.qualcomm.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANCEDSTATUSCODES 250-EXPN 250-VRFY 250-DSN 250-X-RCPTLIMIT 10000 250-ETRN 250 HELP MAIL FROM: SIZE=790 RCPT TO: 250 2.1.0 Sender syntax Ok 250 2.1.5 Ok; can accomodate 790 byte message BDAT 790 LAST Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 15:26:30 -0700 To: glenn@cybertech.apple.com From: Glenn Anderson Subject: test testing. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed EIMS timed out waiting for a response to the data it sent. As a result, EIMS can only assume that the message as not been received so retries it again later. I also tried doing a manual send using Telnet and press-gopher.uchicago.edu just closed the connection after receiving the BDAT data, it did not return a response. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Sat May 24 06:10:50 1997 Received: from TIMEDOUT@ra4000-p23.qualcomm.com (port 3592 [129.46.54.103]) by nic.funet.fi with ESMTP id <1510-11207>; Sat, 24 May 1997 06:04:29 +0300 Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 16:11:25 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 16:11:01 -0700 To: Roy C Bixler From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: Matti Aarnio , zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message >> 16312r BDAT 2522 LAST >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool While ZMail may think it is sending a response, AIMS/EIMS is never getting it. I got this from a captured session: Fri, 23 May 1997 15:30:37 -0700 128.135.234.36 Connecting to press-gopher.uchicago.edu Fri, 23 May 1997 15:30:38 -0700 128.135.234.36 SMTPC connection opened 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT ready at Fri, 23 May 1997 17:30:44 -0500 EHLO dhcp-849726459.qualcomm.com 250-press-gopher.uchicago.edu Hello dhcp-849726459.qualcomm.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANCEDSTATUSCODES 250-EXPN 250-VRFY 250-DSN 250-X-RCPTLIMIT 10000 250-ETRN 250 HELP MAIL FROM: SIZE=790 RCPT TO: 250 2.1.0 Sender syntax Ok 250 2.1.5 Ok; can accomodate 790 byte message BDAT 790 LAST Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 15:26:30 -0700 To: glenn@cybertech.apple.com From: Glenn Anderson Subject: test testing. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed EIMS timed out waiting for a response to the data it sent. As a result, EIMS can only assume that the message as not been received so retries it again later. I also tried doing a manual send using Telnet and press-gopher.uchicago.edu just closed the connection after receiving the BDAT data, it did not return a response. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Sat May 24 06:17:31 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 19280 [128.214.46.64]) by nic.funet.fi with ESMTP id <1192-16120>; Sat, 24 May 1997 06:13:31 +0300 Received: from ra4000-p23.qualcomm.com (ra4000-p23.qualcomm.com [129.46.54.103]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-3.01) with ESMTP id GAA04731 for ; Sat, 24 May 1997 06:09:48 +0300 (EET DST) Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 16:11:25 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 16:11:01 -0700 To: Roy C Bixler From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: Matti Aarnio , zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message >> 16312r BDAT 2522 LAST >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool While ZMail may think it is sending a response, AIMS/EIMS is never getting it. I got this from a captured session: Fri, 23 May 1997 15:30:37 -0700 128.135.234.36 Connecting to press-gopher.uchicago.edu Fri, 23 May 1997 15:30:38 -0700 128.135.234.36 SMTPC connection opened 220 press-gopher.uchicago.edu ZMailer Server 2.99.48p2 #1 ESMTP+IDENT ready at Fri, 23 May 1997 17:30:44 -0500 EHLO dhcp-849726459.qualcomm.com 250-press-gopher.uchicago.edu Hello dhcp-849726459.qualcomm.com 250-SIZE 0 250-8BITMIME 250-PIPELINING 250-CHUNKING 250-ENHANCEDSTATUSCODES 250-EXPN 250-VRFY 250-DSN 250-X-RCPTLIMIT 10000 250-ETRN 250 HELP MAIL FROM: SIZE=790 RCPT TO: 250 2.1.0 Sender syntax Ok 250 2.1.5 Ok; can accomodate 790 byte message BDAT 790 LAST Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.com with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700 X-Sender: glenn@mail.aec.co.nz Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 23 May 1997 15:26:30 -0700 To: glenn@cybertech.apple.com From: Glenn Anderson Subject: test testing. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed EIMS timed out waiting for a response to the data it sent. As a result, EIMS can only assume that the message as not been received so retries it again later. I also tried doing a manual send using Telnet and press-gopher.uchicago.edu just closed the connection after receiving the BDAT data, it did not return a response. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Sun May 25 10:02:35 1997 Received: from root@acli.interlog.com (port 28180 [198.53.146.23]) by nic.funet.fi with ESMTP id <13091-25241>; Sun, 25 May 1997 10:02:22 +0300 Received: from acli@localhost (port 18452 [127.0.0.1]) by acli.interlog.com with SMTP id <16193-12151>; Sun, 25 May 1997 03:01:09 +0000 Date: Sun, 25 May 1997 03:00:48 -0400 (EDT) From: Ambrose Li Sender: acli@acli.interlog.com Reply-To: acli@mingpaoxpress.com To: zmailer@nic.funet.fi Subject: ident in zmailer Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Is anyone having consistent success with Zmailer's ident feature? My zmailer (2.99.48p2 on Linux) only gives me intermittant success. Zmailer sometimes gives me IDENT-NONSENSE when the other end runs an operational identd which does not give nonsense results. (Both my home machine and the machines at work run zmailer and working identd's, but both ends identify the other end as giving out IDENT-NONSENSE.) I am suspecting something wrong with my configuration, but I don't know what to look for. I do use --with-bundled-libresolv. -- Ambrose Li ~ acli@acli%interlog.com (Hello, world!\n) ~ http://www.interlog.com/~acli/~acli/ ~ print flush quit From zmailer-owner@nic.funet.fi Mon May 26 14:01:49 1997 Received: from TIMEDOUT@dfw-ix10.ix.netcom.com (port 36237 [206.214.98.10]) by nic.funet.fi with ESMTP id <12530-16425>; Mon, 26 May 1997 14:01:12 +0300 Received: (from smap@localhost) by dfw-ix10.ix.netcom.com (8.8.4/8.8.4) id CAA11156; Mon, 26 May 1997 02:16:44 -0500 (CDT) Received: from dal-tx7-44.ix.netcom.com(207.94.122.172) by dfw-ix10.ix.netcom.com via smap (V1.3) id smad08933; Mon May 26 02:08:10 1997 Message-Id: <3.0.1.32.19970525223053.006a85b8@popd.ix.netcom.com> X-Sender: Joeisme@popd.ix.netcom.com (Unverified) X-Mailer: Windows Eudora Light Version 3.0.1 (32) Date: Sun, 25 May 1997 22:30:53 -0500 To: herman273@juno.com From: Joe Isme Subject: this can be the best day of your life! Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Return-Path: This can be the best day of your life! Do You Know? john 3:16 16 "For God so loved the world that He gave His only begotten Son, that whoever believes in Him should not perish but have everlasting life. (NKJ) Who is Good? Rom 3:10 10 As it is written: "There is none righteous, no, not one; (NKJ) Who Has Sinned? Rom 3:23 23 for all have sinned and fall short of the glory of God, (NKJ) Where Sin Came From Rom 5:12 12 Therefore, just as through one man sin entered the world, and death through sin, and thus death spread to all men, because all sinned-- (NKJ) God's Price on Sin Rom 6:23 23 For the wages of sin is death, but the gift of God is eternal life in Christ Jesus our Lord. (NKJ) Our Way Out Rom 5:8 8 But God demonstrates His own love toward us, in that while we were still sinners, Christ died for us. (NKJ) Rom 10:9 9 that if you confess with your mouth the Lord Jesus and believe in your heart that God has raised Him from the dead, you will be saved. (NKJ) Rom 10:13 13 For "whoever calls on the name of the Lord shall be saved." (NKJ) Will you call upon the Lord now, and ask Him to save you? Then, pray this prayer: 'Lord Jesus, I confess to You that I am a sinner. I believe Jesus died for my sins and rose from the dead. The best way I know how, I ask you to forgive my sin, and I personally accept You as my Savior. From this moment on, I am going to depend on Jesus for my salvation, Amen.' From zmailer-owner@nic.funet.fi Mon May 26 19:43:07 1997 Received: by nic.funet.fi id <12887-19601>; Mon, 26 May 1997 19:42:17 +0300 Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) From: Matti Aarnio To: glenn@aims.aec.co.nz (Glenn Anderson) Date: Mon, 26 May 1997 19:42:08 +0300 (EET DST) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com In-Reply-To: from "Glenn Anderson" at May 23, 97 04:11:01 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3200 Message-Id: <19970526164217Z12887-19601+191@nic.funet.fi> Return-Path: > >> 16312w 250 2.1.5 Ok; can accomodate 2522 byte message > >> 16312r BDAT 2522 LAST > >> 16312w 250 2.6.0 S.nSqHV26582 Roger, stored 2704 bytes into spool > > While ZMail may think it is sending a response, AIMS/EIMS is never getting > it. I got this from a captured session: Btw: ZMailer != ZMail -- be carefull, the latter is a "newcomer", which is commercialized MUSH -- a MUA... I suspect the sender sends more than the exact number of bytes it is claims to send, as otherwise it would get reply pipeline flush... Could you try to send to my office workstation ? To address: mea@mea.tmt.tele.fi My system has some extra log tests in it to see what happens. It will also flush the reply pipe in every case, so your end should get a report back. The problem with BDAT is that it is EXTREMELY sensitive to the correct count of transferred bytes. My pipeline optimi- zation in the BDAT processing was such that if the transfer was successfull, and the input is not empty, then there is no immediate "replyflush". By the way, how many systems in the network have BDAT service at all ? ... I tried to count the number of chars in the message you logged, did I get it right, or did I under/overshoot a couple bytes ? At the end there are two Xes after your signature, of which I don't know, were they original CRLFs in the message ? (The "M" at the end of the line represents CR of the CRLF pair.) > BDAT 790 LAST --------------- start ------------ Received: from [129.46.137.21] (127.0.0.1) by dhcp-849726459.qualcomm.comM with ESMTP (Eudora Internet Mail Server 1.2b2); Fri, 23 May 1997 15:26:54 -0700M X-Sender: glenn@mail.aec.co.nzM Message-Id: M Mime-Version: 1.0M Content-Type: text/plain; charset="us-ascii"M Date: Fri, 23 May 1997 15:26:30 -0700M To: glenn@cybertech.apple.comM From: Glenn Anderson M Subject: testM M testing.M M -------------------------------------------------------------M Glenn Anderson glenn@aims.aec.co.nzM Macintosh Internet Programmer Phone (mobile) +64 25 518661M Eudora Internet Mail Server:M ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/M -------------------------------------------------------------M X X -------------- end at 790 chars ------- > > > > Fri, 23 May 1997 15:35:50 -0700 128.135.234.36 SMTPC connection closed > > EIMS timed out waiting for a response to the data it sent. As a result, > EIMS can only assume that the message as not been received so retries it > again later. I also tried doing a manual send using Telnet and > press-gopher.uchicago.edu just closed the connection after receiving the > BDAT data, it did not return a response. > > Glenn. > > ------------------------------------------------------------- > Glenn Anderson glenn@aims.aec.co.nz > Macintosh Internet Programmer Phone (mobile) +64 25 518661 > Eudora Internet Mail Server: > ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ > ------------------------------------------------------------- /Matti Aarnio phone: +358-20402082 ( UTC: 0700-2400 ) From zmailer-owner@nic.funet.fi Mon May 26 23:49:15 1997 Received: from NO-IDENT-SERVICE@port432-well.ihug.co.nz (port 3080 [202.49.241.178]) by nic.funet.fi with ESMTP id <1193-19601>; Mon, 26 May 1997 23:48:16 +0300 Received: from [202.49.241.178] (127.0.0.1) by port432-well.ihug.co.nz with ESMTP (Apple Internet Mail Server 2.0b16); Tue, 27 May 1997 08:48:04 +1200 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: <19970526164217Z12887-19601+191@nic.funet.fi> References: from "Glenn Anderson" at May 23, 97 04:11:01 pm Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 27 May 1997 08:48:04 +1200 To: Matti Aarnio From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: > I suspect the sender sends more than the exact number of bytes > it is claims to send, as otherwise it would get reply pipeline > flush... > > Could you try to send to my office workstation ? To address: > mea@mea.tmt.tele.fi > My system has some extra log tests in it to see what happens. > It will also flush the reply pipe in every case, so your end > should get a report back. Whatever you have done, it fixed the problem. I had no problem sending messages to that address using BDAT. No error messages were reported either. > The problem with BDAT is that it is EXTREMELY sensitive to > the correct count of transferred bytes. My pipeline optimi- > zation in the BDAT processing was such that if the transfer > was successfull, and the input is not empty, then there is > no immediate "replyflush". As EIMS works internally with messages internally stored in raw RFC 822 format, all it does is send the length of the message in the BDAT command, then send the message. It leaves no posibility of the size being wrong. It was actually a lot easier to implement BDAT than the normal transfer method as there is no need to pad lines starting with a '.' or search for . at the end. > By the way, how many systems in the network have BDAT service > at all ? EIMS 1.x supports sending using BDAT, but not receiving. Future versions will support receiving as well. I don't know of any other systems that use it. Let me know if you want to do any compatibility testing in future. > I tried to count the number of chars in the message you logged, > did I get it right, or did I under/overshoot a couple bytes ? > At the end there are two Xes after your signature, of which I > don't know, were they original CRLFs in the message ? > (The "M" at the end of the line represents CR of the CRLF pair.) Yes, there are two CRLF pairs at the end where the Xes are. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Mon May 26 23:50:27 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 53512 [128.214.46.64]) by nic.funet.fi with ESMTP id <1719-19601>; Mon, 26 May 1997 23:49:47 +0300 Received: from port432-well.ihug.co.nz (port432-well.ihug.co.nz [202.49.241.178]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-3.12) with ESMTP id XAA02618; Mon, 26 May 1997 23:49:33 +0300 (EET DST) Received: from [202.49.241.178] (127.0.0.1) by port432-well.ihug.co.nz with ESMTP (Apple Internet Mail Server 2.0b16); Tue, 27 May 1997 08:48:04 +1200 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: <19970526164217Z12887-19601+191@nic.funet.fi> References: from "Glenn Anderson" at May 23, 97 04:11:01 pm Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 27 May 1997 08:48:04 +1200 To: Matti Aarnio From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: > I suspect the sender sends more than the exact number of bytes > it is claims to send, as otherwise it would get reply pipeline > flush... > > Could you try to send to my office workstation ? To address: > mea@mea.tmt.tele.fi > My system has some extra log tests in it to see what happens. > It will also flush the reply pipe in every case, so your end > should get a report back. Whatever you have done, it fixed the problem. I had no problem sending messages to that address using BDAT. No error messages were reported either. > The problem with BDAT is that it is EXTREMELY sensitive to > the correct count of transferred bytes. My pipeline optimi- > zation in the BDAT processing was such that if the transfer > was successfull, and the input is not empty, then there is > no immediate "replyflush". As EIMS works internally with messages internally stored in raw RFC 822 format, all it does is send the length of the message in the BDAT command, then send the message. It leaves no posibility of the size being wrong. It was actually a lot easier to implement BDAT than the normal transfer method as there is no need to pad lines starting with a '.' or search for . at the end. > By the way, how many systems in the network have BDAT service > at all ? EIMS 1.x supports sending using BDAT, but not receiving. Future versions will support receiving as well. I don't know of any other systems that use it. Let me know if you want to do any compatibility testing in future. > I tried to count the number of chars in the message you logged, > did I get it right, or did I under/overshoot a couple bytes ? > At the end there are two Xes after your signature, of which I > don't know, were they original CRLFs in the message ? > (The "M" at the end of the line represents CR of the CRLF pair.) Yes, there are two CRLF pairs at the end where the Xes are. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Mon May 26 23:56:23 1997 Received: from NO-IDENT-SERVICE@port432-well.ihug.co.nz (port 4360 [202.49.241.178]) by nic.funet.fi with ESMTP id <1328-19601>; Mon, 26 May 1997 23:55:34 +0300 Received: from [202.49.241.178] (127.0.0.1) by port432-well.ihug.co.nz with ESMTP (Apple Internet Mail Server 2.0b16); Tue, 27 May 1997 08:55:18 +1200 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: <19970526164217Z12887-19601+191@nic.funet.fi> References: from "Glenn Anderson" at May 23, 97 04:11:01 pm Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 27 May 1997 08:55:18 +1200 To: Matti Aarnio From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: Some further info, I put some extra debugging code into my server here, and when sending a message to zmailer@nic.funet.fi or rcb@press-gopher.uchicago.edu the receiving ZMailer server closes it's end of the connection after receiving the message. This may explain why the response is not being returned. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Mon May 26 23:57:25 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 55816 [128.214.46.64]) by nic.funet.fi with ESMTP id <1203-1427>; Mon, 26 May 1997 23:56:51 +0300 Received: from port432-well.ihug.co.nz (port432-well.ihug.co.nz [202.49.241.178]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-3.12) with ESMTP id XAA02627; Mon, 26 May 1997 23:56:36 +0300 (EET DST) Received: from [202.49.241.178] (127.0.0.1) by port432-well.ihug.co.nz with ESMTP (Apple Internet Mail Server 2.0b16); Tue, 27 May 1997 08:55:18 +1200 X-Sender: glenn@mail.aec.co.nz Message-Id: In-Reply-To: <19970526164217Z12887-19601+191@nic.funet.fi> References: from "Glenn Anderson" at May 23, 97 04:11:01 pm Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 27 May 1997 08:55:18 +1200 To: Matti Aarnio From: Glenn Anderson Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com Return-Path: Some further info, I put some extra debugging code into my server here, and when sending a message to zmailer@nic.funet.fi or rcb@press-gopher.uchicago.edu the receiving ZMailer server closes it's end of the connection after receiving the message. This may explain why the response is not being returned. Glenn. ------------------------------------------------------------- Glenn Anderson glenn@aims.aec.co.nz Macintosh Internet Programmer Phone (mobile) +64 25 518661 Eudora Internet Mail Server: ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ ------------------------------------------------------------- From zmailer-owner@nic.funet.fi Tue May 27 11:42:46 1997 Received: from root@mea.tmt.tele.fi (port 39762 [194.251.252.33]) by nic.funet.fi with ESMTP id <13252-29284>; Tue, 27 May 1997 11:42:37 +0300 Received: from mea@localhost by mea.tmt.tele.fi id <231510-23914>; Tue, 27 May 1997 11:42:23 +0300 Subject: Re: BDAT test 2 To: glenn@aims.aec.co.nz (Glenn Anderson) Date: Tue, 27 May 1997 11:42:19 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from Glenn Anderson at "May 27, 97 08:34:21 am" Sender: Matti Aarnio From: Matti Aarnio Phone: +358-20402082 (office, with redirection to cellular) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1711 Message-Id: <19970527084223Z231510-23914+4@mea.tmt.tele.fi> Return-Path: (Cc: ZMailer list) Ok, both BDAT test transmissions were successfull. Here is what the SMTP-server logged: 24421r BDAT 787 LAST 24421w 250 2.6.0 S.nWTAX231510 Roger, got 787 bytes in the last chunk, stored 1057 bytes into spool 24421# -- pipeline input: 0 bytes That is, the BDAT data was precisely sized. This version of ZMailer I am running has unconditional reply flush at the end of the BDAT processing. ZMailer 2.99.49 release will have it too. However I still don't understand, why the conditional reply flush didn't happen with "pipeline input: 0 bytes" condition at the end of the BDAT subroutine (as it should have..) Codehackers info: At the end of the subroutine smtp_bdata() (smtpserver/smtpdata.c, or smtpserver/smtpserver.c before the source split..) there is a test: if (!s_hasinput(SS)) typeflush(SS); Remove the test, run the typeflush() unconditionally. (At the printout '0 bytes' means that test succeeds, because it is the return value of the s_hasinput() routine..) /Matti Aarnio > From glenn@aims.aec.co.nz Mon May 26 23:35:21 1997 > X-Sender: glenn@mail.aec.co.nz > Message-Id: > Date: Tue, 27 May 1997 08:34:21 +1200 > To: mea@mea.tmt.tele.fi > From: Glenn Anderson > Subject: BDAT test 2 > > testing. > > ------------------------------------------------------------- > Glenn Anderson glenn@aims.aec.co.nz > Macintosh Internet Programmer Phone (mobile) +64 25 518661 > Eudora Internet Mail Server: > ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ > ------------------------------------------------------------- > > From zmailer-owner@nic.funet.fi Tue May 27 12:06:53 1997 Received: from TIMEDOUT@falco.its.uu.se (port 18705 [130.238.131.113]) by nic.funet.fi with ESMTP id <12697-7094>; Tue, 27 May 1997 12:06:38 +0300 Received: from alba.its.uu.se ([130.238.131.102]) by falco.its.uu.se with ESMTP id <2056-25638>; Tue, 27 May 1997 11:06:09 +0100 X-Mailer: exmh version 2.0gamma 1/27/96 To: Matti Aarnio cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se Subject: Re: Bug in sm still truncates messages on AIX (New bug) In-reply-to: Your message of "Fri, 23 May 1997 22:33:19 +0300." <19970523193323Z1316-13715+253@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 27 May 1997 11:05:58 +0200 From: Martin Wendel Message-Id: <97May27.110609+0100msz.2056-25638+1@falco.its.uu.se> Return-Path: > ... Umm.. Ah.. check_7bit_cleanness() may be the culprit here. > It leaves the loop sometimes before it can read until the EOF, > and I presume that your emails are with 8-bit chars, and/or > you have "Content-Conversion: prohibited" headers... Yes, that's correct. And this also explains why readalready always was set to sizeof(let_buffer) when the error occurred. The same bug seems to be in the other transports aswell (smtp, mailbox). -- E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS Tel: 018 18 77 80 Box 887 Int: +46 18 18 77 80 S-751 08 Uppsala Fax: +46 18 18 79 10 Sweden http://www.uu.se/Internservice/ue/personal/MartinWendel.html From zmailer-owner@nic.funet.fi Tue May 27 12:52:49 1997 Received: by nic.funet.fi id <12671-7094>; Tue, 27 May 1997 12:52:37 +0300 Subject: Re: Bug in sm still truncates messages on AIX (New bug) From: Matti Aarnio To: martin.wendel@its.uu.se (Martin Wendel) Date: Tue, 27 May 1997 12:52:33 +0300 (EET DST) Cc: zmailer@nic.funet.fi, Torbjorn.Wictorin@its.uu.se In-Reply-To: <97May27.110609+0100msz.2056-25638+1@falco.its.uu.se> from "Martin Wendel" at May 27, 97 11:05:58 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 862 Message-Id: <19970527095237Z12671-7094+164@nic.funet.fi> Return-Path: > > ... Umm.. Ah.. check_7bit_cleanness() may be the culprit here. > > It leaves the loop sometimes before it can read until the EOF, > > and I presume that your emails are with 8-bit chars, and/or > > you have "Content-Conversion: prohibited" headers... > > Yes, that's correct. And this also explains why readalready always > was set to sizeof(let_buffer) when the error occurred. The same > bug seems to be in the other transports aswell (smtp, mailbox). I know. Fixed in 2.99.49b6 available at my workstation. > -- > E-Mail: Martin.Wendel@ITS.UU.SE Snail-Mail: ITS > Tel: 018 18 77 80 Box 887 > Int: +46 18 18 77 80 S-751 08 Uppsala > Fax: +46 18 18 79 10 Sweden > http://www.uu.se/Internservice/ue/personal/MartinWendel.html /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 27 15:25:50 1997 Received: by nic.funet.fi id <12537-23540>; Tue, 27 May 1997 15:24:47 +0300 Subject: Re: duplicate email from Apple/Eudora Internet servers (fwd) From: Matti Aarnio To: glenn@aims.aec.co.nz (Glenn Anderson) Date: Tue, 27 May 1997 15:24:38 +0300 (EET DST) Cc: rcb@press-gopher.uchicago.edu, zmailer@nic.funet.fi, eims-bugs@qualcomm.com In-Reply-To: from "Glenn Anderson" at May 27, 97 08:55:18 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 969 Message-Id: <19970527122447Z12537-23540+160@nic.funet.fi> Return-Path: > Some further info, I put some extra debugging code into my server here, and > when sending a message to zmailer@nic.funet.fi or > rcb@press-gopher.uchicago.edu the receiving ZMailer server closes it's end > of the connection after receiving the message. This may explain why the > response is not being returned. Indeed. It gets a SIGSEGV when it tries to do: printf( "%s", some_integer ) My newer betas have different spoolid reporting schemes, but apparently both nic.funet.fi, and press-gopher are running a bit older edition. Funet machine upgrades in a moment.. > Glenn. > > ------------------------------------------------------------- > Glenn Anderson glenn@aims.aec.co.nz > Macintosh Internet Programmer Phone (mobile) +64 25 518661 > Eudora Internet Mail Server: > ftp://ftp.eudora.com/eudora/servers/mac/eudora_ims/ > ------------------------------------------------------------- /Matti Aarnio From zmailer-owner@nic.funet.fi Tue May 27 18:59:23 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 46033 [194.67.3.135]) by nic.funet.fi with ESMTP id <12854-22645>; Tue, 27 May 1997 18:58:59 +0300 Received: from NO-IDENT-SERVICE@ariel (port 41743 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68048-314>; Tue, 27 May 1997 19:58:30 +0400 Date: Tue, 27 May 1997 19:58:16 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: probs in .49 To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: First and not really important: now `zmailer' script tries to use `echo -n' when reporting start progress. This dows not work in many echos around, e.g. with Solaris /bin/echo. Second: if you configure smtpserver to verify addresses, it would not accept mail anything: crosser@chronos:/src/zmailer/zmailer-2.99.49b6/$ telnet ariel 25 Trying 194.67.3.216... Connected to ariel. Escape character is '^]'. 220 ariel ZMailer Server 2.99.49b6 #1 ESMTP+IDENT (nulltrans) ready at Tue, 27 May 1997 19:54:52 +0400 helo chronos 250 ariel Hello chronos mail from: got EOF! got EOF! 500 **INTERNAL*ERROR** Eugene From zmailer-owner@nic.funet.fi Tue May 27 19:53:59 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 48338 [194.67.3.135]) by nic.funet.fi with ESMTP id <12439-18610>; Tue, 27 May 1997 19:53:46 +0300 Received: from NO-IDENT-SERVICE@ariel (port 41851 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68049-29127>; Tue, 27 May 1997 20:53:19 +0400 Date: Tue, 27 May 1997 20:53:09 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: And the main problem with .49 - smtp still does not work at all To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: with beta6, smtp still does not work with the same symptoms: May 27 20:43:37 ariel smtp[21201]: S.nWkwY23343: to=, delay=00:00:21, xdelay=00:00:01, mailer=smtp, stat=error2 smtp; 500 (nameserver data inconsistency. No MX, no address: 'online.ru', errno=Error 0) Solaris 2.5.1 with gcc. Eugene From zmailer-owner@nic.funet.fi Wed May 28 12:05:07 1997 Received: by nic.funet.fi id <2363-17318>; Wed, 28 May 1997 12:03:55 +0300 Subject: zmailer-2.99.48.patch3.test released ... From: Matti Aarnio To: rcb@press-gopher.uchicago.edu (Roy C Bixler) Date: Wed, 28 May 1997 12:03:54 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Roy C Bixler" at May 27, 97 04:10:52 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1365 Message-Id: <19970528090355Z2363-17318+172@nic.funet.fi> Return-Path: ... > Does this mean another place to patch in the 2.99.48p2 source tree or can > I ignore this? I already placed the patch you suggested in your earlier > e-mail in the 'smtpdata.c' source file and that does solve the problem. > Right now, because of the reported problems with 2.99.49b6, I am inclined > to stand pat with 2.99.48p2+. In fact, IMHO, it would be a good idea to > put out a 3rd patch for 2.99.48 with the 'smtpdata.c' fix. Anyway, as > usual, thanks for the good responce! Ok, I made zmailer-2.99.48.patch3.test available at my workstation ( ftp://mea.tmt.tele.fi/ ). I did compile test it at my workstation both with and without HAVE_MMAP... Report on its working, and I will make official release by storeing it into ftp.funet.fi. * config.h.in, configure.in, configure: - Pulled these from 2.99.49beta7 * transports/sm/sm.c, transports/smtp/smtp.c, transports/mailbox/mailbox.c: "readalready" cache handling in case the MMAP is not recognized properly (IBM AIX has problems at the auto- detection of that feature..) * smtpserver/smtpserver.c, smtpserver/smtpcmds.c, smtpserver/smtpdata.c: - Wrong parameters in one logging call at the BDAT processing, - Rearranged typeflush() calls, and pipeline status reporting. > Roy > rcb@press-gopher.uchicago.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Wed May 28 19:30:11 1997 Received: by nic.funet.fi id <3043-17318>; Wed, 28 May 1997 19:28:50 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49beta7 at my workstation... Message-Id: <19970528162850Z3043-17318+318@nic.funet.fi> Date: Wed, 28 May 1997 19:28:42 +0300 Return-Path: ... with emphasis on trying to get (with extensive debug logging) getaddrinfo() to fail. Initially I did see failures, but after adding the logging features, I haven't been able to repeat it :-( With it, "mailx -v user@somewhere" logs a lot of resolver things to the verbose trace at the submitters screen. I doubt that simply adding the logging routines have altered behaviour to any extent, but I can't quite guarantee it. (Especially as such would require same GCC bug at intel, AND at Sparc systems...) Eugene ? /Matti Aarnio From zmailer-owner@nic.funet.fi Thu May 29 00:36:09 1997 Received: from NO-IDENT-SERVICE@fddisunsite.oit.unc.edu (port 2507 [152.2.254.81]) by nic.funet.fi with ESMTP id <2190-13499>; Thu, 29 May 1997 00:35:36 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by calzone.oit.unc.edu (8.8.5/8.8.5) with ESMTP id RAA07873 for ; Wed, 28 May 1997 17:35:28 -0400 (EDT) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id NAA00398 for ; Wed, 28 May 1997 13:03:17 -0400 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Wed, 28 May 1997 13:03:17 -0400 (EDT) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: zmailer@nic.funet.fi Subject: db question and portability.h patch Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hiya. I'm trying to compile zmailer-2.99.48p2 on a Solaris 2.6 host running gcc-2.7.2.2 and db-1.8.5. My db.h file lives in /usr/local/include, but configure doesn't find it when searching for database packages. I've edited config.h by hand and tried to compile, but to no avail. Any suggestions? Oh, I had trouble with libresolv/portability.h (using --with-bundled-libresolv option to configure). Turns out I needed to comment out the following section: /* #if defined(SVR4) && defined(sun) extern int gethostname __P((char *, size_t)); #endif */ - donald From zmailer-owner@nic.funet.fi Thu May 29 01:27:35 1997 Received: by nic.funet.fi id <3004-5038>; Thu, 29 May 1997 01:27:13 +0300 Subject: Re: db question and portability.h patch From: Matti Aarnio To: balld@sunsite.unc.edu (Donald Ball) Date: Thu, 29 May 1997 01:26:58 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Donald Ball" at May 28, 97 01:03:17 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1219 Message-Id: <19970528222713Z3004-5038+22@nic.funet.fi> Return-Path: > Hiya. I'm trying to compile zmailer-2.99.48p2 on a Solaris 2.6 host > running gcc-2.7.2.2 and db-1.8.5. My db.h file lives in > /usr/local/include, but configure doesn't find it when searching for > database packages. I've edited config.h by hand and tried to compile, but > to no avail. Any suggestions? Solaris 2.6 ? Early-Access beta ? I have to ask an account at an Sol2.6EA machine we have at my work.. I didn't think there was any high priority at getting the ZMailer running in it :) Anyway, the autoconf(ig) isn't very good at finding things automatically when they are in abnormal positions. "abnormal" in this context is /usr/local/ instead of /usr/ ... How many alternate trees should be supported automatically, and when to add one more --with-bsd-db-include= --with-bsd-db-lib= or some such ? > Oh, I had trouble with libresolv/portability.h (using > --with-bundled-libresolv option to configure). Turns out I needed to > comment out the following section: Yes, there has been talk about it in the bind-workders list. > /* > #if defined(SVR4) && defined(sun) > extern int gethostname __P((char *, size_t)); > #endif > */ > > - donald /Matti Aarnio -- wishing you good night From zmailer-owner@nic.funet.fi Thu May 29 03:57:01 1997 Received: from NO-IDENT-SERVICE@fddisunsite.oit.unc.edu (port 37875 [152.2.254.81]) by nic.funet.fi with ESMTP id <2136-5038>; Thu, 29 May 1997 03:56:51 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by calzone.oit.unc.edu (8.8.5/8.8.5) with ESMTP id UAA02068; Wed, 28 May 1997 20:56:36 -0400 (EDT) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id UAA01566; Wed, 28 May 1997 20:56:32 -0400 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Wed, 28 May 1997 20:56:32 -0400 (EDT) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: db question and portability.h patch In-Reply-To: <19970528222713Z3004-5038+22@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: > Solaris 2.6 ? Early-Access beta ? > I have to ask an account at an Sol2.6EA machine we > have at my work.. I didn't think there was any > high priority at getting the ZMailer running in it :) Yeah, we're on the EA program. Solaris 2.6 seems pretty decent so far, I've had more troubles with gcc-2.7.2.2... web server performance in particular is quite a bit better than Solaris 2.5.1, though I haven't compared to 2.5.1+ISS. Anyway... > Anyway, the autoconf(ig) isn't very good at finding > things automatically when they are in abnormal positions. > "abnormal" in this context is /usr/local/ instead > of /usr/ ... How many alternate trees should be > supported automatically, and when to add one more > --with-bsd-db-include= > --with-bsd-db-lib= > or some such ? Well, the problem appears to be a bit more subtle than I originally thought - I copied the db.h file to /usr/include just for kicks and the configure script still doesn't find it! I had to mangle the db.h file a little bit to get db-1.8.5 to compile with gcc-2.7.2.2 (problem with a new int_types.h file - actually, I get warnings about it when compiling zmailer too but no errors), but not too seriously. What can I do here? - donald From zmailer-owner@nic.funet.fi Thu May 29 08:57:22 1997 Received: by nic.funet.fi id <2197-16542>; Thu, 29 May 1997 08:57:14 +0300 Subject: Re: db question and portability.h patch From: Matti Aarnio To: balld@sunsite.unc.edu (Donald Ball) Date: Thu, 29 May 1997 08:57:04 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Donald Ball" at May 28, 97 08:56:32 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1109 Message-Id: <19970529055714Z2197-16542+104@nic.funet.fi> Return-Path: ... > > Anyway, the autoconf(ig) isn't very good at finding > > things automatically when they are in abnormal positions. > > "abnormal" in this context is /usr/local/ instead > > of /usr/ ... How many alternate trees should be > > supported automatically, and when to add one more > > --with-bsd-db-include= > > --with-bsd-db-lib= > > or some such ? > > Well, the problem appears to be a bit more subtle than I originally > thought - I copied the db.h file to /usr/include just for kicks and the > configure script still doesn't find it! I had to mangle the db.h file a > little bit to get db-1.8.5 to compile with gcc-2.7.2.2 (problem with a new > int_types.h file - actually, I get warnings about it when compiling > zmailer too but no errors), but not too seriously. What can I do here? The autoconfig tests to do several linkings too to verify that proper library is available. Location for libdb.a/libdb.so is propably: /lib Browsing thru config.log should tell you what did happen during the configuration process. > - donald /Matti Aarnio -- wishing you good morning From zmailer-owner@nic.funet.fi Thu May 29 18:38:31 1997 Received: from NO-IDENT-SERVICE@outbound.Princeton.EDU (port 41869 [128.112.128.84]) by nic.funet.fi with ESMTP id <2115-13499>; Thu, 29 May 1997 18:38:15 +0300 Received: from lists.Princeton.EDU ([128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <541264-2996>; Thu, 29 May 1997 11:37:55 -0400 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370079.s2-1>; Thu, 29 May 1997 11:37:45 -0400 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id LAA13294 for ; Thu, 29 May 1997 11:37:41 -0400 Message-Id: <199705291537.LAA13294@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: problem with MX handling Organization: Princeton University, Network Systems, Class. 56 X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 29 May 1997 11:37:41 EDT From: "Michael R. Gettes" Return-Path: I have zmailer-2.99.45 on solaris-2.5.1. I am sending to foobar@kff.org [ip=205.187.85.60, and an MX of tfs_gate.kff.org]. zmailer seems to NOT honor the MX and only send to the ip address above. Does anyone else see this behavior? Is this fixed in a later version?? Thanks! /mrg From zmailer-owner@nic.funet.fi Thu May 29 22:52:42 1997 Received: from NO-IDENT-SERVICE@outbound.Princeton.EDU (port 54187 [128.112.128.84]) by nic.funet.fi with ESMTP id <2200-13499>; Thu, 29 May 1997 22:49:13 +0300 Received: from IDENT-NOT-QUERIED@lists.Princeton.EDU (port 3028 [128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <541762-28172>; Thu, 29 May 1997 15:48:54 -0400 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370095.s1-2>; Thu, 29 May 1997 15:48:35 -0400 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id PAA05817 for ; Thu, 29 May 1997 15:48:34 -0400 Message-Id: <199705291948.PAA05817@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: Re: problem with MX handling Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 29 May 1997 15:48:34 EDT From: "Michael R. Gettes" Return-Path: On Thu, 29 May 1997 12:14:12 -0500 rcb said: > On Thu, 29 May 1997, Michael R. Gettes wrote: > > I have zmailer-2.99.45 on solaris-2.5.1. > > > > I am sending to foobar@kff.org [ip=205.187.85.60, and an MX of > > tfs_gate.kff.org]. zmailer seems to NOT honor the MX and only > > send to the ip address above. > > It works properly under Z-mailer 2.99.48patch3. > Roy > rcb@press-gopher.uchicago.edu > I got a response from Roy Bixler saying that it worked on 2.99.48.p3 so, I installed this version... there were existing mails awaiting delivery to kff.org... and under 2.99.48.p3 they are NOT transmitting. I would appreciate any clues on how to proceed. /mrg From zmailer-owner@nic.funet.fi Thu May 29 23:37:57 1997 Received: by nic.funet.fi id <2519-5038>; Thu, 29 May 1997 23:33:05 +0300 Subject: Re: problem with MX handling From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Thu, 29 May 1997 23:32:58 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199705291537.LAA13294@yo.Princeton.EDU> from "Michael R. Gettes" at May 29, 97 11:37:41 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 543 Message-Id: <19970529203305Z2519-5038+306@nic.funet.fi> Return-Path: > I have zmailer-2.99.45 on solaris-2.5.1. > > I am sending to foobar@kff.org [ip=205.187.85.60, and an MX of > tfs_gate.kff.org]. zmailer seems to NOT honor the MX and only > send to the ip address above. > > Does anyone else see this behavior? Is this fixed in a later > version?? Have a look again at the MX target... The underscore is invalid in the DNS objects, and some newer resolvers enforce that rule yielding no address (propably an error) when doing getaddrbyname() > Thanks! > > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 30 00:39:53 1997 Received: from NO-IDENT-SERVICE@fddisunsite.oit.unc.edu (port 60378 [152.2.254.81]) by nic.funet.fi with ESMTP id <2402-461>; Fri, 30 May 1997 00:36:45 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by calzone.oit.unc.edu (8.8.5/8.8.5) with ESMTP id RAA13618; Thu, 29 May 1997 17:36:37 -0400 (EDT) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id RAA05615; Thu, 29 May 1997 17:36:28 -0400 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Thu, 29 May 1997 17:36:28 -0400 (EDT) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: How to change delivery agent in zmailer In-Reply-To: <19970529203305Z2519-5038+306@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: So I finally got zmailer compiled and installed (although I did have to tweak the Makefile in the man directory a teensy bit - changed "-m444" to "-m 444"). Now I want to change the delivery agent so that it will deliver to cyrus imapd instead of mailboxes in /var/mail. There apparantly is a way, since there's a small mention of it in pm.conf, but very little in the way of actual documentation. Anyone else out there have zmailer configured for use with cyrus? What's the secret? - donald From zmailer-owner@nic.funet.fi Fri May 30 14:56:11 1997 Received: by nic.funet.fi id <2768-461>; Fri, 30 May 1997 14:55:26 +0300 Subject: Re: How to change delivery agent in zmailer From: Matti Aarnio To: balld@sunsite.unc.edu (Donald Ball) Date: Fri, 30 May 1997 14:55:22 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Donald Ball" at May 29, 97 05:36:28 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1607 Message-Id: <19970530115526Z2768-461+266@nic.funet.fi> Return-Path: > So I finally got zmailer compiled and installed (although I did have to > tweak the Makefile in the man directory a teensy bit - changed "-m444" to > "-m 444"). Now I want to change the delivery agent so that it will > deliver to cyrus imapd instead of mailboxes in /var/mail. There apparantly > is a way, since there's a small mention of it in pm.conf, but very little > in the way of actual documentation. Anyone else out there have zmailer > configured for use with cyrus? What's the secret? At the zmailer-2.99.48*/proto/scheduler.conf file there is: ------- local/* interval=5m idlemax=9m # Originally we had 3 hour expiry, but if your local system goes to # a fizz (freezes, that is), your local mail may start to bounce # before anybody notices anything... expiry=3d # want 20 channel slots, but only one HOST maxchannel=15 maxring=5 # # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! command="mailbox -8" # Or with CYRUS server the following might do: #command="sm -8c $channel cyrus" # Or with PROCMAIL as the local delivery agent: #command="sm -8c $channel procm" ------- You have the choice... At the proto/sm.conf is the ultimate running method to run the cyrus, but to get that far, it is necessary to choose right way to schedule it at first. Do note that at the scheduler.conf there are also channels to handle local pipes, and local file writes independently from the general use mailbox store. That because neither the procmail, nor cyrus are very happy to do them... > - donald /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 30 19:08:02 1997 Received: from NO-IDENT-SERVICE@mercury.uniserve.com (port 35086 [204.191.197.248]) by nic.funet.fi with ESMTP id <16249-27824>; Fri, 30 May 1997 19:06:03 +0300 Received: from kst ([204.244.242.135]) by mercury.uniserve.com with ESMTP id JAA10438 for ; Fri, 30 May 1997 09:05:50 -0700 (PDT) Message-Id: <199705301605.JAA10438@mercury.uniserve.com> From: "Kevin Traas" To: "zmailer" Subject: Config Question Date: Fri, 30 May 1997 08:56:42 -0700 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: These are probably pretty simple questions; however, I'm quite new to ZMailer, so even a pointer to the right FM would be appreciated.... I'm currently running 2.99.48p2 on Debian GNU/Linux. All is working fine; however, I've got two things I'd like to change: 1. I'd like to hide the system-name on outgoing mail. ZMailer is using the FQDN - how do I change that? 2. On my old system (SCO Unix), the mail system was configured to recognise when multiple copies of the same message were being sent to the same addressee and would, therefore, suppress additional copies. i.e. if I were a member of two mailing lists on the local system and someone send a message addressed to me and to both lists, I'd receive only one copy of the message. Currently, on my new system, I am receiving three messages. Can ZMailer be configured to do this? Thanks for any help, Kevin Traas Systems Analyst Edmondson Roper CA http://www.eroper.bc.ca From zmailer-owner@nic.funet.fi Fri May 30 19:37:03 1997 Received: by nic.funet.fi id <2974-4234>; Fri, 30 May 1997 19:33:49 +0300 Subject: Re: Config Question From: Matti Aarnio To: KTraas@uniserve.com (Kevin Traas) Date: Fri, 30 May 1997 19:27:40 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199705301605.JAA10438@mercury.uniserve.com> from "Kevin Traas" at May 30, 97 08:56:42 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1506 Message-Id: <19970530163349Z2974-4234+214@nic.funet.fi> Return-Path: > These are probably pretty simple questions; however, I'm quite new to > ZMailer, so even a pointer to the right FM would be appreciated.... > > I'm currently running 2.99.48p2 on Debian GNU/Linux. All is working fine; > however, I've got two things I'd like to change: > > 1. I'd like to hide the system-name on outgoing mail. ZMailer is using > the FQDN - how do I change that? INSTALL -file, paragraph 7 speaking of 'mydomain' Also note paragraph 8 speaking of 'localnames' file. > 2. On my old system (SCO Unix), the mail system was configured to > recognise when multiple copies of the same message were being sent to the > same addressee and would, therefore, suppress additional copies. i.e. if I > were a member of two mailing lists on the local system and someone send a > message addressed to me and to both lists, I'd receive only one copy of the > message. Currently, on my new system, I am receiving three messages. Can > ZMailer be configured to do this? This is more hairy situation. It does the suppression in case the lists are ZMailer internal, but if any of them involves running external programs (i.e. Majordomo), they are entirely separate entities. To answer fully, I do need to know your list setup. Are the lists in the $MAILSHARE/lists/ -directory, or somewhere else ? If they are elsewere, where ? How ? > Thanks for any help, > > Kevin Traas > Systems Analyst > Edmondson Roper CA > http://www.eroper.bc.ca /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 30 19:51:06 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 6550 [194.67.3.135]) by nic.funet.fi with ESMTP id <2305-4234>; Fri, 30 May 1997 19:46:37 +0300 Received: from NO-IDENT-SERVICE@ariel (port 48600 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68050-10488>; Fri, 30 May 1997 20:34:45 +0400 Date: Fri, 30 May 1997 20:34:40 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: mailbox in .49b7 To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Matti, '-H' option in mailbox program (allow 8bit in headers) stopped working. Moreover, attemped QP encoding is incorrect: *any* 8bit character is converted to '=FF'! Have to defer installing in on the main production relay... Also, noticed a strange thing: a few messages are stuck in the scheduler queue for smtp delivery, and mailq only shows them occationaly. Here are 'mailq' commands run *immediately* one after one: crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq smtp/it.ru: 68047-25772: (retry in 4s, expires in 2d22h, tries=1) crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq smtp/nic.funet.fi: 68060-25772: (retry in 52s, expires in 1d20h, tries=1) crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq smtp/bm.com: 68058-25772: (retry in 1m, expires in 1d20h, tries=1) smtp/nic.funet.fi: 68060-25772: (retry in 46s, expires in 1d20h, tries=1) crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ Eugene From zmailer-owner@nic.funet.fi Fri May 30 19:51:23 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 6550 [194.67.3.135]) by nic.funet.fi with ESMTP id <2076-27824>; Fri, 30 May 1997 19:46:23 +0300 Received: from NO-IDENT-SERVICE@ariel (port 46148 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68060-25772>; Thu, 29 May 1997 17:23:42 +0400 Date: Thu, 29 May 1997 17:23:32 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: mailbox -l gone? To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: is it intentional that -l option does not work in .49? Eugene From zmailer-owner@nic.funet.fi Fri May 30 20:27:56 1997 Received: by nic.funet.fi id <2286-4234>; Fri, 30 May 1997 20:25:26 +0300 Subject: Re: mailbox in .49b7 From: Matti Aarnio To: crosser@online.ru Date: Fri, 30 May 1997 20:25:17 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 30, 97 08:34:40 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1971 Message-Id: <19970530172526Z2286-4234+219@nic.funet.fi> Return-Path: > Matti, > > '-H' option in mailbox program (allow 8bit in headers) stopped working. > Moreover, attemped QP encoding is incorrect: *any* 8bit character is > converted to '=FF'! Have to defer installing in on the main production > relay... Propably character signedness problem. A diff at the end. (I wonder if any others are lurking in the corners..) (mailbox, sm, and smtp programs are armoured ok, but I can't quickly verify all of their libraries..) > Also, noticed a strange thing: a few messages are stuck in the scheduler > queue for smtp delivery, and mailq only shows them occationaly. Here > are 'mailq' commands run *immediately* one after one: > > crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq > smtp/it.ru: > 68047-25772: (retry in 4s, expires in 2d22h, tries=1) > crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq > smtp/nic.funet.fi: > 68060-25772: (retry in 52s, expires in 1d20h, tries=1) > crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ mailq > smtp/bm.com: > 68058-25772: (retry in 1m, expires in 1d20h, tries=1) > smtp/nic.funet.fi: > 68060-25772: (retry in 46s, expires in 1d20h, tries=1) > crosser@chronos:/src/zmailer/zmailer-2.99.49b7/$ The $LOGDIR/scheduler -logfile does tell something. All queue dropouts are listed there. > Eugene /Matti Aarnio --- transports/libta/mime2headers.c~ Fri May 23 01:17:15 1997 +++ transports/libta/mime2headers.c Fri May 30 20:07:46 1997 @@ -251,12 +251,13 @@ q += strlen(q); for ( ; *s && (*s != ' ' && *s != '\t' && *s != ')'); ++s) { - if (*s < ' ' || *s > 126 || - *s == '=' || *s == '?' || *s == '_') { - sprintf(q,"=%02X",*s); + int c = (*s) & 0xFF; + if (c < ' ' || c > 126 || + c == '=' || c == '?' || c == '_') { + sprintf(q, "=%02X", c); q += 3; } else - *q++ = *s; + *q++ = c; } strcpy(q,"?="); q += 2; strcpy(q,(void*)s); From zmailer-owner@nic.funet.fi Fri May 30 20:42:30 1997 Received: by nic.funet.fi id <2563-22436>; Fri, 30 May 1997 20:38:18 +0300 Subject: Re: mailbox -l gone? From: Matti Aarnio To: crosser@online.ru Date: Fri, 30 May 1997 20:38:04 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 29, 97 05:23:32 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 759 Message-Id: <19970530173818Z2563-22436+201@nic.funet.fi> Return-Path: > is it intentional that -l option does not work in .49? There is only one place where the log-event is done, and it means the flow must go thru it every time... Thus perhaps following will do it ? > Eugene /Matti Aarnio --- transports/mailbox/mailbox.c~ Mon May 26 20:41:31 1997 +++ transports/mailbox/mailbox.c Fri May 30 20:35:08 1997 @@ -1667,6 +1667,9 @@ fsync(fdmail); #endif + + time_reset: + /* This is actually Linux-thing, we reset the last access time to be that of before we did read the file, last modification into current moment -- At Linux 1.3.8x the system is a bit @@ -1688,8 +1691,6 @@ #endif fflush(logfp); } - - time_reset: if (*(rp->addr->user) != TO_PIPE && keepatime) { #ifdef HAVE_UTIME From zmailer-owner@nic.funet.fi Fri May 30 20:46:16 1997 Received: from TIMEDOUT@quena.rcp.net.pe (port 15286 [161.132.12.3]) by nic.funet.fi with SMTP id <16288-22436>; Fri, 30 May 1997 20:44:25 +0300 Received: by quena.rcp.net.pe via sendmail with stdio id for zmailer@nic.funet.fi; Fri, 30 May 97 12:38:59 -0400 (EDT) (/\##/\ Smail3.1.30.13 #30.5 built 30-sep-96) Message-Id: From: vadillo@quena.rcp.net.pe (Enrique Vadillo) Subject: Error: norealname ?!?!? To: zmailer@nic.funet.fi Date: Fri, 30 May 1997 12:38:59 -0400 (EDT) Cc: crosser@average.org, mea@mea.tmt.tele.fi, tom@sdf.com, ken@spry.com, mss1@cornell.edu PGP-FingerPrint: 55 B9 83 D2 61 71 E6 6B 1E CE FD B5 F7 AA F1 B5 X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1072 Return-Path: Hello, I am getting this error message after successfully installing Zmailer-2.99.48: [ayllu/usr/zmail/zmailer-2.99.48]# sendmail -bt ZMailer router (2.99.48 #1: Thu May 29 20:48:26 CST 1997) vadillo@ayllu:/usr/zmail/zmailer-2.99.48/router Copyright 1992 Rayan S. Zachariassen Copyright 1992-1997 Matti Aarnio z# router operator : address: operator (((local operator operator@ayllu.rcp.net.pe default_attributes))) z# router operator@www.rcp.net.pe : address: operator@www.rcp.net.pe newsgroup: not found (((error norealname operator@www.rcp.net.pe default_attributes))) z# I don't manage or have newsgroups installed in my system, i don't even want newsgroups suport of any kind, why do i get this error message? how can i get thru it? nntp is something i don't use here at all. If you have any clue, please tip me. I will appreciate a LOT. Enrique Vadillo- -- RCP - Internet Peru Fax: +51 1 241-1320 Web Site: http://www.rcp.net.pe (PERU) Mirror Web Site: http://ekeko.rcp.net.pe (USA) From zmailer-owner@nic.funet.fi Fri May 30 20:47:42 1997 Received: from NO-IDENT-SERVICE@mercury.uniserve.com (port 10760 [204.191.197.248]) by nic.funet.fi with ESMTP id <2431-27824>; Fri, 30 May 1997 20:45:37 +0300 Received: from kst ([204.244.242.135]) by mercury.uniserve.com with ESMTP id KAA29664; Fri, 30 May 1997 10:45:26 -0700 (PDT) Message-Id: <199705301745.KAA29664@mercury.uniserve.com> From: "Kevin Traas" To: "Matti Aarnio" Cc: Subject: Re: Config Question Date: Fri, 30 May 1997 10:38:44 -0700 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1155 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-Path: > > 2. On my old system (SCO Unix), the mail system was configured to > > recognise when multiple copies of the same message were being sent to the > > same addressee and would, therefore, suppress additional copies. i.e. if I > > were a member of two mailing lists on the local system and someone send a > > message addressed to me and to both lists, I'd receive only one copy of the > > message. Currently, on my new system, I am receiving three messages. Can > > ZMailer be configured to do this? > > This is more hairy situation. > It does the suppression in case the lists are ZMailer internal, but > if any of them involves running external programs (i.e. Majordomo), > they are entirely separate entities. > > To answer fully, I do need to know your list setup. Are the lists > in the $MAILSHARE/lists/ -directory, or somewhere else ? If they > are elsewere, where ? How ? Sorry, that was quite an oversite on my part. Obviously you'd need to know this info.... Right now, I'm just using the ZMailer to manage my mailing lists. As you mentioned above, my lists are in $MAILSHARE/lists/. However, ZMailer isn't suppressing duplicates on these internal lists.... Sometime in the future, I'll be moving to majordomo for list management; however. Thanks, Kevin Traas Systems Analyst Edmondson Roper CA http://www.eroper.bc.ca From zmailer-owner@nic.funet.fi Fri May 30 21:03:52 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 54423 [194.67.3.135]) by nic.funet.fi with ESMTP id <2206-27824>; Fri, 30 May 1997 21:02:27 +0300 Received: from NO-IDENT-SERVICE@ariel (port 48903 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68047-11036>; Fri, 30 May 1997 22:02:14 +0400 Date: Fri, 30 May 1997 22:01:59 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: mailbox -l gone? To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970530173818Z2563-22436+201@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > > is it intentional that -l option does not work in .49? > > There is only one place where the log-event is done, > and it means the flow must go thru it every time... > Thus perhaps following will do it ? > --- transports/mailbox/mailbox.c~ Mon May 26 20:41:31 1997 > +++ transports/mailbox/mailbox.c Fri May 30 20:35:08 1997 No, mailbox command does not *accept* -l option, issueing `usage' message. [BTW, the set of options in the `usage' message does not match the set of options actually accepted by getopt() call] Eugene From zmailer-owner@nic.funet.fi Fri May 30 21:30:19 1997 Received: from NO-IDENT-SERVICE@antares.utu.fi (port 43169 [130.232.1.2]) by nic.funet.fi with ESMTP id <15443-4234>; Fri, 30 May 1997 21:28:42 +0300 Received: by utu.fi id <37901-6208>; Fri, 30 May 1997 21:27:21 +0300 Subject: Re: Error: norealname ?!?!? In-Reply-To: from Enrique Vadillo at "May 30, 97 12:38:59 pm" From: Matti Aarnio To: vadillo@quena.rcp.net.pe (Enrique Vadillo) Date: Fri, 30 May 1997 21:27:15 +0300 (EET DST) Cc: zmailer@nic.funet.fi, crosser@average.org, mea@mea.tmt.tele.fi, tom@sdf.com, ken@spry.com, mss1@cornell.edu X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 713 Message-Id: <19970530182721Z37901-6208+1787@utu.fi> Return-Path: > Hello, > > I am getting this error message after successfully installing Zmailer-2.99.48: You will need to apply patches 1 thru 3 to the source tree, then do the whole chain again from configuring to ... 'make clean', 'make', and 'make install'. The important bits (for this) are the small changes in the proto/cf/*.cf -files that are in the patch1, but the others are also needed. Now I am too tired to remember, if the 'make install' will automatically install the $MAILSHARE/cf/ -files, or not. ... > Enrique Vadillo- > -- > RCP - Internet Peru > Fax: +51 1 241-1320 > Web Site: http://www.rcp.net.pe (PERU) > Mirror Web Site: http://ekeko.rcp.net.pe (USA) /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 30 21:33:16 1997 Received: by nic.funet.fi id <15796-27824>; Fri, 30 May 1997 21:31:25 +0300 Subject: Re: mailbox -l gone? From: Matti Aarnio To: crosser@online.ru Date: Fri, 30 May 1997 21:31:15 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at May 30, 97 10:01:59 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 838 Message-Id: <19970530183125Z15796-27824+242@nic.funet.fi> Return-Path: > > > is it intentional that -l option does not work in .49? > > > > There is only one place where the log-event is done, > > and it means the flow must go thru it every time... > > Thus perhaps following will do it ? > > > --- transports/mailbox/mailbox.c~ Mon May 26 20:41:31 1997 > > +++ transports/mailbox/mailbox.c Fri May 30 20:35:08 1997 > > No, mailbox command does not *accept* -l option, issueing `usage' message. > [BTW, the set of options in the `usage' message does not match the set of > options actually accepted by getopt() call] Doh? (as is the popular word in US TV..) I just tried following two mailbox calls at my workstation: mailbox -8 -l /tmp/mailbox.log mailbox -l /tmp/mailbox.log -8 Umm.. Did you, by change try: mailbox -l/tmp/mailbox.log ??? > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Fri May 30 22:07:49 1997 Received: from NO-IDENT-SERVICE@press-gopher.uchicago.edu (port 64999 [128.135.234.36]) by nic.funet.fi with ESMTP id <15384-22436>; Fri, 30 May 1997 22:06:15 +0300 Received: from rcb@localhost (fake: uid#100@localhost) by press-gopher.uchicago.edu with SMTP id <26532-17078>; Fri, 30 May 1997 14:06:04 -0500 Date: Fri, 30 May 1997 14:05:53 -0500 (CDT) Sender: Roy C Bixler From: Roy C Bixler To: zmailer@nic.funet.fi cc: Enrique Vadillo Subject: Re: Error: norealname ?!?!? In-Reply-To: <19970530182721Z37901-6208+1787@utu.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Fri, 30 May 1997, Matti Aarnio wrote: > Now I am too tired to remember, if the 'make install' will > automatically install the $MAILSHARE/cf/ -files, or not. If I want this, I find that I have to do a 'rm -rf $MAILSHARE/cf' command before 'make install'. Roy rcb@press-gopher.uchicago.edu From zmailer-owner@nic.funet.fi Sat May 31 10:45:54 1997 Received: by nic.funet.fi id <15621-8636>; Sat, 31 May 1997 10:44:49 +0300 Subject: Re: Config Question From: Matti Aarnio To: KTraas@uniserve.com (Kevin Traas) Date: Sat, 31 May 1997 10:44:41 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199705301745.KAA29664@mercury.uniserve.com> from "Kevin Traas" at May 30, 97 10:38:44 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1518 Message-Id: <19970531074449Z15621-8636+97@nic.funet.fi> Return-Path: .... > Sorry, that was quite an oversite on my part. Obviously you'd need to know > this info.... > > Right now, I'm just using the ZMailer to manage my mailing lists. As you > mentioned above, my lists are in $MAILSHARE/lists/. However, ZMailer isn't > suppressing duplicates on these internal lists.... > > Sometime in the future, I'll be moving to majordomo for list management; > however. Ok, I had a brief tour over the configuration scripts. The info is buried in there into the comments. proto/cf/process.cf: Need to add 'db flush expansions' to other 'db flush'es it does at the start of process(). proto/cf/standard.cf: Comment away the 'db flush expansions'. See the associated comment text :-) It explains this all. After the editing, you need to explicitely install the script files: # cp -p proto/cf/*.cf $MAILSHARE/cf/ (because 'make install' will not do it..) (You can, of course, edit files in the $MAILSHARE/cf/ ...) Once you begin to use external programs to filter the lists, message's address associations are broken, and sending to two lists at same command line becomes all the sudden equivalent of sending the message individually to each of the lists; except of the RFC-822 headers, which don't have anything to do with message routing, anyway... Thus the message-wide address duplication filtering as built above breaks down. > Thanks, > > Kevin Traas > Systems Analyst > Edmondson Roper CA > http://www.eroper.bc.ca /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jun 2 18:36:44 1997 Received: from root@lupinella.troll.no (port 29447 [195.0.254.19]) by nic.funet.fi with ESMTP id <2667-27382>; Mon, 2 Jun 1997 18:35:27 +0300 Received: by troll.no id <45640-21325>; Mon, 2 Jun 1997 17:35:14 +0200 Sender: agulbra@troll.no To: zmailer@nic.funet.fi Subject: lists using zmailer Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 02 Jun 1997 17:35:12 +0200 Message-ID: Lines: 23 Return-Path: I heard there's been some discussion about running lists with zmailer here in the past few days. I have a setup I like quite well now, involving two perl scripts written specifically for zmailer. They assume that there's a capable human on-site who looks after the lists, and concentrate on doing the -simple- stuff and leave a little to the admin, rather than add complexity across the board to try to do as much as possible (as Majordomo does) or add DWIM features (as smartlist does). They also have reasonable antispam and anti-looping features, and use DSN to limit the amount of bounces. I've been meaning to send the scripts to Matti as contribs, but would like to have a little broader testing first. If you'd like to be the second site to run these scripts, before they're publicly available, write me. (I'm not currently subscribed to the zmailer list. I just sent off a subscription request so I should catch all replies, but please cc me directly just in case.) --Arnt From zmailer-owner@nic.funet.fi Mon Jun 2 19:12:37 1997 Received: from TIMEDOUT@wit393101.student.utwente.nl (port 2823 [130.89.235.11]) by nic.funet.fi with ESMTP id <2653-27382>; Mon, 2 Jun 1997 19:11:28 +0300 Received: from sim@localhost (port 40198 [127.0.0.1]) by wit393101.student.utwente.nl with SMTP id <32388-16943>; Mon, 2 Jun 1997 18:11:44 +0000 Date: Mon, 2 Jun 1997 18:11:30 +0200 (MET DST) From: Florian Overkamp Reply-To: sim@wit393101.student.utwente.nl To: Arnt Gulbrandsen cc: zmailer@nic.funet.fi Subject: Re: lists using zmailer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On 2 Jun 1997, Arnt Gulbrandsen wrote: > I've been meaning to send the scripts to Matti as contribs, but would > like to have a little broader testing first. If you'd like to be the > second site to run these scripts, before they're publicly available, > write me. I'm running some small lists on my machine with the zmailer lists and a linux/zmailer hacked ListServ for subbing and unsubbing. I'd like to atleast have a look at the scripts, but from your description I stil don't get any idea as to what they do and don't do. /* Florian Overkamp */ /* PGP Key-ID 233FC67D */ /* PGP Fingerprint 00 3F F5 63 DD FB 5D B7 0A 16 0A EC F9 BD 01 97 */ From zmailer-owner@nic.funet.fi Mon Jun 2 19:21:58 1997 Received: from root@lupinella.troll.no (port 3844 [195.0.254.19]) by nic.funet.fi with ESMTP id <2244-30324>; Mon, 2 Jun 1997 19:20:51 +0300 Received: by troll.no id <45646-21324>; Mon, 2 Jun 1997 18:20:37 +0200 Sender: agulbra@troll.no To: sim@wit393101.student.utwente.nl Cc: zmailer@nic.funet.fi Subject: Re: lists using zmailer References: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 02 Jun 1997 18:20:35 +0200 In-Reply-To: Florian Overkamp's message of Mon, 2 Jun 1997 18:11:30 +0200 (MET DST) Message-ID: Lines: 18 Return-Path: Florian Overkamp > I'm running some small lists on my machine with the zmailer lists and a > linux/zmailer hacked ListServ for subbing and unsubbing. I'd like to > atleast have a look at the scripts, but from your description I stil don't > get any idea as to what they do and don't do. I'll send them to you. One script handles -request mail, processing and answering some (with cc to the admins), answering some without doing any processing, and leaving a final small fraction for the admins to handle. The other one looks for mail loops, bounces, unsubscribes, spam and perhaps some other bogus mail, and resends the rest to the lists' subscribers with an appropriate DSN. It does not use any alias, but sends the message to all of the subscribers itself. --Arnt From zmailer-owner@nic.funet.fi Tue Jun 3 00:17:41 1997 Received: from TIMEDOUT@quena.rcp.net.pe (port 11453 [161.132.12.3]) by nic.funet.fi with SMTP id <2506-17469>; Tue, 3 Jun 1997 00:16:10 +0300 Received: by quena.rcp.net.pe via sendmail with stdio id for zmailer@nic.funet.fi; Mon, 2 Jun 97 16:13:21 -0400 (EDT) (/\##/\ Smail3.1.30.13 #30.5 built 30-sep-96) Message-Id: From: vadillo@quena.rcp.net.pe (Enrique Vadillo) Subject: read error on SMTP process To: zmailer@nic.funet.fi Date: Mon, 2 Jun 1997 16:13:19 -0400 (EDT) Cc: tom@sdf.com PGP-FingerPrint: 55 B9 83 D2 61 71 E6 6B 1E CE FD B5 F7 AA F1 B5 X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2542 Return-Path: Hi, I have flawlessly compiled and installed zmailer-2.99.48p3 under Solaris 2.5.1 , but i still have one problem, when i send mail to one local user, i get this message from the sender's host (which uses Smail 3.2): 06/02/97 14:58:49: [m0wYf7j-000oBLC] Deferred TO:operador@relay2.rcp.net.pe ROUTER:inet_hosts TRANSPORT:smtp ERROR:(ERR151) transport smtp: 499 read error from remote SMTP process relay2.rcp.net.pe is the host that now uses zmailer-2.99.48p3. I get this using 'sendmail -bt' for local user 'operador': [relay2/usr/spool/zmail]# sendmail -bt ZMailer router (2.99.48 #1: Thu May 29 20:48:26 CST 1997) vadillo@ayllu:/usr/zmail/zmailer-2.99.48/router Copyright 1992 Rayan S. Zachariassen Copyright 1992-1997 Matti Aarnio z# router operador : address: operador (((local operador operador@relay2.rcp.net.pe default_attributes))) z# I have the following in my system: drwxrwxrwt 3 root mail 512 Jun 2 15:22 /var/mail/ and under the $POSTOFFICE directory: drwxr-xr-- 2 root other 512 May 29 18:55 deferred/ drwxrwxrwx 2 root other 512 Jun 2 12:51 log/ drwxr-x--- 2 root other 512 May 29 18:55 postman/ drwxrwxrwt 2 root other 512 Jun 2 15:49 public/ drwxr-x--- 2 root other 512 May 29 18:55 queue/ drwxr-xr-x 2 root other 512 May 29 18:55 router/ drwxr-xr-x 2 root other 512 May 29 18:55 scheduler/ drwxr-xr-x 2 root other 512 May 29 18:55 transport/ The undelivered message can be found in the 'public' directory, what can be wrong with the reading processes? any ideas? the 'smtpserver' log showed this for this message: ----------------------------------------------------------------------- 25719# connection from TIMEDOUT@tucume.rcp.net.pe (port 41197) 25719w 220 relay2.rcp.net.pe ZMailer Server 2.99.48 #1 ESMTP+IDENT ready at Mon , 2 Jun 1997 15:09:41 -0400 25719# remote from [161.132.12.7] 25719r HELO tucume.rcp.net.pe 25719w 250 relay2.rcp.net.pe Hello tucume.rcp.net.pe 25719r MAIL FROM: 25719w 250 2.1.0 Sender syntax Ok 25719r RCPT TO: 25719w 250 2.1.5 Recipient address syntax Ok 25719r DATA 25719w 354 Start mail input; end with . ----------------------------------------------------------------------- Enrique Vadillo- -- RCP - Internet Peru Fax: +51 1 241-1320 Web Site: http://www.rcp.net.pe (PERU) Mirror Web Site: http://ekeko.rcp.net.pe (USA) From zmailer-owner@nic.funet.fi Tue Jun 3 10:34:22 1997 Received: by nic.funet.fi id <2122-17469>; Tue, 3 Jun 1997 10:33:15 +0300 Subject: Re: read error on SMTP process From: Matti Aarnio To: vadillo@quena.rcp.net.pe (Enrique Vadillo) Date: Tue, 3 Jun 1997 10:33:11 +0300 (EET DST) Cc: zmailer@nic.funet.fi, tom@sdf.com In-Reply-To: from "Enrique Vadillo" at Jun 2, 97 04:13:19 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1530 Message-Id: <19970603073315Z2122-17469+109@nic.funet.fi> Return-Path: > Hi, > > I have flawlessly compiled and installed zmailer-2.99.48p3 under Solaris > 2.5.1 , but i still have one problem, when i send mail to one local user, > i get this message from the sender's host (which uses Smail 3.2): .... > I have the following in my system: > > drwxrwxrwt 3 root mail 512 Jun 2 15:22 /var/mail/ > > and under the $POSTOFFICE directory: > > drwxr-xr-- 2 root other 512 May 29 18:55 deferred/ > drwxrwxrwx 2 root other 512 Jun 2 12:51 log/ > drwxr-x--- 2 root other 512 May 29 18:55 postman/ > drwxrwxrwt 2 root other 512 Jun 2 15:49 public/ > drwxr-x--- 2 root other 512 May 29 18:55 queue/ > drwxr-xr-x 2 root other 512 May 29 18:55 router/ The "router" directory protection needs to be 1777 (like also with "public"). And as you apparently put logs to that same directory-tree, I suggest you add the T-bit to your 777 at "log" directory too, because plain 777 is way too open... What protections do you have at the $MAILSHARE directory there ? (I really should write a program to do automatic protection verifications, apparently it really has a call..) > drwxr-xr-x 2 root other 512 May 29 18:55 scheduler/ > drwxr-xr-x 2 root other 512 May 29 18:55 transport/ ... > Enrique Vadillo- > -- > RCP - Internet Peru > Fax: +51 1 241-1320 > Web Site: http://www.rcp.net.pe (PERU) > Mirror Web Site: http://ekeko.rcp.net.pe (USA) /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jun 3 20:14:43 1997 Received: from root@m45bmi.Dave-World.net (port 58672 [206.26.2.45]) by nic.funet.fi with ESMTP id <2977-17469>; Tue, 3 Jun 1997 20:13:01 +0300 Received: from bozo.comanche.dyn.ml.org (bozo.comanche.dyn.ml.org [192.168.1.2]) by linux.comanche.dyn.ml.org (8.8.4/8.8.4) with ESMTP id MAA05656 for ; Tue, 3 Jun 1997 12:12:08 -0500 Message-ID: <33945042.C3FA8F2C@prairienet.org> Date: Tue, 03 Jun 1997 12:11:30 -0500 From: Seymore Buttz X-Mailer: Mozilla 4.0b5 [en] (Win95; I) MIME-Version: 1.0 To: zmailer@nic.funet.fi Subject: help/unsubscribe? X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: I've sent some help & unsubscribe msgs to zmailer-request but haven't heard back. Is that the correct address for such requests, or is there any other problem? Lemme know, thanks. From zmailer-owner@nic.funet.fi Wed Jun 4 13:18:42 1997 Received: from root@tuminfo2.informatik.tu-muenchen.de (port 10679 [131.159.0.81]) by nic.funet.fi with ESMTP id <15403-30801>; Wed, 4 Jun 1997 13:14:54 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109545-21460>; Wed, 4 Jun 1997 12:14:27 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <18961-10746>; Wed, 4 Jun 1997 12:14:06 +0200 Date: Wed, 4 Jun 1997 12:14:03 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt X-Sender: goldt@sunrbg6 To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Alive ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi Matti, i am still waiting for my signed pgp key. Meanwhile i have a new bug for you. Somebody here wants to send a mail to someone.@t-online.de ^^^ Now zmailer/router says this is not RFC822/9??, but that's not true. This adress exists (not someone, but with a trailing '.') and other mailsystems deliver mails to this adress. Greets, Sven -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey From zmailer-owner@nic.funet.fi Wed Jun 4 18:19:39 1997 Received: from TIMEDOUT@tshimshan.phys.ualberta.ca (port 26643 [129.128.7.76]) by nic.funet.fi with ESMTP id <15588-25544>; Wed, 4 Jun 1997 18:16:54 +0300 Received: by tshimshan.Phys.UAlberta.Ca id <8047-2>; Wed, 4 Jun 1997 09:13:57 -0600 Date: Wed, 4 Jun 1997 09:13:43 -0600 (MDT) Sender: James MacKinnon From: James MacKinnon X-Sender: jmack@tshimshan To: zmailer@nic.funet.fi Subject: Re: Alive ? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 4 Jun 1997, Sven Goldt wrote: > Hi Matti, > > i am still waiting for my signed pgp key. Meanwhile i have a new bug > for you. Somebody here wants to send a mail to someone.@t-online.de > ^^^ > Now zmailer/router says this is not RFC822/9??, but that's not true. > This adress exists (not someone, but with a trailing '.') and other > mailsystems deliver mails to this adress. A couple of other similar ones I've seen recently: from and from both likely not RFC822/9xx, but "NETWiRE!" does seem to be what 'Pegasus Mail for Win32 (v2.53/R1)' is using at that site... Seems these stupid stupid MESSDOS box mail setups lack a lot of sanity checking. Cheers, -- James S. MacKinnon Office: P-139 Avahd-Bhatia Physics Lab Team Physics Voice : (403) 492-8226 University of Alberta email : Jim.MacKinnon@Phys.UAlberta.CA Edmonton, Canada T6G 2N5 WWW: http://www.phys.ualberta.ca/~jmack/jmack.html From zmailer-owner@nic.funet.fi Wed Jun 4 18:26:26 1997 Received: by nic.funet.fi id <16073-25544>; Wed, 4 Jun 1997 18:24:20 +0300 Subject: Re: What about the book? From: Matti Aarnio To: vadillo@quena.rcp.net.pe (Enrique Vadillo) Date: Wed, 4 Jun 1997 18:24:10 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Enrique Vadillo" at Jun 3, 97 01:01:07 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1841 Message-Id: <19970604152420Z16073-25544+215@nic.funet.fi> Return-Path: > Gentle Matti, > > As i had already told you, i have succeeded configuring the Zmailer, > now what i am interested about is *better* documentation in order to > install it in our busy mail gateways. At the moment, no. I recall you asked how fast the processing is with ZMailer ? You noted that usually the processing delay was about 60 seconds. That is true for lightly loaded system, but it is caused by various systems checking for new jobs only every 20-30 seconds. If they detect new jobs, they are FAST at assimilating, and processing every new job there is. (Remember: Latency != low throughput) Once the job-list from previous scan has been exhausted, the spool area is checked again for possible new jobs, and only if no new jobs are there, the router will sleep for 20-30 seconds. Then at the scheduler things are different, and the work assimi- lation is a bit slower with mandatory delays in between directory scans -- but no delays in handling working transporters. All in all, this way you can handle way over 1 million messages per day without ever getting your system load average very high. > Do you have better documentation or (better) some book coming up? We are looking into it. Initially there won't be any printed books, however text for one may appear presuming we (= my employer) agree on hiring technical writer to help me. We intend to use Linux-DOC SGML to create it with idea in mind, that such base material is easier to turn into a book. (And as you may know, Linux-DOC SGML is used to create Postscript, ASCII, as well as HTML representations of the same material.) > Thanks. > > Enrique Vadillo- > -- > RCP - Internet Peru > Fax: +51 1 241-1320 > Web Site: http://www.rcp.net.pe (PERU) > Mirror Web Site: http://ekeko.rcp.net.pe (USA) /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Jun 4 18:29:41 1997 Received: from NO-IDENT-SERVICE@mariner.interdynamix.com (port 45749 [206.75.211.66]) by nic.funet.fi with ESMTP id <15363-17617>; Wed, 4 Jun 1997 18:28:44 +0300 Received: from pongo.interdynamix.com ([192.168.1.3]) by mariner.interdynamix.com with SMTP id <495568-224>; Wed, 4 Jun 1997 09:32:45 -0600 Message-ID: <339589B4.3E8C@interdynamix.com> Date: Wed, 04 Jun 1997 09:28:52 -0600 From: Dean Michaels Organization: InterDynamiX Systems Corp X-Mailer: Mozilla 3.01 (Win95; I) MIME-Version: 1.0 To: James MacKinnon CC: zmailer@nic.funet.fi Subject: Re: Alive ? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: > from > > and from I'm not sure if these really need fixing, First example should be quoted <"Lola & Connie"@vol.it> Second example screams uucp/smtp/whatever. Dean. From zmailer-owner@nic.funet.fi Wed Jun 4 21:19:56 1997 Received: from NO-IDENT-SERVICE@[206.161.253.66] (port 46693 [206.161.253.66]) by nic.funet.fi with ESMTP id <2257-25544>; Wed, 4 Jun 1997 21:18:44 +0300 Received: from NO-IDENT-SERVICE@localhost (port 34395 [127.0.0.1]) by roadrunner.realbig.com with SMTP id <133181-24329>; Wed, 4 Jun 1997 14:14:49 +0000 Date: Wed, 4 Jun 1997 14:14:49 -0400 (EDT) From: Andy Poling To: Dean Michaels cc: James MacKinnon , zmailer@nic.funet.fi Subject: Re: Alive ? In-Reply-To: <339589B4.3E8C@interdynamix.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 4 Jun 1997, Dean Michaels wrote: > > from > > > > and from > > I'm not sure if these really need fixing, > > First example should be quoted <"Lola & Connie"@vol.it> > > Second example screams uucp/smtp/whatever. Yes - never (ever) assume that because some LAN (the PC is silent) based mail system allows a mail address it is legal. They're collectively (in)famous for their protocol and RFC violations. You can research what's legal by reading the RFC's... -Andy Global Auctions From zmailer-owner@nic.funet.fi Thu Jun 5 18:54:25 1997 Received: from NO-IDENT-SERVICE@outbound.Princeton.EDU (port 38357 [128.112.128.84]) by nic.funet.fi with ESMTP id <3020-20444>; Thu, 5 Jun 1997 18:50:13 +0300 Received: from IDENT-NOT-QUERIED@lists.Princeton.EDU (port 1371 [128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <542001-15981>; Thu, 5 Jun 1997 11:49:54 -0400 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370112.s2-1>; Thu, 5 Jun 1997 11:49:53 -0400 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id LAA25577 for ; Thu, 5 Jun 1997 11:49:47 -0400 Message-Id: <199706051549.LAA25577@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: hold channel is a memory hog Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 05 Jun 1997 11:49:46 EDT From: "Michael R. Gettes" Return-Path: running zmailer zmailer-2.99.48 (pl3) I am once again having a problem with the hold channel utilizing 117MB of swap and constantly running. I hope this is sufficient info as I have to kill this process to get mail going again since I am out of memory. mailq shows: hold/*/0 hold/ns:adlibv.adelphi.edu/any/0 R=1 A=1006242 P=16950 HA=0s FA=0s OF=1 QA=2d2h hold/ns:icdgate.hanscom.af.mil/any/0 R=1 A=2 W=662s QA=2h8m32s hold/ns:sable.adelphi.edu/any/0 R=1 A=967772 QA=2d2h hold/ns:epfl2.epflbalto.org/any/0 R=6 A=6 W=23810s QA=11h57m59s channelwait Threads: 4 Msgs: 9 Procs: 1 Idle: 0 Plim: 20 Flim: 150 A truss shows the following (this is sunos5.5.1): write(1, " # h u n g r y\n", 8) = 8 read(0, " S / Q / 5 4 1 9 6 0 - 1".., 8192) = 43 getpid() = 16950 [29418] open("S/Q/541960-16959", O_RDWR) = 3 fstat(3, 0xEFFFF218) = 0 fcntl(3, F_SETFD, 0x00000001) = 0 mmap(0x00000000, 13823, PROT_READ|PROT_WRITE, MAP_SHARED, 3, 0) = 0xEF750000 read(3, " i 5 4 1 9 6 0 - 1 6 9".., 13823) = 13823 open("../queue/S/Q/541960-16959", O_RDONLY) = 4 fstat(4, 0xEFFFF218) = 0 fcntl(4, F_SETFD, 0x00000001) = 0 mmap(0x00000000, 11491, PROT_READ, MAP_SHARED, 4, 0) = 0xEF5B0000 getpid() = 16950 [29418] write(1, " 5 4 1 9 6 0 / 7 8 4 6\t".., 139) = 139 time() = 865525645 fstat(5, 0xEFFFAC88) = 0 time() = 865525645 getpid() = 16950 [29418] putmsg(5, 0xEFFFA340, 0xEFFFA334, 0) = 0 fstat(6, 0xEFFFA1F0) = 0 fcntl(6, F_SETLK, 0xEFFFA29C) Err#11 EAGAIN munmap(0xEF5B0000, 11491) = 0 munmap(0xEF750000, 13823) = 0 close(3) = 0 close(4) = 0 write(1, " # h u n g r y\n", 8) = 8 read(0, " S / Q / 5 4 1 9 6 0 - 1".., 8192) = 42 From zmailer-owner@nic.funet.fi Thu Jun 5 19:27:44 1997 Received: from root@acli.interlog.com (port 5908 [198.53.146.23]) by nic.funet.fi with ESMTP id <2587-30582>; Thu, 5 Jun 1997 19:26:31 +0300 Received: from acli@localhost (port 35091 [127.0.0.1]) by acli.interlog.com with SMTP id <16193-12151>; Thu, 5 Jun 1997 12:26:22 +0000 Date: Thu, 5 Jun 1997 12:26:03 -0400 (EDT) From: Ambrose Li Sender: acli@acli.interlog.com To: zmailer@nic.funet.fi Subject: spam refusal with crossbar.cf? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I was trying to do spam blocking (blocking known spam sites) from within the crossbar, but after what I have done zmailer refuses to deliver mail to the blocked sites, but still happily deliver mail that originates from those sites. I am clueless about what I did wrong. Someone must have done this before? -- Ambrose Li ~ acli@acli%interlog.com (Hello, world!\n) ~ http://www.interlog.com/~acli/~acli/ ~ print flush quit From zmailer-owner@nic.funet.fi Thu Jun 5 20:41:36 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 18703 [13.1.64.93]) by nic.funet.fi with SMTP id <3006-27508>; Thu, 5 Jun 1997 20:41:18 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <18482(1)>; Thu, 5 Jun 1997 10:40:03 PDT X-NS-Transport-ID: 0800200AF21F374DC9BD Date: Thu, 5 Jun 1997 10:39:24 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: spam refusal with crossbar.cf? In-Reply-to: "acli@acli.interlog:com:Xerox's message of Thu, 5 Jun 1997 09:26:03 PDT" To: acli@acli.interlog.COM cc: zmailer@nic.funet.FI Message-Id: <97Jun5.104003pdt."18482(1)"@alpha.xerox.com> Return-Path: I run my smtpserver with "-s vef", and added code in the "server" function in the router config to check the "from" fields -- I maintain a table of domains that I don't accept mail from, and user@domain names that I won't accept mail from. There are some usernames I won't accept mail from no matter what domain they are in (e.g., "friend@....") \nick From zmailer-owner@nic.funet.fi Thu Jun 5 20:45:19 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 57993 [194.67.3.135]) by nic.funet.fi with ESMTP id <2649-3235>; Thu, 5 Jun 1997 20:44:55 +0300 Received: from NO-IDENT-SERVICE@ariel (port 61098 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68064-24817>; Thu, 5 Jun 1997 21:44:21 +0400 Date: Thu, 5 Jun 1997 21:44:18 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: spam refusal with crossbar.cf? To: Ambrose Li cc: zmailer@nic.funet.fi In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > I was trying to do spam blocking (blocking known spam sites) > from within the crossbar, but after what I have done zmailer > refuses to deliver mail to the blocked sites, but still > happily deliver mail that originates from those sites. > > I am clueless about what I did wrong. Someone must have done > this before? In the latest beta, Matti has included some anti-spamming code in the smtpserver. You may want to check this. Eugene From zmailer-owner@nic.funet.fi Thu Jun 5 21:39:41 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 31492 [204.244.210.193]) by nic.funet.fi with ESMTP id <3035-3235>; Thu, 5 Jun 1997 21:39:28 +0300 Received: from localhost (tom@localhost) by misery.sdf.com (8.8.5/8.8.5) with SMTP id LAA08024; Thu, 5 Jun 1997 11:37:18 -0700 (PDT) X-Authentication-Warning: misery.sdf.com: tom owned process doing -bs Date: Thu, 5 Jun 1997 11:37:17 -0700 (PDT) From: Tom Samplonius To: Nicholas_Briggs.PARC@xerox.com cc: acli@acli.interlog.COM, zmailer@nic.funet.FI Subject: Re: spam refusal with crossbar.cf? In-Reply-To: <97Jun5.104003pdt."18482(1)"@alpha.xerox.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 5 Jun 1997 Nicholas_Briggs.PARC@xerox.com wrote: > I run my smtpserver with "-s vef", and added code in the "server" function in > the router config to check the "from" fields -- I maintain a table of domains > that I don't accept mail from, and user@domain names that I won't accept mail > from. There are some usernames I won't accept mail from no matter what domain > they are in (e.g., "friend@....") > > \nick There would seem to be quite a performance hit for this luxury. It means that you will be forking and execing a router for every incoming message. I guess your server is very powerful. Tom From zmailer-owner@nic.funet.fi Thu Jun 5 22:02:39 1997 Received: by nic.funet.fi id <2329-19826>; Thu, 5 Jun 1997 22:02:05 +0300 Subject: Re: spam refusal with crossbar.cf? From: Matti Aarnio To: crosser@online.ru Date: Thu, 5 Jun 1997 22:01:59 +0300 (EET DST) Cc: acli@acli.interlog.com, zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Jun 5, 97 09:44:18 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 817 Message-Id: <19970605190205Z2329-19826+199@nic.funet.fi> Return-Path: > > I was trying to do spam blocking (blocking known spam sites) > > from within the crossbar, but after what I have done zmailer > > refuses to deliver mail to the blocked sites, but still > > happily deliver mail that originates from those sites. > > > > I am clueless about what I did wrong. Someone must have done > > this before? > > In the latest beta, Matti has included some anti-spamming code in > the smtpserver. You may want to check this. Indeed, but not yet quite enough. My summer assistant is coding more bits for it every day. I just made a snapshot 2.99.49b10 available at my workstation, but it has none of that code yet. It does, however have working getaddrinfo(); at least it works for me at nic.funet.fi without several bogus "DNS inconsistency" claims. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 5 22:44:42 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 47373 [13.1.64.93]) by nic.funet.fi with SMTP id <2369-2000>; Thu, 5 Jun 1997 22:44:31 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <17141(2)>; Thu, 5 Jun 1997 12:44:18 PDT X-NS-Transport-ID: 0800200AF21F374DC9D0 Date: Thu, 5 Jun 1997 12:43:47 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: spam refusal with crossbar.cf? In-Reply-to: "tom@sdf:com:Xerox's message of Thu, 5 Jun 1997 11:37:17 PDT" To: tom@sdf.COM cc: Nicholas_Briggs.PARC@xerox.com, acli@acli.interlog.COM, zmailer@nic.funet.FI Message-Id: <97Jun5.124418pdt."17141(2)"@alpha.xerox.com> Return-Path: I process 80000 messages a day on a slow (85MHz) SPARC 5, running SunOS 4.1.3_U1 From zmailer-owner@nic.funet.fi Fri Jun 6 10:12:13 1997 Received: from NO-IDENT-SERVICE@zeronull.aszi.sztaki.hu (port 17676 [193.225.12.33]) by nic.funet.fi with SMTP id <2279-26504>; Fri, 6 Jun 1997 10:10:57 +0300 Received: by zeronull.aszi.sztaki.hu (5.67a8/SZTAKI-3.13) id AG29360; Fri, 6 Jun 1997 09:11:03 +0200 Date: Fri, 6 Jun 1997 09:11:03 +0200 From: Gabor Kiss Message-Id: <199706060711.AG29360@zeronull.aszi.sztaki.hu> To: zmailer@nic.funet.fi Subject: Re: spam refusal with crossbar.cf? X-Charset: US X-Char-Esc: 0 Return-Path: > There would seem to be quite a performance hit for this luxury. It > means that you will be forking and execing a router for every incoming > message. I guess your server is very powerful. I do it since this February. No performance problems. (Sparccenter 2000 ;-) See my policy-based-filering-database patch at ftp://ftp.sztaki.hu/pub/private/kissg/zmailer. Regards Gabor From zmailer-owner@nic.funet.fi Fri Jun 6 14:40:57 1997 Received: from root@tuminfo2.informatik.tu-muenchen.de (port 59877 [131.159.0.81]) by nic.funet.fi with ESMTP id <2465-26504>; Fri, 6 Jun 1997 14:40:32 +0300 Received: from sunrbg6.informatik.tu-muenchen.de ([131.159.1.33]) by tuminfo2.informatik.tu-muenchen.de with ESMTP id <109569-21464>; Fri, 6 Jun 1997 13:40:24 +0200 Received: by sunrbg6.informatik.tu-muenchen.de id <18946-10746>; Fri, 6 Jun 1997 13:40:08 +0200 Date: Fri, 6 Jun 1997 13:39:57 +0200 (MET DST) Sender: Sven Goldt From: Sven Goldt X-Sender: goldt@sunrbg6 To: zmailer@ftp.funet.fi Subject: smtpserver Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi Matti, while i am still waiting for you to send me my signed pgp key, i have yet another bug. We start the smtpserver from /etc/inetd.conf for some reason and if the mail spool is full, the smtpserver still accepts mail ! Zmailer returns incoming mail with 'unknown recipient' until we free some space. Greets, Sven -- PGP public key at http://www.informatik.tu-muenchen.de/~goldt/pgp/pubkey From zmailer-owner@nic.funet.fi Fri Jun 6 15:47:45 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 34715 [194.67.3.135]) by nic.funet.fi with ESMTP id <2872-3487>; Fri, 6 Jun 1997 15:45:36 +0300 Received: from NO-IDENT-SERVICE@ariel (port 62718 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68066-473>; Fri, 6 Jun 1997 16:45:20 +0400 Date: Fri, 6 Jun 1997 16:45:13 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: .49 beta10 To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970605190205Z2329-19826+199@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > I just made a snapshot 2.99.49b10 available at my workstation, > but it has none of that code yet. It does, however have working > getaddrinfo(); at least it works for me at nic.funet.fi without > several bogus "DNS inconsistency" claims. I've installed it right now, seems to work so far. Eugene From zmailer-owner@nic.funet.fi Fri Jun 6 20:15:16 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 5893 [204.244.210.193]) by nic.funet.fi with ESMTP id <2916-30582>; Fri, 6 Jun 1997 20:13:10 +0300 Received: from localhost (tom@localhost) by misery.sdf.com (8.8.5/8.8.5) with SMTP id QAA11082; Thu, 29 May 1997 16:34:13 -0700 (PDT) X-Authentication-Warning: misery.sdf.com: tom owned process doing -bs Date: Thu, 29 May 1997 16:34:12 -0700 (PDT) From: Tom Samplonius To: Donald Ball cc: Matti Aarnio , zmailer@nic.funet.FI Subject: Re: How to change delivery agent in zmailer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 29 May 1997, Donald Ball wrote: > So I finally got zmailer compiled and installed (although I did have to > tweak the Makefile in the man directory a teensy bit - changed "-m444" to > "-m 444"). Now I want to change the delivery agent so that it will > deliver to cyrus imapd instead of mailboxes in /var/mail. There apparantly > is a way, since there's a small mention of it in pm.conf, but very little > in the way of actual documentation. Anyone else out there have zmailer > configured for use with cyrus? What's the secret? > > - donald No secret at all. Basically make sure that the line in sm.conf for cyrus is enabled, and make sure that you modify the "local/*" section in scheduler.conf to start sm, rather than mailbox. I used to use Zmailer with cyrus for quite a while, but I had ditch Zmailer because it rather unstable at the time. Tom From zmailer-owner@nic.funet.fi Fri Jun 6 20:37:06 1997 Received: by nic.funet.fi id <2763-16702>; Fri, 6 Jun 1997 20:33:31 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: patch for 2.99.49beta10 Message-Id: <19970606173331Z2763-16702+130@nic.funet.fi> Date: Fri, 6 Jun 1997 20:33:28 +0300 Return-Path: ... at my workstation. I just found out that under some conditions the getaddrinfo() was rejecting valid responses. It did test ">=" in places where ">" is correct.. /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Jun 6 21:06:25 1997 Received: from NO-IDENT-SERVICE@calvin.math.vt.edu (port 4749 [128.173.42.128]) by nic.funet.fi with ESMTP id <2893-30224>; Fri, 6 Jun 1997 21:03:47 +0300 Received: from mmead@localhost by calvin.math.vt.edu id <57485-5862>; Fri, 6 Jun 1997 13:50:18 -0400 Subject: blocking mail from specific hosts From: Matthew Mead To: zmailer@nic.funet.fi Date: Fri, 6 Jun 1997 13:50:16 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 483 Message-Id: <19970606175018Z57485-5862+5@calvin.math.vt.edu> Return-Path: I'm pretty new to zmailer, it being the installed mailer where I am now working, and I'm trying to get up to speed on it. Is there a general FAQ out there somewhere that might help me? Right now, I'm specifically looking at trying to block the acceptance of mail from a specific set of hosts. Does anyone have any suggestions? Thanks! -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Fri Jun 6 21:41:20 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 1449 [194.67.3.135]) by nic.funet.fi with ESMTP id <2462-30224>; Fri, 6 Jun 1997 21:38:56 +0300 Received: from NO-IDENT-SERVICE@ariel (port 64020 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68066-5407>; Fri, 6 Jun 1997 22:38:23 +0400 Date: Fri, 6 Jun 1997 22:38:18 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: patch for 2.99.49beta10 To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970606173331Z2763-16702+130@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: I unstalled this one on our production machine. One thing noticed that I dislike: it writes in the syslog 227686._badheader saved for inspection: message header syntax error! and puts a message in POSTOFFICE/postman, not trying to fix the problem and send the message in some way (?). In my case, the problem was in non-critical Reply-To: header (which was broken by broken sendmail on the way). In another .badheader file, I cannot find any error and zmailer did not give any hint about what exactly it disliked. Old more tolerant behavior looks better... Eugene From zmailer-owner@nic.funet.fi Mon Jun 9 20:52:42 1997 Received: from NO-IDENT-SERVICE@outbound.Princeton.EDU (port 36267 [128.112.128.84]) by nic.funet.fi with ESMTP id <16299-23144>; Mon, 9 Jun 1997 20:50:40 +0300 Received: from IDENT-NOT-QUERIED@lists.Princeton.EDU (port 1783 [128.112.129.249]) by outbound.Princeton.EDU with ESMTP id <541955-9319>; Mon, 9 Jun 1997 13:50:02 -0400 Received: from yo.Princeton.EDU ([128.112.233.37]) by lists.Princeton.EDU with ESMTP id <370101.s1-2>; Mon, 9 Jun 1997 13:49:53 -0400 Received: from yo.Princeton.EDU (gettes@localhost) by yo.Princeton.EDU (8.6.12/8.6.12) with ESMTP id NAA09600 for ; Mon, 9 Jun 1997 13:49:48 -0400 Message-Id: <199706091749.NAA09600@yo.Princeton.EDU> X-Mailer: exmh version 1.6.9 8/22/96 To: zmailer@nic.funet.fi Subject: smtp talking to loopback Organization: Princeton University, Network Systems Group X-url: http://www.Princeton.EDU/~gettes Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 09 Jun 1997 13:49:48 EDT From: "Michael R. Gettes" Return-Path: I have found that in smtp.c there is code to prevent me from sending mail back to myself. Why is this considered bad?? What I want to do is send mail back to myself on a different port (where I have a sendmail running) to perform some special processing. Maybe there should be a check for a port other than the SMTP port, and if that is the case, then figure that the maintainer knows what they are doing and let it through. Does this seem reasonable??? /mrg From zmailer-owner@nic.funet.fi Mon Jun 9 22:55:53 1997 Received: from TIMEDOUT@fddisunsite.oit.unc.edu (port 43488 [152.2.254.81]) by nic.funet.fi with ESMTP id <15595-28263>; Mon, 9 Jun 1997 22:54:44 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by calzone.oit.unc.edu (8.8.5/8.8.5) with ESMTP id PAA22621; Mon, 9 Jun 1997 15:54:29 -0400 (EDT) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id TAA14713; Sun, 8 Jun 1997 19:05:19 -0400 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Sun, 8 Jun 1997 19:05:19 -0400 (EDT) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: Tom Samplonius cc: Matti Aarnio , zmailer@nic.funet.FI Subject: Re: How to change delivery agent in zmailer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: I can't get it to work at all. I've tested it with sendmail so I'm pretty sure it's not a cyrus configuration problem. I keep getting user not found errors in the router log file. Any clues whatsoever? - donald On Thu, 29 May 1997, Tom Samplonius wrote: > > On Thu, 29 May 1997, Donald Ball wrote: > > > So I finally got zmailer compiled and installed (although I did have to > > tweak the Makefile in the man directory a teensy bit - changed "-m444" to > > "-m 444"). Now I want to change the delivery agent so that it will > > deliver to cyrus imapd instead of mailboxes in /var/mail. There apparantly > > is a way, since there's a small mention of it in pm.conf, but very little > > in the way of actual documentation. Anyone else out there have zmailer > > configured for use with cyrus? What's the secret? > > > > - donald > > No secret at all. Basically make sure that the line in sm.conf for > cyrus is enabled, and make sure that you modify the "local/*" section in > scheduler.conf to start sm, rather than mailbox. > > I used to use Zmailer with cyrus for quite a while, but I had ditch > Zmailer because it rather unstable at the time. > > Tom > > From zmailer-owner@nic.funet.fi Tue Jun 10 09:02:32 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 25610 [204.244.210.193]) by nic.funet.fi with SMTP id <15773-25203>; Tue, 10 Jun 1997 09:01:21 +0300 Received: from tom by misery.sdf.com with smtp (Exim 1.62 #1) id 0wbFe3-0004rT-00; Mon, 9 Jun 1997 18:22:47 -0700 Date: Mon, 9 Jun 1997 18:22:47 -0700 (PDT) From: Tom Samplonius To: Donald Ball cc: Matti Aarnio , zmailer@nic.funet.FI Subject: Re: How to change delivery agent in zmailer In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Sun, 8 Jun 1997, Donald Ball wrote: > I can't get it to work at all. I've tested it with sendmail so I'm pretty > sure it's not a cyrus configuration problem. I keep getting user not found > errors in the router log file. Any clues whatsoever? > > - donald What does your "local/*" section in scheduler.conf say? Tom > On Thu, 29 May 1997, Tom Samplonius wrote: > > > > > On Thu, 29 May 1997, Donald Ball wrote: > > > > > So I finally got zmailer compiled and installed (although I did have to > > > tweak the Makefile in the man directory a teensy bit - changed "-m444" to > > > "-m 444"). Now I want to change the delivery agent so that it will > > > deliver to cyrus imapd instead of mailboxes in /var/mail. There apparantly > > > is a way, since there's a small mention of it in pm.conf, but very little > > > in the way of actual documentation. Anyone else out there have zmailer > > > configured for use with cyrus? What's the secret? > > > > > > - donald > > > > No secret at all. Basically make sure that the line in sm.conf for > > cyrus is enabled, and make sure that you modify the "local/*" section in > > scheduler.conf to start sm, rather than mailbox. > > > > I used to use Zmailer with cyrus for quite a while, but I had ditch > > Zmailer because it rather unstable at the time. > > > > Tom > > > > > > > From zmailer-owner@nic.funet.fi Tue Jun 10 12:35:16 1997 Received: by nic.funet.fi id <15387-10169>; Tue, 10 Jun 1997 12:33:38 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: ZMailer manual content Message-Id: <19970610093338Z15387-10169+129@nic.funet.fi> Date: Tue, 10 Jun 1997 12:33:36 +0300 Return-Path: Hello, I have been talking with a technical documentation company people about ZMailer manuals, and we came up with following rough structure. Do you have any comments/wishes ? If you look at "the Batbook" (sendmail book), you will find this overall structure hauntingly familiar... /Matti Aarnio Plans for ZMailer manual content: (with guesses on page counts; propably underestimated...) I A Tutorial ( 20 ) 1. Introduction to Email 2. ZMailer's features and facilities 3. Running Zmailer (short) II Build and Install ( 20 ) 1. Autoconfiguration 2. Compilation 3. Installation / Upgradeing 4. Configuring (short) 5. Verifying the system III Administration ( 50 ) 1. DNS and ZMailer 2. Security issues 3. The Queue 4. Logging and Statistics 5. SMTP-server configuration - Policy based filtering 6. Router configuration - Databases - Mailinglists and ~/.forward 7. Scheduler configuration (and transport agents?) 8. sm configuration IV Reference ( 100 ) 1. SMTP-server - Policy filtering language 2. 'sendmail', 'rmail' 3. zmailer(3) 4. Router - Internal functions ( ~ 100 functions ) - ZMSH script language - configuration script writing rules - debugging 5. Scheduler - Configuration "language" - Resource management - mailq 6. Delivery agents - mailbox - hold - smtp - sm - libta -- transport agent support library A.1 Sample router configuration scripts ( 20 ) A.2 Using ZMailer with Mailinglist Managers ( 20 ) A.3 Adding new transport agents ( 10 ) (mail delivery via procmail/cyrus/..) A.4 Internal file dataformats ( 30 ) A.nn SSL language ( 10 ) From zmailer-owner@nic.funet.fi Tue Jun 10 20:23:05 1997 Received: from NO-IDENT-SERVICE@pobox3.funet.fi (port 7458 [128.214.46.64]) by nic.funet.fi with ESMTP id <16057-10169>; Tue, 10 Jun 1997 20:21:45 +0300 Received: from chronos (chronos.sovam.com [194.67.3.135]) by pobox3.funet.fi (8.8.5/8.8.5/FUNET/SGI-3.12) with ESMTP id UAA17781 for ; Tue, 10 Jun 1997 20:21:12 +0300 (EET DST) Received: from NO-IDENT-SERVICE@ariel (port 38764 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68074-339>; Tue, 10 Jun 1997 21:19:58 +0400 Date: Tue, 10 Jun 1997 21:19:46 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: beta10 - too many end up in postman To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: Matti, I hate to say it but I had to rollback to .48 because of the number of messages that end up in POSTOFFICE/postman. There where 725 during the holidays; I cannot examine all of them. When I started .48 and moved them all to the router/, they *all* went thru one way or another (around 200 produced non-delivery reports). I kept a copy of postman contents in case you want more information about these messages. Eugene From zmailer-owner@nic.funet.fi Tue Jun 10 21:30:32 1997 Received: from NO-IDENT-SERVICE@amc-edm.amc.ualberta.ca (port 53263 [129.128.50.68]) by nic.funet.fi with ESMTP id <15564-10169>; Tue, 10 Jun 1997 21:29:12 +0300 Received: from IDENT-NOT-QUERIED@amc-pc34.amc.ab.ca (port 1466 []) by amc-edm.amc.ab.ca with SMTP id <17411-18576>; Tue, 10 Jun 1997 12:28:29 -0600 Date: Tue, 10 Jun 1997 12:28:44 -0600 () From: Daryle Tilroe To: zmailer@nic.funet.fi Subject: smtp seems to be ignoring a 550 return In-Reply-To: Message-ID: X-X-Sender: data@amc-edm.amc.ab.ca MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Using zmailer 2.99.48p2 under SunOS-4.1.4_U1. I don't know if I am missing something obvious but smtp seems to just idle out after receiving a 550 user unknown return from a remote esmtp (and I believe smtpd as well) daemon. The message just gets retried until the 3 day timeout is up. Any ideas? Daryle A. Tilroe -------------- Systems Administrator ------ data@amc.ab.ca Alberta Microelectronic Centre ----------------------- http://www.amc.ab.ca #318, 11315-87 Avenue |---------------------------------------------------| Edmonton, AB /\/\ | Work: (403) 492-3914 Direct: (403) 492-5570 x227 | Canada /_/\_ | Fax: (403) 492-1643 | T6G 2T9 /__\ |---------------------------------------------------| From zmailer-owner@nic.funet.fi Tue Jun 10 23:52:46 1997 Received: from NO-IDENT-SERVICE@crash.novalink.com (port 30217 [192.233.90.218]) by nic.funet.fi with ESMTP id <2154-10169>; Tue, 10 Jun 1997 23:51:37 +0300 Received: from dongeo ([192.233.90.81]) by crash.novalink.com (Post.Office MTA v3.1 release PO203a ID# 0-36389U100L2S100) with SMTP id AAA132 for ; Tue, 10 Jun 1997 17:48:25 -0400 Message-ID: <339DBF29.73BC@novalink.com> Date: Tue, 10 Jun 1997 16:55:05 -0400 From: jradoff@novalink.com (Jon Radoff) Reply-To: jradoff@novalink.com Organization: NovaLink X-Mailer: Mozilla 3.01Gold (WinNT; I) MIME-Version: 1.0 To: zmailer@nic.funet.fi Subject: unsubscribe Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: unsubscribe jradoff@novalink.com From zmailer-owner@nic.funet.fi Wed Jun 11 01:52:49 1997 Received: from root@scapa.cs.ualberta.ca (port 15623 [129.128.4.44]) by nic.funet.fi with ESMTP id <2631-908>; Wed, 11 Jun 1997 01:52:15 +0300 Received: from bruce@localhost (fake: redwater.cs.ualberta.ca) by scapa.cs.ualberta.ca id <13111-11275>; Tue, 10 Jun 1997 16:51:56 -0600 Subject: Re: smtp seems to be ignoring a 550 return From: Bruce Wm Folliott To: zmailer@nic.funet.fi Date: Tue, 10 Jun 1997 16:51:45 -0600 (MDT) In-Reply-To: from "Daryle Tilroe" at Jun 10, 97 12:28:44 pm Phone: (403) 492-4775 X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 961 Message-Id: <19970610225156Z13111-11275+1@scapa.cs.ualberta.ca> Return-Path: > > > Using zmailer 2.99.48p2 under SunOS-4.1.4_U1. > > I don't know if I am missing something obvious but smtp seems to just idle > out after receiving a 550 user unknown return from a remote esmtp (and I > believe smtpd as well) daemon. The message just gets retried until the 3 > day timeout is up. Any ideas? > > Here are a couple of additional observations I made with a similiar problem: Using `Mail -v`, this is the last part of the repeated smtp sessions: 550 ... User unknown ordered deletion of task file from scheduler memory (13113-10396) The far send sends a message which in part says: ----- Transcript of session follows ----- <<< RCPT To: ORCPT=rfc822;brucewilliamfolliott@gpu.srv.ualberta.ca 550 ... User unknown 421 maildrop.srv.ualberta.ca Lost input channel from scapa.cs.ualberta.ca [129.128.4.44] From zmailer-owner@nic.funet.fi Wed Jun 11 18:28:49 1997 Received: by nic.funet.fi id <16295-632>; Wed, 11 Jun 1997 18:27:57 +0300 Subject: Re: smtp seems to be ignoring a 550 return From: Matti Aarnio To: bruce@cs.ualberta.ca (Bruce Wm Folliott) Date: Wed, 11 Jun 1997 18:27:49 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <19970610225156Z13111-11275+1@scapa.cs.ualberta.ca> from "Bruce Wm Folliott" at Jun 10, 97 04:51:45 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 677 Message-Id: <19970611152757Z16295-632+34@nic.funet.fi> Return-Path: > > Using zmailer 2.99.48p2 under SunOS-4.1.4_U1. > > > > I don't know if I am missing something obvious but smtp seems to just idle > > out after receiving a 550 user unknown return from a remote esmtp (and I > > believe smtpd as well) daemon. The message just gets retried until the 3 > > day timeout is up. Any ideas? All the symptoms listed sound like the smtp process is dropping a core at the event. Is there a core at $POSTOFFICE/transport/ directory ? Can I get the backtrace ? (See that you start the scheduler so that it/its children can drop cores..) (2.99.48p3 does not drop the core, nor does it freeze at Linux..) /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Jun 11 18:50:23 1997 Received: from NO-IDENT-SERVICE@crash.novalink.com (port 58634 [192.233.90.218]) by nic.funet.fi with ESMTP id <2136-631>; Wed, 11 Jun 1997 18:48:37 +0300 Received: from dongeo ([192.233.90.81]) by crash.novalink.com (Post.Office MTA v3.1 release PO203a ID# 0-36389U100L2S100) with SMTP id AAA139; Wed, 11 Jun 1997 12:45:26 -0400 Message-ID: <339EC9A3.3919@novalink.com> Date: Wed, 11 Jun 1997 11:52:03 -0400 From: jradoff@novalink.com (Jon Radoff) Reply-To: jradoff@novalink.com Organization: NovaLink X-Mailer: Mozilla 3.01Gold (WinNT; I) MIME-Version: 1.0 To: Matti Aarnio CC: zmailer@nic.funet.fi Subject: Re: smtp seems to be ignoring a 550 return References: <19970611152757Z16295-632+34@nic.funet.fi> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Return-Path: Okay, I give up. How do I unsubscribe from this list? From zmailer-owner@nic.funet.fi Wed Jun 11 19:46:03 1997 Received: by nic.funet.fi id <2298-631>; Wed, 11 Jun 1997 19:44:34 +0300 Subject: Re: smtp talking to loopback From: Matti Aarnio To: gettes@Princeton.EDU (Michael R. Gettes) Date: Wed, 11 Jun 1997 19:44:25 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199706091749.NAA09600@yo.Princeton.EDU> from "Michael R. Gettes" at Jun 9, 97 01:49:48 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1510 Message-Id: <19970611164434Z2298-631+75@nic.funet.fi> Return-Path: > I have found that in smtp.c there is code to prevent me > from sending mail back to myself. Why is this considered > bad?? A last-line defence against MX pointer failure. It is troublesome with explicite routes / when you want to test weird things / ... > What I want to do is send mail back to myself on > a different port (where I have a sendmail running) to > perform some special processing. Maybe there should be > a check for a port other than the SMTP port, and if > that is the case, then figure that the maintainer knows > what they are doing and let it through. Does this > seem reasonable??? Yes. If the destination port != IPPORT_SMTP (25), then it is ok to talk to myself.. (unless the same ZMailer is listening at that weird port, and ...) (I am happy to declare that if sysadmins are doing that sort of tricks, they are all on their own, and ZMailer is better to let them just do it.) I have added this to my 2.99.48p3+ code base too. I will dump 2.99.49b11 (without 'aa') to my workstation anonymous ftp before I head for home today. It contains also partially coded policytest facilities for the smtpserver. It is essentially Gabor Kissigs algorithm, but done entirely in the smtpserver. (But it does not work yet! See comments at the source.) Oh yes, doc/design/zmog.ps is more up to date, than the old info file, but still not completely up to date.. We will be using that text in the new documentation too. > /mrg /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Jun 11 20:11:07 1997 Received: from root@lupinella.troll.no (port 36416 [195.0.254.19]) by nic.funet.fi with ESMTP id <2625-632>; Wed, 11 Jun 1997 20:09:57 +0300 Received: by troll.no id <79573-2487>; Wed, 11 Jun 1997 19:09:38 +0200 Sender: agulbra@troll.no To: jradoff@novalink.com Cc: Matti Aarnio , zmailer@nic.funet.fi Subject: Re: smtp seems to be ignoring a 550 return References: <19970611152757Z16295-632+34@nic.funet.fi> <339EC9A3.3919@novalink.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 11 Jun 1997 19:09:35 +0200 In-Reply-To: jradoff@novalink.com's message of Wed, 11 Jun 1997 11:52:03 -0400 Message-ID: Lines: 13 Return-Path: jradoff@novalink.com (Jon Radoff) > Okay, I give up. How do I unsubscribe from this list? This text is the very tail end of zmailer-2.99.48/README: ...use to subscribe yourself to the list by sending it a message with body content: subscribe zmailer Your Name README leaves unsubscribing as an exercise for the reader. Doesn't look like a very hard exercise, so don't give up yet :) --Arnt From zmailer-owner@nic.funet.fi Wed Jun 11 22:26:49 1997 Received: by nic.funet.fi id <2537-631>; Wed, 11 Jun 1997 22:25:22 +0300 Subject: Re: smtp seems to be ignoring a 550 return From: Matti Aarnio To: bruce@cs.ualberta.ca (Bruce Wm Folliott) Date: Wed, 11 Jun 1997 22:25:11 +0300 (EET DST) Cc: data@amc.ualberta.ca, zmailer@nic.funet.fi In-Reply-To: <19970611161109Z13155-27660+107@scapa.cs.ualberta.ca> from "Bruce Wm Folliott" at Jun 11, 97 10:11:01 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1070 Message-Id: <19970611192522Z2537-631+117@nic.funet.fi> Return-Path: We were able to track this one down, The problem turned out to be mis-detected return type of the sprintf(), which at SunOS 4.x is 'char *', while at Solaris et.al. newer systems it is 'int'. I didn't check if the autoconfig had a hickup this time, or if it is persistent problem, but my test with the function source, and gcc did produce working binary all right... I am not sure there are easy ways to do regression testing so that the autoconfigration can be sure it in itself does not go nuts, and create even greater mess. /Matti Aarnio ... > There is no core file. > > Using `Mail -v`, I can see that smtp sessions are repeated > even though the 550 User unknown condition happens. > > 550 ... User unknown > ordered deletion of task file from scheduler memory (13113-10396) > > It doesn't look like the 'ordered deletion of task file' succeeds. > > I see some $POSTOFFICE/public/v_a10475 type files which have the > verbose listing. I'll send one just in case I have missed somthing .... From zmailer-owner@nic.funet.fi Wed Jun 11 22:42:26 1997 Received: by nic.funet.fi id <650-632>; Wed, 11 Jun 1997 22:40:57 +0300 Subject: Re: patch for 2.99.49beta10 From: Matti Aarnio To: crosser@online.ru Date: Wed, 11 Jun 1997 22:40:51 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Jun 6, 97 10:38:18 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1178 Message-Id: <19970611194057Z650-632+88@nic.funet.fi> Return-Path: > I unstalled this one on our production machine. > One thing noticed that I dislike: it writes in the syslog > > 227686._badheader saved for inspection: message header syntax error! > > and puts a message in POSTOFFICE/postman, not trying to fix > the problem and send the message in some way (?). In my case, > the problem was in non-critical Reply-To: header (which was > broken by broken sendmail on the way). > > In another .badheader file, I cannot find any error and > zmailer did not give any hint about what exactly it disliked. If the files at $POSTOFFICE/postman/ are named: *_* then they ARE processed with warnings, and perhaps sent, perhaps bounced, perhaps.. Anyway, the sender DOES get a reply, and you can delete all such files. Files WITHOUT *_* in their names are the tough nuts, that the router has not been able to handle at all. (No sensible address to return an error message to, no working destination addresses, or some serious system fault. Resubmitting them to router might help, or migh not..) > Old more tolerant behavior looks better... It is still the same as 2.1/2.2.x were. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Jun 11 22:54:37 1997 Received: by nic.funet.fi id <589-632>; Wed, 11 Jun 1997 22:52:47 +0300 Subject: Re: beta10 - too many end up in postman From: Matti Aarnio To: crosser@online.ru Date: Wed, 11 Jun 1997 22:52:43 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Jun 10, 97 09:19:46 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 657 Message-Id: <19970611195247Z589-632+90@nic.funet.fi> Return-Path: > Matti, > > I hate to say it but I had to rollback to .48 because of the number > of messages that end up in POSTOFFICE/postman. There where 725 > during the holidays; I cannot examine all of them. When I started > .48 and moved them all to the router/, they *all* went thru one way > or another (around 200 produced non-delivery reports). > > I kept a copy of postman contents in case you want more information > about these messages. An 'ls' of the directory would be most educative. Also a few of those files that DO NOT have '_' in their names might tell me something. (Propably they are SPAMs..) > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 12 13:59:46 1997 Received: from IDENT-NONSENSE@uni-paderborn.de (port 46558 [131.234.22.30]) by nic.funet.fi with ESMTP id <2348-21399>; Thu, 12 Jun 1997 13:58:48 +0300 Received: from NO-IDENT-SERVICE@krk-fes (port 64633 [131.234.60.10]) by pbinfo.uni-paderborn.de with ESMTP id <52246-18047>; Thu, 12 Jun 1997 12:58:34 +0200 Received: from localhost (localhost [127.0.0.1]) by krk.uni-paderborn.de (8.7.3/8.7.3) with SMTP id MAA24445; Thu, 12 Jun 1997 12:58:27 +0200 (MET DST) Date: Thu, 12 Jun 1997 12:58:27 +0200 (MET DST) From: Swen Thuemmler To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: Re: zmailer-2.99.48.patch3.test released ... In-Reply-To: <19970528090355Z2363-17318+172@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 28 May 1997, Matti Aarnio wrote: > Ok, I made zmailer-2.99.48.patch3.test available at > my workstation ( ftp://mea.tmt.tele.fi/ ). I did compile > test it at my workstation both with and without HAVE_MMAP... > Report on its working, and I will make official release > by storeing it into ftp.funet.fi. I had to make the following patch to rrouter.cf: --- cf.orig/rrouter.cf Tue Jun 10 17:47:58 1997 +++ cf/rrouter.cf Tue Jun 10 18:21:10 1997 @@ -48,7 +48,7 @@ tsift "$address" in # See that it does not start with a pipe ... - [|].+ # Looks like a pipe... Don't mutilate it! + ["][|].+ # Looks like a pipe... Don't mutilate it! break ;; # Now make canonical (.*)<(.*) since one of our users had a forward file with content <"|/usr/local/procmail/bin/procmail -t"@uni-paderborn.de> which caused the router to segfault. I'm not sure, if the patch is correct, an of course the forward file is silly, but it worked before (2.99.43)..., and with the patch, it does work with 2.99.48p3, too. Greetings, Swen From zmailer-owner@nic.funet.fi Fri Jun 13 13:06:25 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 25549 [194.67.3.135]) by nic.funet.fi with ESMTP id <15393-7532>; Fri, 13 Jun 1997 13:05:30 +0300 Received: from NO-IDENT-SERVICE@ariel (port 45060 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68043-22814>; Fri, 13 Jun 1997 14:05:08 +0400 Date: Fri, 13 Jun 1997 14:05:06 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: patch for 2.99.49beta10 To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970611194057Z650-632+88@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > If the files at $POSTOFFICE/postman/ are named: *_* > then they ARE processed with warnings, and perhaps > sent, perhaps bounced, perhaps.. Anyway, the sender > DOES get a reply, and you can delete all such files. > > Files WITHOUT *_* in their names are the tough nuts, Oh, I see... Then, I should not bother too much? I will check if messages with broken headers are actually delivered and if yes, swith to .49 again. Eugene From zmailer-owner@nic.funet.fi Fri Jun 13 17:06:19 1997 Received: by nic.funet.fi id <2393-7532>; Fri, 13 Jun 1997 17:05:20 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49beta11 with smtp policy filters Message-Id: <19970613140520Z2393-7532+373@nic.funet.fi> Date: Fri, 13 Jun 1997 17:05:19 +0300 Return-Path: Hello, At ftp://mea.tmt.tele.fi/zmailer-2.99.49b11.tar.gz are our initial smtp-server resident answers for the SPAM fighting, and relaying control. They are based on Gabor Kissigs ideas, and eat the same config file his does. (But documents are 'a bit' short yet.) Processing is done entirely within the smtpserver without running a router in subprocess. I have added also several 'freeze*' attributes, and support for recognizing usernames ( 'user@domain' and 'user@' ) at the analysis. (The "freeze" means the message is placed into 'POSTOFFICE/deferred/' directory with name of: ${inode}.policyfreeze I am considering creating a new directory for such freeze-outs at the postoffice tree, so that the deferral resubmission would not cause frozen messages to be sent out without administrative action; because often 'zmailer resubmit' is run at cron.. ) The 'makendbm' tool (increasingly inappropriate name) got new options for processing policy-filter db: makendbm -p btree policy-filter < policy.dat It also defaults suffixes to the 3rd parameter (db-file name) which in btree case makes the final name to be: policy-filter.db Similarly the smtpserver does default suffixes by the type, and reference to that file should be: PARAM policydb btree /opt/zmailer/db/policy-filter Be carefull with it, a mistake at the policy input db may block all incoming email! ( And do realize that I am not running it yet myself at my workstation! :-/~ ) I do feel the input policy.dat should be a combination of several input databases, and some standard boiler-plate header/trailers, but as of yet my thoughts there are extremely vague. Suggestions accepted. (For example so that 'localnames' file is one of the inputs, and all domains in 'localnames' are accepted as always valid targets.) (The combination policy should not be coded into 'makendbm', but rather into some SH/PERL utility.) /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jun 16 02:49:34 1997 Received: by nic.funet.fi id <2933-26433>; Mon, 16 Jun 1997 02:49:21 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49 is soon available Message-Id: <19970615234921Z2933-26433+62@nic.funet.fi> Date: Mon, 16 Jun 1997 02:49:17 +0300 Return-Path: ... at ftp://ftp.funet.fi/pub/unix/mail/zmailer/ My all weekend coding and testing marathon has created some sort of documents (!) for the smtp-policy mechanism at the smtpserver. I am using the beast at my workstation, and you are welcome to test its functioning. (Manually, please!) Now it should not allow trivial relaying, and even be able to stop pretty current list of spam-source domains. (proto/policy-builder.sh) While going thru one recent report about problems when some user had an idea of using .forward entry like this: "|some pipe > file"@localhost the router (2.99.48*) had crashed, and the fix had been a change in the MAILSHARE/cf/rrouter.cf. It turned out I had syntactic bug (too simple approach) at 'condquote' routine within the router, and that both the proto/cf/rrouter.cf, and proto/cf/canon.cf did need some changes. When you do 'make install' of 2.99.49, you need to do explicite copy of proto/cf/*.cf into your MAILSHARE/cf/, unless this is your very first installation. Also do note that the POSTOFFICE/ structure got a new directory. A directory called "freezer" into which the smtp-policy can decide to push the messages instead of rejecting all the recipient addresses... /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jun 16 16:45:32 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 7820 [194.67.3.135]) by nic.funet.fi with ESMTP id <2334-26433>; Mon, 16 Jun 1997 16:44:54 +0300 Received: from NO-IDENT-SERVICE@ariel (port 51736 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68046-91>; Mon, 16 Jun 1997 17:44:27 +0400 Date: Mon, 16 Jun 1997 17:44:18 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: non-delivery receipt - to whom? To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: I got a non-delivery report from postmaster@nic.funet.fi in responce to my previous posting to zmailer@nic.funet.fi. I beleive that this is wrong behavior: zmailer-owner@nic.funet.fi *should* have received it. Is it .49 "feature"? Eugene From zmailer-owner@nic.funet.fi Mon Jun 16 17:09:07 1997 Received: by nic.funet.fi id <2393-26433>; Mon, 16 Jun 1997 17:07:46 +0300 Subject: Re: non-delivery receipt - to whom? From: Matti Aarnio To: crosser@online.ru Date: Mon, 16 Jun 1997 17:07:41 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Eugene Crosser" at Jun 16, 97 05:44:18 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 414 Message-Id: <19970616140746Z2393-26433+340@nic.funet.fi> Return-Path: > I got a non-delivery report from postmaster@nic.funet.fi in responce > to my previous posting to zmailer@nic.funet.fi. I beleive that this > is wrong behavior: zmailer-owner@nic.funet.fi *should* have received > it. Is it .49 "feature"? Oh ?? Could you forward it to me ? Or save into a file, and sent that as an attachement, or ... ? Such should not happen, but obviously is.. > Eugene /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jun 16 23:56:45 1997 Received: from NO-IDENT-SERVICE@calvin.math.vt.edu (port 33678 [128.173.42.128]) by nic.funet.fi with ESMTP id <15457-26433>; Mon, 16 Jun 1997 23:55:34 +0300 Received: from mmead@localhost by calvin.math.vt.edu id <57488-10287>; Mon, 16 Jun 1997 16:55:37 -0400 Subject: mail stuck in transport? From: "Matthew C. Mead " To: zmailer@nic.funet.fi Date: Mon, 16 Jun 1997 16:55:23 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 531 Message-Id: <19970616205537Z57488-10287+17@calvin.math.vt.edu> Return-Path: I'm still pretty green with respect to zmailer, so let me ask for help here. I'm testing zmailer on kamin.math.vt.edu. When I mail to "user@math.vt.edu" or "user@calvin.math.vt.edu" (same machine), the mail gets stuck in the transport queue. Here is what I have in my /etc/mail.conf: -- orgdomain=math.vt.edu hostname=kamin.$orgdomain #mydomain= -- Any ideas? Thanks in advance! -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 17 11:25:15 1997 Received: from IDENT-NONSENSE@uni-paderborn.de (port 26240 [131.234.22.30]) by nic.funet.fi with ESMTP id <2975-23506>; Tue, 17 Jun 1997 11:23:32 +0300 Received: from SOCKFAULT1@krk-fes (port 42732 [131.234.60.10]) by pbinfo.uni-paderborn.de with ESMTP id <52606-8175>; Tue, 17 Jun 1997 10:23:15 +0200 Received: from localhost (localhost [127.0.0.1]) by krk.uni-paderborn.de (8.7.3/8.7.3) with SMTP id KAA26545 for ; Tue, 17 Jun 1997 10:23:11 +0200 (MET DST) Date: Tue, 17 Jun 1997 10:23:11 +0200 (MET DST) From: Swen Thuemmler To: zmailer@nic.funet.fi Subject: How to update localnames? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hi, yesterday I changed the db/localnames file (added some entries). This lead to some nasty errors ("trying to talk with myself" and such). I immediately stopped and restarted zmailer, but I wonder what is the correct method to change localnames (and fqdnaliases and aliases). Can I edit a copy and then move it in place or do I have to stop zmailer before changing the file? Normally, I don't want to stop zmailer just for changing one of these files. Now, localnames is an ordered file, so this may be the source of the error, but now I wonder about aliases, which I do edit a lot more often. Is newaliases safe? What are the correct procedures? Thanks, Swen From zmailer-owner@nic.funet.fi Wed Jun 18 02:39:34 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 47241 [128.173.42.130]) by nic.funet.fi with ESMTP id <16213-21916>; Wed, 18 Jun 1997 02:37:57 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148162-589>; Tue, 17 Jun 1997 19:38:18 -0400 Subject: zmailer 2.99.49 router bus errors on FreeBSD 2.1.7 From: "Matthew C. Mead " To: zmailer@nic.funet.fi Date: Tue, 17 Jun 1997 19:38:07 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2694 Message-Id: <19970617233818Z148162-589+3@kamin.math.vt.edu> Return-Path: I'm not sure if this is what's going on in router when a piece of mail gets shoved to it, but this should at least be a start. This is the result of running router -i under gdb and doing a "router you" Any help would be appreciated. Thanks. -matt (gdb) run -i Starting program: /usr/local/zmailer/bin/router -i ZMailer router (2.99.49 #1: Tue Jun 17 13:50:42 EDT 1997) mmead@goof.com:/goof/goof/home/mmead/zmailer-2.99.49/router Copyright 1992 Rayan S. Zachariassen Copyright 1992-1997 Matti Aarnio z# router you : address: you Program received signal SIGBUS, Bus error. 0x3d4c7 in free (cp=0x81dd4) at malloc.c:387 387 UNLINK(p1, bin); (gdb) where #0 0x3d4c7 in free (cp=0x81dd4) at malloc.c:387 #1 0x28c60 in sb_free (n=0) at execute.c:785 #2 0x2944e in runio (ioopp=0xefbf9d30) at execute.c:1016 #3 0x28605 in execute (c=0xefbf9d08, caller=0xefbfadb0, oretcode=0, name=0x1bb1e "") at execute.c:520 #4 0x1bd1b in runcommand (c=0xefbf9d08, pc=0xefbfadb0, retcodep=0xefbfa6c0, cmdname=0x0) at interpret.c:682 #5 0x1ddae in interpret (Vcode=0x78044, Veocode=0x7a228, Ventry=0x7834a, caller=0xefbfadb0, retcodep=0xefbfa6c0, cdp=0x72dd0) at interpret.c:1679 #6 0x28011 in execute (c=0xefbfadb0, caller=0xefbfbd38, oretcode=0, name=0x1bb1e "") at execute.c:412 #7 0x1bd1b in runcommand (c=0xefbfadb0, pc=0xefbfbd38, retcodep=0xefbfb6d8, cmdname=0x7833b "routeuser") at interpret.c:682 #8 0x1ddae in interpret (Vcode=0x7c7f8, Veocode=0x7d8db, Ventry=0x7c900, caller=0xefbfbd38, retcodep=0xefbfb6d8, cdp=0x76f28) at interpret.c:1679 #9 0x28011 in execute (c=0xefbfbd38, caller=0xefbfccc0, oretcode=0, name=0x1bb1e "") at execute.c:412 #10 0x1bd1b in runcommand (c=0xefbfbd38, pc=0xefbfccc0, retcodep=0xefbfc6f0, cmdname=0x7c8f3 "rrouter") at interpret.c:682 #11 0x1ddae in interpret (Vcode=0x732f4, Veocode=0x73ce5, Ventry=0x73b58, caller=0xefbfccc0, retcodep=0xefbfc6f0, cdp=0x67634) at interpret.c:1679 #12 0x28011 in execute (c=0xefbfccc0, caller=0x4fe7c, oretcode=0, name=0x1bb1e "") at execute.c:412 #13 0x1bd1b in runcommand (c=0xefbfccc0, pc=0x4fe7c, retcodep=0xefbfd67c, cmdname=0x73b4c "router") at interpret.c:682 #14 0x1ddae in interpret (Vcode=0x7baf0, Veocode=0x7bb01, Ventry=0x0, caller=0x4fe7c, retcodep=0xefbfd67c, cdp=0x7b4b4) at interpret.c:1679 #15 0x30f2b in zshtoplevel (errname=0x0) at zmsh.c:79 #16 0x1f95 in main (argc=2, argv=0xefbfd748) at router.c:308 -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Wed Jun 18 19:44:58 1997 Received: from NO-IDENT-SERVICE@ph-elec.phys.msu.su (port 33028 [193.232.124.99]) by nic.funet.fi with ESMTP id <15483-21916>; Wed, 18 Jun 1997 19:42:16 +0300 Received: by ph-elec.phys.msu.su id <3905-7881>; Wed, 18 Jun 1997 20:41:42 +0400 Content-Length: 3107 Message-ID: X-Mailer: XFMail 1.0 [p0] on FreeBSD Sender: gena@ph-elec.phys.msu.su Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <19970617233818Z148162-589+3@kamin.math.vt.edu> Date: Wed, 18 Jun 1997 20:39:22 +0400 (MSD) Organization: Physics Department, Moscow State University From: Gennadi Makhmetov To: "Matthew C. Mead" Subject: RE: zmailer 2.99.49 router bus errors on FreeBSD 2.1.7 Cc: zmailer@nic.funet.fi Return-Path: On 17-Jun-97 Matthew C. Mead wrote: > I'm not sure if this is what's going on in router when a piece of >mail gets shoved to it, but this should at least be a start. This is >the result of running router -i under gdb and doing a "router you" > > Any help would be appreciated. Thanks. > > > >-matt > > >(gdb) run -i >Starting program: /usr/local/zmailer/bin/router -i >ZMailer router (2.99.49 #1: Tue Jun 17 13:50:42 EDT 1997) > mmead@goof.com:/goof/goof/home/mmead/zmailer-2.99.49/router >Copyright 1992 Rayan S. Zachariassen >Copyright 1992-1997 Matti Aarnio > >z# router you >: address: you > >Program received signal SIGBUS, Bus error. >0x3d4c7 in free (cp=0x81dd4) at malloc.c:387 >387 UNLINK(p1, bin); >(gdb) where >#0 0x3d4c7 in free (cp=0x81dd4) at malloc.c:387 >#1 0x28c60 in sb_free (n=0) at execute.c:785 >#2 0x2944e in runio (ioopp=0xefbf9d30) at execute.c:1016 >#3 0x28605 in execute (c=0xefbf9d08, caller=0xefbfadb0, oretcode=0, > name=0x1bb1e "") at execute.c:520 >#4 0x1bd1b in runcommand (c=0xefbf9d08, pc=0xefbfadb0, retcodep=0xefbfa6c0, > cmdname=0x0) at interpret.c:682 >#5 0x1ddae in interpret (Vcode=0x78044, Veocode=0x7a228, Ventry=0x7834a, > caller=0xefbfadb0, retcodep=0xefbfa6c0, cdp=0x72dd0) at interpret.c:1679 >#6 0x28011 in execute (c=0xefbfadb0, caller=0xefbfbd38, oretcode=0, > name=0x1bb1e "") at execute.c:412 >#7 0x1bd1b in runcommand (c=0xefbfadb0, pc=0xefbfbd38, retcodep=0xefbfb6d8, > cmdname=0x7833b "routeuser") at interpret.c:682 >#8 0x1ddae in interpret (Vcode=0x7c7f8, Veocode=0x7d8db, Ventry=0x7c900, > caller=0xefbfbd38, retcodep=0xefbfb6d8, cdp=0x76f28) at interpret.c:1679 >#9 0x28011 in execute (c=0xefbfbd38, caller=0xefbfccc0, oretcode=0, > name=0x1bb1e "") at execute.c:412 >#10 0x1bd1b in runcommand (c=0xefbfbd38, pc=0xefbfccc0, retcodep=0xefbfc6f0, > cmdname=0x7c8f3 "rrouter") at interpret.c:682 >#11 0x1ddae in interpret (Vcode=0x732f4, Veocode=0x73ce5, Ventry=0x73b58, > caller=0xefbfccc0, retcodep=0xefbfc6f0, cdp=0x67634) at interpret.c:1679 >#12 0x28011 in execute (c=0xefbfccc0, caller=0x4fe7c, oretcode=0, > name=0x1bb1e "") at execute.c:412 >#13 0x1bd1b in runcommand (c=0xefbfccc0, pc=0x4fe7c, retcodep=0xefbfd67c, > cmdname=0x73b4c "router") at interpret.c:682 >#14 0x1ddae in interpret (Vcode=0x7baf0, Veocode=0x7bb01, Ventry=0x0, > caller=0x4fe7c, retcodep=0xefbfd67c, cdp=0x7b4b4) at interpret.c:1679 >#15 0x30f2b in zshtoplevel (errname=0x0) at zmsh.c:79 >#16 0x1f95 in main (argc=2, argv=0xefbfd748) at router.c:308 > > >-- >Matthew C. Mead Technology Support Staff, > Mathematics Department, >mmead@math.vt.edu Virginia Tech >(540)231-2643 I had the same problem. It was solved by doing configure with option --with-system-malloc --------------------------------- E-Mail: "Gennadi Makhmetov" Date: 18-Jun-97 Time: 20:39:22 From zmailer-owner@nic.funet.fi Wed Jun 18 23:57:50 1997 Received: by nic.funet.fi id <2892-21916>; Wed, 18 Jun 1997 23:57:29 +0300 Subject: Re: zmailer 2.99.49 router bus errors on FreeBSD 2.1.7 From: Matti Aarnio To: gena@ph-elec.phys.msu.su (Gennadi Makhmetov) Date: Wed, 18 Jun 1997 23:57:25 +0300 (EET DST) Cc: mmead@kamin.math.vt.edu, zmailer@nic.funet.fi In-Reply-To: from "Gennadi Makhmetov" at Jun 18, 97 08:39:22 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1719 Message-Id: <19970618205729Z2892-21916+399@nic.funet.fi> Return-Path: This problem, and the solution that people have found for it has prompted me to consider altering defaults around: --with-libmalloc=XXXX where XXXX can be one of: system, malloc_d, malloc Currently the last one is the default, and means the use of the CSRI malloc. These days the system mallocs are usually high performance buddy/arena mallocs, and often can become par with CRSI malloc. How about it ? /Matti Aarnio > On 17-Jun-97 Matthew C. Mead wrote: > > I'm not sure if this is what's going on in router when a piece of > >mail gets shoved to it, but this should at least be a start. This is > >the result of running router -i under gdb and doing a "router you" > > > > Any help would be appreciated. Thanks. > > > >-matt > > > >(gdb) run -i > >Starting program: /usr/local/zmailer/bin/router -i > >ZMailer router (2.99.49 #1: Tue Jun 17 13:50:42 EDT 1997) > > mmead@goof.com:/goof/goof/home/mmead/zmailer-2.99.49/router > >Copyright 1992 Rayan S. Zachariassen > >Copyright 1992-1997 Matti Aarnio > > > >z# router you > >: address: you > > > >Program received signal SIGBUS, Bus error. > >0x3d4c7 in free (cp=0x81dd4) at malloc.c:387 > >387 UNLINK(p1, bin); ...... > >Matthew C. Mead Technology Support Staff, > > Mathematics Department, > >mmead@math.vt.edu Virginia Tech > >(540)231-2643 > > I had the same problem. It was solved by doing configure with option > --with-system-malloc > > --------------------------------- > E-Mail: "Gennadi Makhmetov" > Date: 18-Jun-97 > Time: 20:39:22 > From zmailer-owner@nic.funet.fi Thu Jun 19 00:10:34 1997 Received: from NO-IDENT-SERVICE@fddisunsite.oit.unc.edu (port 53944 [152.2.254.81]) by nic.funet.fi with ESMTP id <15700-30135>; Thu, 19 Jun 1997 00:10:03 +0300 Received: from yangtze.oit.unc.edu (yangtze.oit.unc.edu [152.2.22.118]) by sunsite.unc.edu (8.8.6/8.8.6) with ESMTP id RAA00706; Wed, 18 Jun 1997 17:09:48 -0400 (EDT) Received: from localhost (balld@localhost) by yangtze.oit.unc.edu (8.8.4/8.8.4) with SMTP id RAA28600; Wed, 18 Jun 1997 17:09:48 -0400 X-Authentication-Warning: yangtze.oit.unc.edu: balld owned process doing -bs Date: Wed, 18 Jun 1997 17:09:47 -0400 (EDT) From: Donald Ball X-Sender: balld@yangtze.oit.unc.edu To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: zmailer 2.99.49 sets hostname to "--fqdn"! In-Reply-To: <19970618205729Z2892-21916+399@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Under Solaris 2.5.1, we had a really weird problem last night. The hostname kept getting changed to "--fqdn". We finally tracked it down to zmailer-2.99.49 -- when you put it into rule testing mode, at least, it sets the machine's hostname to "--fqdn". Not cool, not cool at all. - donald From zmailer-owner@nic.funet.fi Thu Jun 19 02:26:18 1997 Received: by nic.funet.fi id <15417-2482>; Thu, 19 Jun 1997 02:26:03 +0300 Subject: Re: zmailer 2.99.49 sets hostname to "--fqdn"! From: Matti Aarnio To: balld@sunsite.unc.edu (Donald Ball) Date: Thu, 19 Jun 1997 02:25:53 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Donald Ball" at Jun 18, 97 05:09:47 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 998 Message-Id: <19970618232603Z15417-2482+44@nic.funet.fi> Return-Path: > Under Solaris 2.5.1, we had a really weird problem last night. The > hostname kept getting changed to "--fqdn". We finally tracked it down to > zmailer-2.99.49 -- when you put it into rule testing mode, at least, it > sets the machine's hostname to "--fqdn". Not cool, not cool at all. .... Oops.. propably your MAILSHARE/router.cf file has lines: domainname=$(/bin/hostname --domain) hostname=$(/bin/hostname --fqdn) That is the last 'Debianism' I can quickly see in the sources. At least some (in this case: Debian) package makers think that their system is the only one in the world, and if I fail to make thorough check (and its related generalization) on the patch I receive, other people end up having problems... Replace that router.cf file with proto/cf/TELE-FI.cf from the source. It is MUCH more generic, and the one I use at a number of hosts. Despite of its name, it has nothing special for Telecom Finland. > - donald /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 19 09:41:09 1997 Received: from NO-IDENT-SERVICE@[206.161.253.66] (port 62756 [206.161.253.66]) by nic.funet.fi with ESMTP id <16048-30135>; Thu, 19 Jun 1997 09:39:45 +0300 Received: from NO-IDENT-SERVICE@localhost (port 14628 [127.0.0.1]) by roadrunner.realbig.com with SMTP id <132904-5061>; Thu, 19 Jun 1997 02:39:50 +0000 Date: Thu, 19 Jun 1997 02:39:48 -0400 (EDT) From: Andy Poling To: Matti Aarnio cc: Gennadi Makhmetov , mmead@kamin.math.vt.edu, zmailer@nic.funet.fi Subject: Re: zmailer 2.99.49 router bus errors on FreeBSD 2.1.7 In-Reply-To: <19970618205729Z2892-21916+399@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Wed, 18 Jun 1997, Matti Aarnio wrote: > This problem, and the solution that people have found for it has > prompted me to consider altering defaults around: > > --with-libmalloc=XXXX > > where XXXX can be one of: system, malloc_d, malloc > Currently the last one is the default, and means > the use of the CSRI malloc. > > These days the system mallocs are usually high performance > buddy/arena mallocs, and often can become par with CRSI > malloc. Exactly. I would consider the system malloc a desireable default. After all anyone who wants to tweak can always change it... -Andy Global Auctions From zmailer-owner@nic.funet.fi Thu Jun 19 10:36:34 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 64203 [193.87.100.12]) by nic.funet.fi with ESMTP id <2781-30135>; Thu, 19 Jun 1997 10:35:28 +0300 Received: by ccduck.ke.sanet.sk id <16222-3>; Thu, 19 Jun 1997 09:34:50 +0200 Subject: RBIFF_ALWAYS From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Thu, 19 Jun 1997 09:34:45 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 422 Message-Id: <97Jun19.093450+0200met_dst.16222-3@ccduck.ke.sanet.sk> Return-Path: HEllo, just downloading 2.99.48 (when will be 3.0 out ?) A question - it took long time to see where is -DRBIFF_ALWAYS defined. Now it takes long time to understand why to do it ? isn't -DRBIFF enough ? -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Thu Jun 19 11:32:48 1997 Received: from IDENT-NONSENSE@dialup15.dynamite.com.au (port 8433 [203.17.154.79]) by nic.funet.fi with ESMTP id <2414-30135>; Thu, 19 Jun 1997 11:31:55 +0300 Received: (from matt@localhost) by firestorm.net.house.au (8.8.5/8.8.5) id SAA06501; Thu, 19 Jun 1997 18:31:14 +1000 Date: Thu, 19 Jun 1997 18:31:14 +1000 Message-Id: <199706190831.SAA06501@firestorm.net.house.au> From: Matt Hawkins To: zmailer@nic.funet.fi Subject: router.cf (Was: Re: zmailer 2.99.49 sets hostname to "--fqdn"!) In-Reply-To: <19970618232603Z15417-2482+44@nic.funet.fi> References: <19970618232603Z15417-2482+44@nic.funet.fi> X-Mailer: VM 6.31 under 19.15p4 XEmacs Lucid Mime-Version: 1.0 (generated by tm-edit 7.106) Content-Type: text/plain; charset=US-ASCII Return-Path: Matti Aarnio writes: > Replace that router.cf file with proto/cf/TELE-FI.cf > from the source. It is MUCH more generic, and the one > I use at a number of hosts. Despite of its name, it > has nothing special for Telecom Finland. Hmm.. didn't seem that way for me when I tried 2.99.48p2 on my Linux box (running kernel v2.0.30) For some reason I could never figure out (and I tried very hard to as I needed some excuse to not revert to *yuk* sendmail) my hostname was set to the domain name - the non-existant private-internal-network name in fact - and all my mail was bounced for 8 hours or more. The return address was set to the non-existant domain, too, which no doubt irritated a lot of postmasters around the world. My problem with mail is that we have a network here at home with the 192.168.0.0 reserved IP range, and a fake domain "net.house.au" All our mail comes from our university account(s) via POP3, and it's nice to be able to send mail from home and masquerade as one of the university servers where we really do exist. I setup zmailer on my machine after a brief fling with qmail which started to play up (like not listening on any port, let alone 25) It appeared to be working fine, yet the next morning after over 300 mails bounced the error was that it couldn't deliver to matt@net.house.au because the domain didn't exist. Our internal DNS was working fine, the MX record for my machine was set up fine, zmailer appeared to be set up fine to recieve mail for firestorm.net.house.au (my machine) so I have no idea why it truncated my hostname, nor why it couldn't resolve from a DNS that was working fine from a shell prompt with dig & nslookup. I'm going to try .49 now - maybe that fixed something :) I'm really tired of sendmail taking forever to send/recieve/deliver mail, and I'd also like to use zmailer on a commercial host of mine, but I can't if it does silly things like the above :( -- Matt PS: if, God forbid, the problem is still there - is it okay to post my zmailer configs to the list? I'm not in the habit of posting long files to large mailing lists... From zmailer-owner@nic.funet.fi Thu Jun 19 11:56:53 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 27084 [193.87.100.12]) by nic.funet.fi with ESMTP id <2206-30135>; Thu, 19 Jun 1997 11:56:09 +0300 Received: by ccduck.ke.sanet.sk id <16239-2>; Thu, 19 Jun 1997 10:55:30 +0200 Subject: procmail as delivery program From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Thu, 19 Jun 1997 10:55:24 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 442 Message-Id: <97Jun19.105530+0200met_dst.16239-2@ccduck.ke.sanet.sk> Return-Path: Hello, some time ago I asked if anyone uses procmail as delivery program to mailboxes. I tried it and It has one problem - local mail is delivered to pipes too and procmail can't handle it (no such user: "|/...") Any help ? -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Thu Jun 19 12:01:07 1997 Received: from root@cal026031.student.utwente.nl (port 26373 [130.89.228.161]) by nic.funet.fi with ESMTP id <15378-30135>; Thu, 19 Jun 1997 11:59:59 +0300 Received: from lists@cal026031.student.utwente.nl (port 21509 [130.89.228.161]) by cal026031.student.utwente.nl with SMTP id <156854-14149>; Thu, 19 Jun 1997 10:59:45 +0200 Date: Thu, 19 Jun 1997 10:59:32 +0200 (MET DST) From: Mark Visser X-Sender: lists@Cal026031.student.utwente.nl Reply-To: mark@cal026031.student.utwente.nl To: Matus.Uhlar@tuke.sk cc: zmailer Subject: Re: procmail as delivery program In-Reply-To: <97Jun19.105530+0200met_dst.16239-2@ccduck.ke.sanet.sk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 19 Jun 1997, Matus Uhlar wrote: > Hello, > > some time ago I asked if anyone uses procmail as delivery program to > mailboxes. I tried it and It has one problem - local mail is delivered to > pipes too and procmail can't handle it (no such user: "|/...") > Any help ? Yep. This was a problem in 2.99.48 and before (at least here).. It was fixed for me in 2.99.48patch2 Try that one, or try 2.99.49 Mark ------------------------------------------------------------------------- Mark Visser | Student Civiele Technologie & Management Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl Telephone: 053-4895038 | ------------------------------------------------------------------------- Warning: You can get rid of all the bugs by disabling them from the main menu. From zmailer-owner@nic.funet.fi Thu Jun 19 12:22:59 1997 Received: from root@wit393101.student.utwente.nl (port 43012 [130.89.235.11]) by nic.funet.fi with ESMTP id <15444-30135>; Thu, 19 Jun 1997 12:21:53 +0300 Received: from sim@localhost (port 35076 [127.0.0.1]) by wit393101.student.utwente.nl with SMTP id <9930-180>; Thu, 19 Jun 1997 11:21:32 +0000 Date: Thu, 19 Jun 1997 11:21:18 +0200 (MET DST) From: Florian Overkamp Sender: sim@wit393101.student.utwente.nl To: Mark Visser cc: Matus.Uhlar@tuke.sk, zmailer Subject: Re: procmail as delivery program In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Thu, 19 Jun 1997, Mark Visser wrote: > > some time ago I asked if anyone uses procmail as delivery program to > > mailboxes. I tried it and It has one problem - local mail is delivered to > > pipes too and procmail can't handle it (no such user: "|/...") > > Any help ? > > Yep. > > This was a problem in 2.99.48 and before (at least here).. > It was fixed for me in 2.99.48patch2 There have been hints on how to use procmail with zmailer since a long time. I had it running based on those hints with versions earlier than .48 It was a bit more tricky than with the current release though... /* Florian Overkamp */ /* PGP Key-ID 233FC67D */ /* PGP Fingerprint 00 3F F5 63 DD FB 5D B7 0A 16 0A EC F9 BD 01 97 */ From zmailer-owner@nic.funet.fi Thu Jun 19 15:58:42 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 28621 [193.87.100.12]) by nic.funet.fi with ESMTP id <2510-2482>; Thu, 19 Jun 1997 15:57:43 +0300 Received: by ccduck.ke.sanet.sk id <16239-2>; Thu, 19 Jun 1997 14:57:23 +0200 Subject: smtp core when delivering to M$ exchange From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Thu, 19 Jun 1997 14:57:21 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1109 Message-Id: <97Jun19.145723+0200met_dst.16239-2@ccduck.ke.sanet.sk> Return-Path: Hello, in last few days we have a little problem - smtp transport agent dies when it tries to deliver mail to machine with Micro$oft exchange. Here's the log: 12748r 220 idea.gratex.sk Microsoft Exchange Internet Mail Service 5.0.1457.7 ready 12748w EHLO ccsun.tuke.sk 12748r 250-idea.gratex.sk supports the following extensions: 12748r 250-XAUTH 12748r 250-XEXCH50 12748r 250-HELP 12748r 250-VRFY 12748r 250-SIZE 0 12748r 250 DSN 12748# 68129-1: We tried to upgrade to new version - we have 2.97 and we tryed to copy just smtp from 2.99.48pl3 but the same happened (only log changed) We tried to turn on debug to see more. but that same happened. just after # number: log entry smtp dumped core. Any fast help ? (some mails have only a few hours to return back :-S) Thanks. (Isn't the problem in "SIZE 0" ?) -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Thu Jun 19 17:40:14 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 60109 [193.87.100.12]) by nic.funet.fi with ESMTP id <2570-30135>; Thu, 19 Jun 1997 17:38:21 +0300 Received: by ccduck.ke.sanet.sk id <16239-4>; Thu, 19 Jun 1997 16:37:54 +0200 Subject: smtp core dumps... From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Thu, 19 Jun 1997 16:37:50 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 998 Message-Id: <97Jun19.163754+0200met_dst.16239-4@ccduck.ke.sanet.sk> Return-Path: This is more info about smtp core dumps by delivering to exchange server 1. we use SunOS 4.1.2 compiled with gcc 2. gdb says (gdb) backtrace #0 0xf77323c0 in nowalgnd () #1 0xf290 in diagnostic (rp=0x7, rc=-127774888, timeout=189576, fmt=0xd
) at diagnostic.c:281 #2 0x35 in ?? () Cannot access memory at address 0x2563259b. that line looks like if (logreport) { /* Right, we have a honour to append our diagnostics to the transport specification file ourselves */ int oldfl = fcntl(rp->desc->ctlfd, F_GETFL); ### int len = 80 + strlen(notarybuf) + strlen(message); ### int rc; long oldalarm; char *sbuf; I'll maybe try more but I don't know what to try :-S -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Thu Jun 19 18:18:25 1997 Received: by nic.funet.fi id <15705-2482>; Thu, 19 Jun 1997 18:17:10 +0300 Subject: Re: smtp core dumps... From: Matti Aarnio To: Matus.Uhlar@tuke.sk Date: Thu, 19 Jun 1997 18:17:07 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <97Jun19.163754+0200met_dst.16239-4@ccduck.ke.sanet.sk> from "Matus Uhlar" at Jun 19, 97 04:37:50 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 799 Message-Id: <19970619151710Z15705-2482+301@nic.funet.fi> Return-Path: > This is more info about smtp core dumps by delivering to exchange server Just what I suspected. I have made 2.99.48.patch5 available (and zmailer-2.99.48p5.tar.gz too) with fixes to this problem, and also with altered defaults on 'libmalloc'. (at ftp.funet.fi) I am also preparing 2.99.49.patch1, but it may take a while before it is ready. (And we are looking into opening 3.0 tree with heavily modified router memory management facilities. It will be -- eh, 'big task'..) ... > I'll maybe try more but I don't know what to try :-S > -- > Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia > E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar > And don't forget: I work for SANET, but talk for myself, dude... ;-) /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 19 18:35:56 1997 Received: by nic.funet.fi id <15469-2482>; Thu, 19 Jun 1997 18:34:24 +0300 Subject: Re: router.cf (Was: Re: zmailer 2.99.49 sets hostname to "--fqdn"!) From: Matti Aarnio To: matt@blitzen.canberra.edu.au (Matt Hawkins) Date: Thu, 19 Jun 1997 18:34:16 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199706190831.SAA06501@firestorm.net.house.au> from "Matt Hawkins" at Jun 19, 97 06:31:14 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2007 Message-Id: <19970619153424Z15469-2482+304@nic.funet.fi> Return-Path: ... > My problem with mail is that we have a network here at home with the > 192.168.0.0 reserved IP range, and a fake domain "net.house.au" > All our mail comes from our university account(s) via POP3, and it's > nice to be able to send mail from home and masquerade as one of the > university servers where we really do exist. > I setup zmailer on my machine after a brief fling with qmail which > started to play up (like not listening on any port, let alone 25) > It appeared to be working fine, yet the next morning after over 300 > mails bounced the error was that it couldn't deliver to > matt@net.house.au because the domain didn't exist. > Our internal DNS was working fine, the MX record for my machine was > set up fine, zmailer appeared to be set up fine to recieve mail for > firestorm.net.house.au (my machine) so I have no idea why it truncated > my hostname, nor why it couldn't resolve from a DNS that was working > fine from a shell prompt with dig & nslookup. I have this file at my workstation: ----- /etc/mail.conf ------- # Zmailer host dependent information # # Where am I? orgdomain=tmt.tele.fi # Who am I? hostname=mea.$orgdomain #hostname=$orgdomain # What do I tell people? #mydomain=$orgdomain mydomain=$hostname # World hack! #world=world.$orgdomain -------------------------------- I have had it for ages... (or rather, nic.funet.fi has had similar for ages..) At the INSTALL file, chapter 7 starts with talk about this file... > I'm going to try .49 now - maybe that fixed something :) I'm really > tired of sendmail taking forever to send/recieve/deliver mail, and I'd > also like to use zmailer on a commercial host of mine, but I can't if > it does silly things like the above :( > > -- Matt > PS: if, God forbid, the problem is still there - is it okay to post my > zmailer configs to the list? I'm not in the habit of posting long files > to large mailing lists... Post them to me, or make them FTP or HTTP downloadable. /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 19 20:22:22 1997 Received: by nic.funet.fi id <946-30135>; Thu, 19 Jun 1997 20:20:02 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49 patch1 Message-Id: <19970619172002Z946-30135+394@nic.funet.fi> Date: Thu, 19 Jun 1997 20:19:55 +0300 Return-Path: Hello, ChangeLog tells: Thu Jun 19 19:41:47 1997 Matti Aarnio * Makefile.in Version 2.99.49p1 * configure.in: Reordered (and changed default) of --with-libmalloc= option parameters * lib/selfaddrs.c: Detected that SunSoft SunPro C 3.0 didn't define '__svr4__', but does '__SVR4' -- automatic Solaris support failed due to that.. * libsh/listmalloc.c: Debugged, and got it finally working.. (something for Zmailer 3.0 ...) * smtpserver/smtpcmds.c: Freed memory blocks that were not malloc()ed at all.. * transports/libta/diagnostics.c: 'notarybuf' can be a NULL pointer, prepare for it! * utils/makendbm/makendbm.c: Just some casts to please SunOS 4.1.x compilation Available at ftp://ftp.funet.fi/pub/unix/mail/zmailer/zmailer-2.99.49.patch1 /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jun 19 22:32:03 1997 Received: from root@babbage.mingpaoxpress.com (port 10747 [205.150.120.3]) by nic.funet.fi with ESMTP id <667-30135>; Thu, 19 Jun 1997 22:30:19 +0300 Received: by www.mingpaoxpress.com id <1930-289>; Thu, 19 Jun 1997 15:29:31 -0400 Received: by www.mingpaoxpress.com id <1929-289>; Thu, 19 Jun 1997 15:29:16 -0400 Path: acli Subject: Re: router.cf (Was: Re: zmailer 2.99.49 sets hostname to "--fqdn"!) Distribution: local Organization: Ming Pao Daily News (Canada) Message-ID: References: <19970618232603Z15417-2482+44@nic.funet.fi> <199706190831.SAA06501@firestorm.net.house.au> Date: Thu, 19 Jun 1997 19:29:02 GMT From: Ambrose Li To: zmailer@nic.funet.fi Reply-To: Ambrose Li Return-Path: In article <199706190831.SAA06501@firestorm.net.house.au>, Matt Hawkins wrote: >Matti Aarnio writes: > > Replace that router.cf file with proto/cf/TELE-FI.cf > > from the source. It is MUCH more generic, and the one > > I use at a number of hosts. Despite of its name, it > > has nothing special for Telecom Finland. > >Hmm.. didn't seem that way for me when I tried 2.99.48p2 on my Linux >box (running kernel v2.0.30) TELE-FI.cf works fine for my installations. At home I also use the 192.168/16 reserved IP range with fake domain names. It never gave me any problems. -- Ambrose C. Li / Please use home address Programmer-analyst / Ming Pao Newspapers (Canada) Ltd. / for private mail +1 416-321-0088 1355 Huntingwood Drive, Scarborough ON, Canada M1S 3J1 From zmailer-owner@nic.funet.fi Fri Jun 20 11:18:25 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 6356 [193.87.100.12]) by nic.funet.fi with ESMTP id <2499-26500>; Fri, 20 Jun 1997 11:17:45 +0300 Received: by ccduck.ke.sanet.sk id <16222-4>; Fri, 20 Jun 1997 10:14:56 +0200 Subject: Re: smtp core dumps... From: Matus Uhlar To: mea@nic.funet.fi (Matti Aarnio) Date: Fri, 20 Jun 1997 10:14:52 +0200 (MET DST) Cc: zmailer@nic.funet.fi (zmailer) Reply-To: Matus.Uhlar@tuke.sk In-Reply-To: <19970619151710Z15705-2482+301@nic.funet.fi> from "Matti Aarnio" at Jun 19, 97 06:17:07 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2052 Message-Id: <97Jun20.101456+0200met_dst.16222-4@ccduck.ke.sanet.sk> Return-Path: -> > This is more info about smtp core dumps by delivering to exchange server -> -> Just what I suspected. I have made 2.99.48.patch5 available -> (and zmailer-2.99.48p5.tar.gz too) with fixes to this problem, -> and also with altered defaults on 'libmalloc'. (at ftp.funet.fi) I installed patch4 and patch5 but this happened: cd ../libident; make libident.a gcc -O2 -DHAVE_CONFIG_H -I../include -I.. -c identuser.c identuser.c:55: conflicting types for `ident_tcpport' identuser.h:12: previous declaration of `ident_tcpport' *** Error code 1 ident_tcpport is defined in identuser.c as int and identuser.h as short because in identuser.c.orig it was short, I changed it to short again gcc -O2 -DHAVE_CONFIG_H -I.. -I../include -target sun4 -c getaddrinfo.c getaddrinfo.c: In function `resolver_lookup_addr': getaddrinfo.c:227: `h_errno' undeclared (first use this function) getaddrinfo.c:227: (Each undeclared identifier is reported only once getaddrinfo.c:227: for each function it appears in.) *** Error code 1 this it worse... I didn't find it anywhere... only in /usr/include/netdb.h: /* * Error return codes from gethostbyname() and gethostbyaddr() * (left in extern int h_errno). */ so I added into libc/getaddrinfo.c #include #ifndef EAI_AGAIN #include "netdb6.h" #endif /* SunOS 4.* doesn't define this */ extern int h_errno; The same for libc/getnameinfo.c -> I am also preparing 2.99.49.patch1, but it may take a while -> before it is ready. hmmm what else changed from 48 to 49 that it's not the same ? -> (And we are looking into opening 3.0 tree with heavily modified -> router memory management facilities. It will be -- eh, 'big task'..) hmmm making zmailer wirking without any bugs would be maybe bigger task :-S (sorry, i even still want zmailer instead of zmailer) -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Fri Jun 20 17:56:36 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 1421 [128.173.42.130]) by nic.funet.fi with ESMTP id <3033-18270>; Fri, 20 Jun 1997 17:55:33 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-589>; Fri, 20 Jun 1997 10:55:43 -0400 Subject: procmail wants to do aliases!? From: "Matthew C. Mead " To: zmailer@nic.funet.fi Date: Fri, 20 Jun 1997 10:55:40 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 974 Message-Id: <19970620145543Z148156-589+8@kamin.math.vt.edu> Return-Path: Well, after the bout of messages regarding using procmail as the mail delivery agent, I decided to try it. I've been using procmail for the MDA under sendmail for about 2 years now with no problems. Unfortunately, it did not work. When mailing to majordomo for processing, something happened during which the mail bounced, and it was attempting to send it to postmaster. I stopped zmailer services and checked some things out - it appears that procmail was trying to deliver to postmaster, which on my system anyway, is an alias. Anyone know why? I changed the definition of mailbox to procm in scheduler.conf as follows: command="sm -8c $channel procm" Whereas it used to be the stock command="mailbox -8" Any help would be greatly appreciated! I like not having .forwards cluttered with |procmail, etc. Thanks! -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Sat Jun 21 16:21:07 1997 Received: from NO-IDENT-SERVICE@trixie.mse.jhu.edu (port 41742 [128.220.8.14]) by nic.funet.fi with ESMTP id <2629-21535>; Sat, 21 Jun 1997 16:20:37 +0300 Received: from see-ya ([128.220.8.107]) by trixie.mse.jhu.edu with SMTP id <39189-5889>; Sat, 21 Jun 1997 09:20:13 -0400 Date: Sat, 21 Jun 1997 09:19:53 -0400 (Eastern Daylight Time) From: Tim DiLauro To: Swen Thuemmler cc: zmailer@nic.funet.fi Subject: Re: How to update localnames? In-Reply-To: Message-ID: X-X-Sender: timmo@racerx.mse.jhu.edu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On Tue, 17 Jun 1997, Swen Thuemmler wrote: > yesterday I changed the db/localnames file (added some entries). This lead > to some nasty errors ("trying to talk with myself" and such). I > immediately stopped and restarted zmailer, but I wonder what is the > correct method to change localnames (and fqdnaliases and aliases). Can I > edit a copy and then move it in place or do I have to stop zmailer before > changing the file? Normally, I don't want to stop zmailer just for > changing one of these files. Now, localnames is an ordered file, so this > may be the source of the error, but now I wonder about aliases, which I > do edit a lot more often. Is newaliases safe? What are the correct > procedures? Swen: Here's what I do. I edit db/localnames and save it to a temporary file. Then I "sort" it back into db/localnames (you can do all this without an extra file from within emacs). Finally, I restart the router(s) with "zmailer router". -timmo Tim DiLauro Milton S. Eisenhower Library Library Systems Jack Johns Hopkins University (410) 516-5263 3400 N. Charles Street timmo@jhu.edu Baltimore, MD 21218 From zmailer-owner@nic.funet.fi Sun Jun 22 01:38:04 1997 Received: from IDENT-NONSENSE@claven.cse.psu.edu (port 52620 [130.203.3.50]) by nic.funet.fi with ESMTP id <2436-12163>; Sun, 22 Jun 1997 01:37:30 +0300 Received: from jovial.cse.psu.edu (groenvel@jovial.cse.psu.edu [130.203.9.26]) by cse.psu.edu (8.8.5/8.7.3) with ESMTP id SAA15904; Sat, 21 Jun 1997 18:37:22 -0400 (EDT) Message-Id: <199706212237.SAA15904@cse.psu.edu> X-Mailer: exmh version 1.6.9 8/22/96 X-uri: X-work-address: Manufacturing Technology - Systems Development Group PO Box 30 The Applied Research Lab State College, PA 16804 X-work-phone: +1 814 863 9896 (Voice) +1 814 863 1183 (Fax) To: Matti Aarnio cc: zmailer@nic.funet.fi Subject: autoconf bug w/ 2.99.49 patch1 In-reply-to: Your message of "Thu, 19 Jun 1997 20:19:55 +0300." <19970619172002Z946-30135+394@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 21 Jun 1997 18:37:23 EDT From: John D Groenveld Return-Path: Generated this distinctly non-Solaris function in SMTP-UUCP.conf. mailconf () { orgdomain=$(/bin/hostname --domain) hostname=$(/bin/hostname --fqdn) # This is what it will say on out mail read mydomain ; Sun, 22 Jun 1997 05:31:55 +0300 Subject: Re: autoconf bug w/ 2.99.49 patch1 From: Matti Aarnio To: groenvel@cse.psu.edu (John D Groenveld) Date: Sun, 22 Jun 1997 05:31:51 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199706212237.SAA15904@cse.psu.edu> from "John D Groenveld" at Jun 21, 97 06:37:23 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1072 Message-Id: <19970622023155Z2530-31789+93@nic.funet.fi> Return-Path: > Generated this distinctly non-Solaris function in SMTP-UUCP.conf. > > mailconf () { > orgdomain=$(/bin/hostname --domain) > hostname=$(/bin/hostname --fqdn) > # This is what it will say on out mail > read mydomain } > > :( Whooops.. Copy TELE-FI.cf into your router.cf, and optionally remove 'usenet' from the list of supported protocols. At 2.99.49p2 and 2.99.48p6 the TELE-FI.cf will be copied into SMTP+UUCP.cf .. (or their *.in versions thereof) ... ok, 2.99.49p2 is now available at my workstation, but I will do a bunch of other system tests before I place it into ftp.funet.fi (and thus finalize it) I am yet to verify that this one compiles cleanly at SunOS 4.1.x systems. It is now an early morning, and I have been up all night -- not that mid-summer nights in Finland are very long for that matter... Even at latitude of 61 degrees north, where I live. (I refrain from giving a short lecture in astronomy here ;) ) > John > groenvel@cse.psu.edu /Matti Aarnio From zmailer-owner@nic.funet.fi Sun Jun 22 14:24:55 1997 Received: by nic.funet.fi id <2536-12163>; Sun, 22 Jun 1997 14:24:14 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49.patch2 available. Message-Id: <19970622112414Z2536-12163+223@nic.funet.fi> Date: Sun, 22 Jun 1997 14:24:11 +0300 Return-Path: Hello, After testing the configuration, compilation, installation and running at Linux 2.1.x/RedHat 4.1 and SunOS 4.1.x (with gcc 2.7.2.2), I am fairly confident that this should work also at other systems. Now back to be a patio potato in Finnish summer -- short as it is, but quite pleasantly warm :-) (circa 90F/25C) /Matti Aarnio diff -u -r zmailer-2.99.49p1/ChangeLog zmailer-2.99.49p2/ChangeLog --- zmailer-2.99.49p1/ChangeLog Thu Jun 19 19:50:30 1997 +++ zmailer-2.99.49p2/ChangeLog Sun Jun 22 05:14:45 1997 @@ -1,3 +1,32 @@ +Sun Jun 22 04:57:02 1997 Matti Aarnio + + * Makefile.in: + Version 2.99.49p2 + + * configure.in: + sprintf() return type testing fix + + * include/hostenv.h: + Too many files need to include , thus + we added it into "hostenv.h"... (for SunOS 4.1.x) + + * libc/getaddrinfo.c, libc/getnameinfo.c: + "extern int h_errno;" added + + * proto/cf/SMTP+UUCP.cf.in + More suitable, and more generic version + + * smtpserver/smtpcmds.c: + A fix in ORCPT pseudo-formation in case the original + sending system did not use the option.. + + * support/vacarion/vacation.c: + Properly placed 'extern char * strerror();' + + * transports/libta/diagnostic.c: + Again optimize int returning sprintf better, than + the old char * returning... + From zmailer-owner@nic.funet.fi Sun Jun 22 19:16:47 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 17551 [128.173.42.130]) by nic.funet.fi with ESMTP id <2079-31789>; Sun, 22 Jun 1997 19:16:16 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-6480>; Sun, 22 Jun 1997 12:16:29 -0400 Subject: Re: How to update localnames? From: "Matthew C. Mead " To: timmo@RacerX.mse.jhu.edu (Tim DiLauro) Date: Sun, 22 Jun 1997 12:16:21 -0400 (EDT) Cc: swen@uni-paderborn.de, zmailer@nic.funet.fi In-Reply-To: from "Tim DiLauro" at Jun 21, 97 09:19:53 am X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1209 Message-Id: <19970622161629Z148156-6480+1@kamin.math.vt.edu> Return-Path: Tim DiLauro writes: > On Tue, 17 Jun 1997, Swen Thuemmler wrote: > > yesterday I changed the db/localnames file (added some entries). This lead > > to some nasty errors ("trying to talk with myself" and such). I > > immediately stopped and restarted zmailer, but I wonder what is the > > correct method to change localnames (and fqdnaliases and aliases). Can I > > edit a copy and then move it in place or do I have to stop zmailer before > > changing the file? Normally, I don't want to stop zmailer just for > > changing one of these files. Now, localnames is an ordered file, so this > > may be the source of the error, but now I wonder about aliases, which I > > do edit a lot more often. Is newaliases safe? What are the correct > > procedures? > Swen: > Here's what I do. > I edit db/localnames and save it to a temporary file. Then I "sort" it > back into db/localnames (you can do all this without an extra file from > within emacs). Finally, I restart the router(s) with "zmailer router". You can also do it in vi: once done editing, just do a :1,$!sort -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Sun Jun 22 22:10:07 1997 Received: by nic.funet.fi id <3031-12163>; Sun, 22 Jun 1997 22:09:24 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: zmailer 2.99.48 patch6 available Message-Id: <19970622190924Z3031-12163+334@nic.funet.fi> Date: Sun, 22 Jun 1997 22:09:14 +0300 Return-Path: With this patch, 2.99.48 series and 2.99.49 series are in same condition -- except that 2.99.48 does not contain smtpserver policyfilter stuff. /Matti Aarnio From zmailer-owner@nic.funet.fi Mon Jun 23 03:52:34 1997 Received: from IDENT-NONSENSE@claven.cse.psu.edu (port 23515 [130.203.3.50]) by nic.funet.fi with ESMTP id <2533-2140>; Mon, 23 Jun 1997 03:52:24 +0300 Received: from jovial.cse.psu.edu (groenvel@jovial.cse.psu.edu [130.203.9.26]) by cse.psu.edu (8.8.5/8.7.3) with ESMTP id UAA22159 for ; Sun, 22 Jun 1997 20:51:59 -0400 (EDT) Message-Id: <199706230051.UAA22159@cse.psu.edu> X-Mailer: exmh version 1.6.9 8/22/96 X-uri: X-work-address: Manufacturing Technology - Systems Development Group PO Box 30 The Applied Research Lab State College, PA 16804 X-work-phone: +1 814 863 9896 (Voice) +1 814 863 1183 (Fax) To: zmailer@nic.funet.fi Subject: identd and bounces Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 22 Jun 1997 20:52:01 EDT From: John D Groenveld Return-Path: How do I disable identd lookups so my headers don't look like Received: from IDENT-NONSENSE@foo.bar.org? How do I get bounces to be CC'd to postmaster? I appologize if I missed these features, I'm coming from 2.99.25 and I didnt see anything in the list archive or via a grep of the docs. Thanks, John groenvel@cse.psu.edu From zmailer-owner@nic.funet.fi Mon Jun 23 19:37:17 1997 Received: from root@lupinella.troll.no (port 22042 [195.0.254.19]) by nic.funet.fi with ESMTP id <2306-24515>; Mon, 23 Jun 1997 19:34:15 +0300 Received: by troll.no id <79769-4817>; Mon, 23 Jun 1997 18:33:49 +0200 Sender: agulbra@troll.no To: John D Groenveld Cc: zmailer@nic.funet.fi Subject: Re: identd and bounces References: <199706230051.UAA22159@cse.psu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 23 Jun 1997 18:33:45 +0200 In-Reply-To: John D Groenveld's message of Sun, 22 Jun 1997 20:52:01 EDT Message-ID: Lines: 14 Return-Path: John D Groenveld > How do I disable identd lookups so my headers don't look like > Received: from IDENT-NONSENSE@foo.bar.org? Remove the -a flag from the smtp server in /etc/zmailer.conf (smtpserver man page has more interesting stuff). > How do I get bounces to be CC'd to postmaster? Edit some or all of your .../forms/* files. The ones in zmailer 2.99.48 have "ADR Cc: The Post Office " all over the place; do something similar. --Arnt From zmailer-owner@nic.funet.fi Tue Jun 24 15:30:53 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 50834 [128.173.42.130]) by nic.funet.fi with ESMTP id <2990-15893>; Tue, 24 Jun 1997 15:30:18 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10530>; Tue, 24 Jun 1997 08:30:34 -0400 Subject: procmail as local MDA pointers? From: "Matthew C. Mead" To: zmailer@nic.funet.fi Date: Tue, 24 Jun 1997 08:30:33 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 426 Message-Id: <19970624123034Z148156-10530+1@kamin.math.vt.edu> Return-Path: I'd like some information on using procmail as the local MDA - anyone have any pointers? Specifically, I would like to prevent zmailer from passing mail onto procmail before it's finished resolving addresses through the aliases file. Any help would be greatly appreciated. Thanks! -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 16:32:58 1997 Received: by nic.funet.fi id <2629-7394>; Tue, 24 Jun 1997 16:30:12 +0300 Subject: Re: procmail as local MDA pointers? From: Matti Aarnio To: mmead@kamin.math.vt.edu (Matthew C. Mead) Date: Tue, 24 Jun 1997 16:30:00 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624123034Z148156-10530+1@kamin.math.vt.edu> from "Matthew C. Mead" at Jun 24, 97 08:30:33 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1305 Message-Id: <19970624133012Z2629-7394+398@nic.funet.fi> Return-Path: > I'd like some information on using procmail as the local MDA - > anyone have any pointers? Specifically, I would like to prevent zmailer > from passing mail onto procmail before it's finished resolving addresses > through the aliases file. Any help would be greatly appreciated. > Thanks! With 2.99.48p6 / 2.99.49p2 the standard configuration will ALMOAST do it all. You need to alter MAILSHARE/scheduler.conf entry of: local/* ... # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! command="mailbox -8" # Or with CYRUS server the following might do: #command="sm -8c $channel cyrus" # Or with PROCMAIL as the local delivery agent: #command="sm -8c $channel procm" Comment the first "command=" entry away, and uncomment the last one. Then all deliveries to local users will go thru procmail, AND if users want to run pipes / store to files thru their .forward files, they can do that too. (That is why there are tag-matchers for "local/file*" and "local/pipe*" before "local/*" in that scheduler.conf file; those are for cases which must not run thru procmail..) > -matt > -- > Matthew C. Mead Technology Support Staff, > Mathematics Department, > mmead@math.vt.edu Virginia Tech /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jun 24 16:37:05 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 9968 [193.87.100.12]) by nic.funet.fi with ESMTP id <2670-15893>; Tue, 24 Jun 1997 16:36:41 +0300 Received: by ccduck.ke.sanet.sk id <16251-2>; Tue, 24 Jun 1997 15:36:09 +0200 Subject: Re: procmail wants to do aliases!? From: Matus Uhlar To: mmead@kamin.math.vt.edu (Matthew C. Mead) Date: Tue, 24 Jun 1997 15:36:04 +0200 (MET DST) Cc: zmailer@nic.funet.fi (zmailer) Reply-To: Matus.Uhlar@tuke.sk In-Reply-To: <19970620145543Z148156-589+8@kamin.math.vt.edu> from "Matthew C. Mead" at Jun 20, 97 10:55:40 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1424 Message-Id: <97Jun24.153609+0200met_dst.16251-2@ccduck.ke.sanet.sk> Return-Path: -> Well, after the bout of messages regarding using procmail as the -> mail delivery agent, I decided to try it. I've been using procmail for -> the MDA under sendmail for about 2 years now with no problems. -> Unfortunately, it did not work. When mailing to majordomo for -> processing, something happened during which the mail bounced, and it was -> attempting to send it to postmaster. I stopped zmailer services and -> checked some things out - it appears that procmail was trying to deliver -> to postmaster, which on my system anyway, is an alias. Anyone know why? -> I changed the definition of mailbox to procm in scheduler.conf as -> follows: -> -> command="sm -8c $channel procm" -> -> Whereas it used to be the stock -> -> command="mailbox -8" -> look at sm.conf procm sSPfnm /usr/local/bin/procmail procmail -a $h -d $u -t it means that procmail will get "-d " and the line command="sm -8c $channel procm" means that ANY local delivery will be passed to sm... so procmail should know how to deliver to pipes/programs or we should tell sm or scheduler only to pass user delivery to procmail. I'd vote for the first way. I'll mail to provcmail maintainers. -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Tue Jun 24 16:47:34 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 37267 [128.173.42.130]) by nic.funet.fi with ESMTP id <2269-7394>; Tue, 24 Jun 1997 16:47:07 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10529>; Tue, 24 Jun 1997 09:47:19 -0400 Subject: Re: procmail as local MDA pointers? From: "Matthew C. Mead" To: mea@nic.funet.fi (Matti Aarnio) Date: Tue, 24 Jun 1997 09:47:16 -0400 (EDT) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624133012Z2629-7394+398@nic.funet.fi> from "Matti Aarnio" at Jun 24, 97 04:30:00 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1787 Message-Id: <19970624134719Z148156-10529+2@kamin.math.vt.edu> Return-Path: Matti Aarnio writes: > > I'd like some information on using procmail as the local MDA - > > anyone have any pointers? Specifically, I would like to prevent zmailer > > from passing mail onto procmail before it's finished resolving addresses > > through the aliases file. Any help would be greatly appreciated. > > Thanks! > With 2.99.48p6 / 2.99.49p2 the standard configuration will > ALMOAST do it all. You need to alter MAILSHARE/scheduler.conf > entry of: > > local/* > ... > # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT > # conversion on flight! > command="mailbox -8" > # Or with CYRUS server the following might do: > #command="sm -8c $channel cyrus" > # Or with PROCMAIL as the local delivery agent: > #command="sm -8c $channel procm" > > Comment the first "command=" entry away, and uncomment the last one. > > Then all deliveries to local users will go thru procmail, AND > if users want to run pipes / store to files thru their .forward > files, they can do that too. (That is why there are tag-matchers > for "local/file*" and "local/pipe*" before "local/*" in that > scheduler.conf file; those are for cases which must not run thru > procmail..) Thanks for the reply. Unfortunately, this is exactly what I did. I commented out command="mailbox -8" and uncommented command="sm -8c $channel procm" in the local/* definition. It was then that I discovered procmail was being asked to deliver mail to aliases, which clearly wouldn't work. Any ideas? What I want to have done is procmail be used *only* for user's mailboxes, and zmailer's mailbox be used for pipes and files. -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 16:49:43 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 37267 [128.173.42.130]) by nic.funet.fi with ESMTP id <2630-15893>; Tue, 24 Jun 1997 16:49:10 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10530>; Tue, 24 Jun 1997 09:49:19 -0400 Subject: Re: procmail wants to do aliases!? From: "Matthew C. Mead" To: Matus.Uhlar@tuke.sk Date: Tue, 24 Jun 1997 09:49:17 -0400 (EDT) Cc: zmailer@nic.funet.fi In-Reply-To: <97Jun24.153609+0200met_dst.16251-2@ccduck.ke.sanet.sk> from "Matus Uhlar" at Jun 24, 97 03:36:04 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1779 Message-Id: <19970624134919Z148156-10530+2@kamin.math.vt.edu> Return-Path: Matus Uhlar writes: > -> Well, after the bout of messages regarding using procmail as the > -> mail delivery agent, I decided to try it. I've been using procmail for > -> the MDA under sendmail for about 2 years now with no problems. > -> Unfortunately, it did not work. When mailing to majordomo for > -> processing, something happened during which the mail bounced, and it was > -> attempting to send it to postmaster. I stopped zmailer services and > -> checked some things out - it appears that procmail was trying to deliver > -> to postmaster, which on my system anyway, is an alias. Anyone know why? > -> I changed the definition of mailbox to procm in scheduler.conf as > -> follows: > -> command="sm -8c $channel procm" > -> Whereas it used to be the stock > -> command="mailbox -8" > look at sm.conf > procm sSPfnm /usr/local/bin/procmail procmail -a $h -d $u -t > it means that procmail will get "-d " > and the line > command="sm -8c $channel procm" > means that ANY local delivery will be passed to sm... > so procmail should know how to deliver to pipes/programs or we should tell > sm or scheduler only to pass user delivery to procmail. > I'd vote for the first way. I'll mail to provcmail maintainers. Yes, the first way is the correct way to handle it - procmail should never be asked to do anything but handle the delivery to a specific user's mailbox, subject to their own .procmailrc of course. However, I'm not following how the line from scheduler.conf (command="sm -8c $channel procm") means ANY local delivery should be sent to procmail when I only define it in local/*. -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 16:54:22 1997 Received: by nic.funet.fi id <2453-7394>; Tue, 24 Jun 1997 16:54:05 +0300 Subject: Re: procmail as local MDA pointers? From: Matti Aarnio To: mmead@kamin.math.vt.edu (Matthew C. Mead) Date: Tue, 24 Jun 1997 16:54:04 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624134719Z148156-10529+2@kamin.math.vt.edu> from "Matthew C. Mead" at Jun 24, 97 09:47:16 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1117 Message-Id: <19970624135405Z2453-7394+411@nic.funet.fi> Return-Path: ... > Thanks for the reply. Unfortunately, this is exactly what I > did. I commented out command="mailbox -8" and uncommented > command="sm -8c $channel procm" in the local/* definition. It was then > that I discovered procmail was being asked to deliver mail to aliases, > which clearly wouldn't work. Any ideas? What I want to have done is > procmail be used *only* for user's mailboxes, and zmailer's mailbox be > used for pipes and files. The alias expansions et.al. happen entirely inside the router. Therefore there must be something else wrong. (The mailbox is also uterly unable to do alias expansions, thus if it works, while the procmail does not...) Could you show (me) a verbose trace of such a delivery test ? Supply also your current scheduler.conf. $ echo test | mailx -v -s test some-test-address > mail.log 2>&1 Follow that log thru -- last line should be: scheduler done processing ... and them mail it to me. > -matt > -- > Matthew C. Mead Technology Support Staff, > Mathematics Department, > mmead@math.vt.edu Virginia Tech /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jun 24 17:41:39 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 43411 [128.173.42.130]) by nic.funet.fi with ESMTP id <2518-7394>; Tue, 24 Jun 1997 17:40:38 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10529>; Tue, 24 Jun 1997 10:40:50 -0400 Subject: Re: procmail as local MDA pointers? From: "Matthew C. Mead" To: mea@nic.funet.fi (Matti Aarnio) Date: Tue, 24 Jun 1997 10:40:36 -0400 (EDT) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624135405Z2453-7394+411@nic.funet.fi> from "Matti Aarnio" at Jun 24, 97 04:54:04 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 9500 Message-Id: <19970624144050Z148156-10529+4@kamin.math.vt.edu> Return-Path: Matti Aarnio writes: > > Thanks for the reply. Unfortunately, this is exactly what I > > did. I commented out command="mailbox -8" and uncommented > > command="sm -8c $channel procm" in the local/* definition. It was then > > that I discovered procmail was being asked to deliver mail to aliases, > > which clearly wouldn't work. Any ideas? What I want to have done is > > procmail be used *only* for user's mailboxes, and zmailer's mailbox be > > used for pipes and files. > The alias expansions et.al. happen entirely inside the router. > Therefore there must be something else wrong. > (The mailbox is also uterly unable to do alias expansions, thus > if it works, while the procmail does not...) Well, something fishy's going on, cause aliases are not expanding prior to hitting procmail: > Could you show (me) a verbose trace of such a delivery test ? > Supply also your current scheduler.conf. > $ echo test | mailx -v -s test some-test-address > mail.log 2>&1 mail.log: ----- router processed message 3721-14928 l <19970624143549Z3721-14928+3@goof.com> e "matthew c. mead" e mmead@goof.com s local mmead mmead@goof.com 10000 r local phoenix phoenix 10000 N ORCPT=rfc822;phoenix headers rewritten using 'internet' function: Received: from mmead@localhost by goof.com id <3721-14928>; Tue, 24 Jun 1997 10:35:49 -0400 From: "matthew c. mead" To: phoenix@goof.com Subject: test Message-Id: <19970624143549Z3721-14928+3@goof.com> Date: Tue, 24 Jun 1997 10:35:36 -0400 router done processing 3721-14928 /usr/local/bin/procmail procmail -f mmead@goof.com -a phoenix -d phoenix Received: from mmead@localhost by goof.com id <3721-14928>; Tue, 24 Jun 1997 10:35:49 -0400 From: "matthew c. mead" To: phoenix@goof.com Subject: test Message-Id: <19970624143549Z3721-14928+3@goof.com> Date: Tue, 24 Jun 1997 10:35:36 -0400 Return-Path: Diagnostic: [exit status 67 (addressee unknown)] scheduler done processing 3721-14928 ----- scheduler.conf: ----- # Do not edit this file, instead edit /goof/goof/home/mmead/zmailer-2.99.49/proto/scheduler.conf # # # Scheduler configuration file # # The scheduler reads this file on startup or when it receives a SIGUSR1 signal # # Every channel/host combination in recipient addresses will be sifted through # the clauses matched in this file, picking up parameters until a clause that # specifies a command. Everything is free-form with three requirements: # Clauses (i.e. the channel/host pattern) start at the beginning of a line. # Clause contents (i.e. the parameters) don't. # Components are separated by whitespace. # NB! For command paths, the "current directory" is MAILBIN/ta # Boilerplate default parameters */* interval=1m idlemax=4m # Max idle for SMTP connections is 5 minutes, don't exceed that! # # (Unless smtp channel becomes a bit smarter on handling it..) # # expire messages after 3 days without full delivery expiry=3d # when the scheduler gets to the end of the retry sequence, # it starts over at some random point in the middle. The # numbers are factors of the scheduling interval. retries="1 1 2 3 5 8 13 21 34" # no default limits on simultaneous transport agents or # connections to a particular host maxchannel=0 maxring=20 # maxta=250 # OSF/1 has 4096 files per process. Don't overbook :-) # # skew is maximum number of tries before the retry time is # aligned to a standard boundary (seconds modulo interval). skew=1 # default uid/gid of transport agents user=root group=daemon # # A flag telling about queue-order.. # ageorder overfeed=150 # Deferred delivery is handled by this transport agent. Deferrals are low # priority, but they tend to bunch up. The 1 channel slot means there will # be lots of contention, and typical checking intervals will be a bit higher # than what is specified (due to waiting for a free slot). hold/* interval=5m maxchannel=1 command=hold # BITNET delivery methods defrt1/* maxchannel=3 command="sm -c $channel defrt1" bsmtp3/* maxchannel=3 command="sm -c $channel bsmtp3" bsmtp3nd/* maxchannel=3 command="sm -c $channel bsmtp3nd" bsmtp3rfc/* maxchannel=3 command="sm -c $channel bsmtp3" bsmtp3ndrfc/* maxchannel=3 command="sm -c $channel bsmtp3nd" # Local delivery: files, processes, user mail local/file* interval=5m idlemax=9m # Originally we had 3 hour expiry, but if your local system goes to # a fizz (freezes, that is), your local mail may start to bounce # before anybody notices anything... expiry=3d # want 20 channel slots, but only one HOST maxchannel=15 maxring=5 # # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! (Can't use CYRUS, nor PROCMAIL here!) command="mailbox -8" local/pipe* interval=5m idlemax=9m # Originally we had 3 hour expiry, but if your local system goes to # a fizz (freezes, that is), your local mail may start to bounce # before anybody notices anything... expiry=3d # want 20 channel slots, but only one HOST maxchannel=15 maxring=5 # # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! (Can't use CYRUS, nor PROCMAIL here!) command="mailbox -8" local/* interval=5m idlemax=9m # Originally we had 3 hour expiry, but if your local system goes to # a fizz (freezes, that is), your local mail may start to bounce # before anybody notices anything... expiry=3d # want 20 channel slots, but only one HOST maxchannel=15 maxring=5 # # Do MIME text/plain; Quoted-Printable -> text/plain; 8BIT # conversion on flight! #command="mailbox -8" # Or with CYRUS server the following might do: #command="sm -8c $channel cyrus" # Or with PROCMAIL as the local delivery agent: command="sm -8c $channel procm" # smtpx is a channel where the delivery is done without checking at MXes; # rather only on A/AAAA (address) entries: smtpx/* maxchannel=90 maxring=10 command="smtp -c smtpx -x -s" # Sometimes we may want to PUNT all out to somewhere without regarding # on what the routing said: # # smtp/* # maxchannel=199 # maxring=5 # command="smtp -F [192.89.123.25] -l /var/log/mail/smtp.punt" # This is a FAST EXPIRY test case.. Will always cause bounce, btw.. smtp/*-gw.funet.fi maxchannel=0 maxring=5 expiry=1m interval=15s retries="1" skew=1 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.rutgers.edu maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.edu maxchannel=199 maxring=20 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.com maxchannel=199 maxring=30 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.uk maxchannel=199 maxring=8 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.ca maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.se maxchannel=199 maxring=7 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.dk maxchannel=199 maxring=7 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.is maxchannel=199 maxring=5 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.no maxchannel=199 maxring=7 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.de maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.uk maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.gov maxchannel=199 maxring=5 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.mil maxchannel=199 maxring=5 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.net maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" smtp/*.org maxchannel=199 maxring=10 command="smtp -s" # -l /var/log/mail/smtp" # Within FUNET we have a bit longer expiry.. smtp/*funet.fi maxchannel=199 maxring=9 # maxta=2 interval=10m retries="1 1 2 3 5 8 13 21 34" skew=1 # Do FORCED MIME-decoding into C-T-E: 8BIT command="smtp -8sl /var/log/mail/smtp" # Within our organization we care more about speed and capacity than connections # The maxchannel value should be larger than the value used by smtp/*, to avoid # some potential state and phase problems in the queues. smtp/*.fi maxchannel=199 maxring=20 interval=10m retries="1 1 2 3 5 8 13 21 34" skew=1 command="smtp -s" # -l /var/log/mail/smtp" # Connections to the outside shouldn't duplicate effort so we only allow one # per destination. smtp/* maxchannel=199 maxring=50 command="smtp -s" # -l /var/log/mail/smtp" # Error messages. Delivery can be retried at leisure. error/* interval=5m idlemax=2m maxchannel=5 command=errormail # UUCP delivery. The "sm" transport agent picks the first host it sees and # will select further recipient addresses with that host only. We tell # the scheduler this with the "byhost" boolean, to avoid a staggered delivery # effect if the scheduler has to discover this on its own. uucp/* maxchannel=5 command="sm -8c $channel uucp" # News delivery. Hostname is always the same here. usenet/* maxchannel=2 command="sm -8c $channel usenet" # UBC EAN X.400 gateway. See comment at UUCP. ean/* maxchannel=1 command="sm -c $channel ean" ----- > Follow that log thru -- last line should be: > scheduler done processing ... > and them mail it to me. Here's phoenix in the /var/zmailer/db/aliases: phoenix: "|/usr/local/majordomo/wrapper resend -l phoenix -h goof.com phoenix-yumayumayumayuga0" -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 18:34:07 1997 Received: by nic.funet.fi id <2751-7394>; Tue, 24 Jun 1997 18:32:56 +0300 Subject: Re: procmail as local MDA pointers? From: Matti Aarnio To: mmead@kamin.math.vt.edu (Matthew C. Mead) Date: Tue, 24 Jun 1997 18:32:44 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624144050Z148156-10529+4@kamin.math.vt.edu> from "Matthew C. Mead" at Jun 24, 97 10:40:36 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1579 Message-Id: <19970624153256Z2751-7394+448@nic.funet.fi> Return-Path: > Matti Aarnio writes: > > > > Thanks for the reply. Unfortunately, this is exactly what I > > > did. I commented out command="mailbox -8" and uncommented > > > command="sm -8c $channel procm" in the local/* definition. It was then > > > that I discovered procmail was being asked to deliver mail to aliases, > > > which clearly wouldn't work. Any ideas? What I want to have done is > > > procmail be used *only* for user's mailboxes, and zmailer's mailbox be > > > used for pipes and files. > > > The alias expansions et.al. happen entirely inside the router. > > Therefore there must be something else wrong. > > (The mailbox is also uterly unable to do alias expansions, thus > > if it works, while the procmail does not...) > > Well, something fishy's going on, cause aliases are not > expanding prior to hitting procmail: I called Matthew, and asked him to do a couple tests. He had stopped the router, and in interactive mode the router gave successfull routing result. Then he started the ZMailer, and everything worked as they should. I am inclined to believe, this problem was somehow related to the sequence of doing commands: # newaliases # zmailer start If the aliases database is not at all available at the time of the router start, the router will not find it latter, but must be started again (no need to stop it before start..): # zmailer start router > -- > Matthew C. Mead Technology Support Staff, > Mathematics Department, > mmead@math.vt.edu Virginia Tech > (540)231-2643 /Matti Aarnio From zmailer-owner@nic.funet.fi Tue Jun 24 18:44:45 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 54163 [128.173.42.130]) by nic.funet.fi with ESMTP id <2239-7394>; Tue, 24 Jun 1997 18:43:42 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10529>; Tue, 24 Jun 1997 11:43:52 -0400 Subject: Re: procmail as local MDA pointers? From: "Matthew C. Mead" To: mea@nic.funet.fi (Matti Aarnio) Date: Tue, 24 Jun 1997 11:43:48 -0400 (EDT) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624153256Z2751-7394+448@nic.funet.fi> from "Matti Aarnio" at Jun 24, 97 06:32:44 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 1919 Message-Id: <19970624154352Z148156-10529+6@kamin.math.vt.edu> Return-Path: Matti Aarnio writes: > > Matti Aarnio writes: > > > > Thanks for the reply. Unfortunately, this is exactly what I > > > > did. I commented out command="mailbox -8" and uncommented > > > > command="sm -8c $channel procm" in the local/* definition. It was then > > > > that I discovered procmail was being asked to deliver mail to aliases, > > > > which clearly wouldn't work. Any ideas? What I want to have done is > > > > procmail be used *only* for user's mailboxes, and zmailer's mailbox be > > > > used for pipes and files. > > > The alias expansions et.al. happen entirely inside the router. > > > Therefore there must be something else wrong. > > > (The mailbox is also uterly unable to do alias expansions, thus > > > if it works, while the procmail does not...) > > Well, something fishy's going on, cause aliases are not > > expanding prior to hitting procmail: > I called Matthew, and asked him to do a couple tests. > He had stopped the router, and in interactive mode the > router gave successfull routing result. > Then he started the ZMailer, and everything worked as they > should. > I am inclined to believe, this problem was somehow related > to the sequence of doing commands: > # newaliases > # zmailer start > If the aliases database is not at all available at the time > of the router start, the router will not find it latter, > but must be started again (no need to stop it before start..): > # zmailer start router Thanks again for the help! I must say it was quite a surprise to get an actual phone call regarding this! I've run some more tests and things seem to be hunky dory, and users bursting out 400 200K messages within the space of 5 minutes don't bring the system to its knees like sendmail did! Thanks! -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 19:27:25 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 62611 [128.173.42.130]) by nic.funet.fi with ESMTP id <2749-7394>; Tue, 24 Jun 1997 19:25:57 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10528>; Tue, 24 Jun 1997 12:26:07 -0400 Subject: procmail as MDA still not working From: "Matthew C. Mead" To: zmailer@nic.funet.fi Date: Tue, 24 Jun 1997 12:26:06 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 2412 Message-Id: <19970624162607Z148156-10528+4@kamin.math.vt.edu> Return-Path: Well, I've now caught it in the act again. :( I've got a fairly large mailing list on this machine handled by majordomo. I just found these in my syslog LOG_MAIL facility output: Jun 24 12:22:29 goof procmail[17218]: Unknown user "shy-k9s-approval" Jun 24 12:22:29 goof sm[16074]: S.nfzEi3968: to=, delay=00:00:23, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] Jun 24 12:22:29 goof procmail[17219]: Unknown user "postmaster" Jun 24 12:22:29 goof sm[16334]: S.nfzEu3739: to=, delay=00:00:11, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] Jun 24 12:22:30 goof procmail[17220]: Unknown user "shy-k9s-approval" Jun 24 12:22:30 goof sm[16074]: S.nfzEu3840: to=, delay=00:00:12, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] Jun 24 12:22:30 goof procmail[17221]: Unknown user "postmaster" Jun 24 12:22:30 goof sm[16074]: S.nfzEs3776: to=, delay=00:00:14, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] Jun 24 12:22:31 goof procmail[17222]: Unknown user "shy-k9s-approval" Jun 24 12:22:31 goof sm[16334]: S.nfzEr3793: to=, delay=00:00:16, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] Jun 24 12:22:32 goof procmail[17223]: Unknown user "postmaster" Jun 24 12:22:32 goof sm[16074]: S.nfzEq3967: to=, delay=00:00:18, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] Jun 24 12:22:33 goof procmail[17224]: Unknown user "shy-k9s-approval" Jun 24 12:22:33 goof sm[16334]: S.nfzEw3844: to=, delay=00:00:13, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] So clearly, aliases are not being expanded prior to getting to procmail, at least in some circumstances. Anyone have any ideas? I've gotta move back to sendmail for the time being to keep these lists alive. -matt -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Tue Jun 24 19:33:28 1997 Received: from NO-IDENT-SERVICE@kamin.math.vt.edu (port 63635 [128.173.42.130]) by nic.funet.fi with ESMTP id <2642-7394>; Tue, 24 Jun 1997 19:32:31 +0300 Received: from mmead@localhost by kamin.math.vt.edu id <148156-10530>; Tue, 24 Jun 1997 12:32:37 -0400 Subject: Re: procmail as MDA still not working From: "Matthew C. Mead" To: mmead@kamin.math.vt.edu (Matthew C. Mead) Date: Tue, 24 Jun 1997 12:32:24 -0400 (EDT) Cc: zmailer@nic.funet.fi In-Reply-To: <19970624162607Z148156-10528+4@kamin.math.vt.edu> from "Matthew C. Mead" at Jun 24, 97 12:26:06 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Length: 8444 Message-Id: <19970624163237Z148156-10530+4@kamin.math.vt.edu> Return-Path: Matthew C. Mead writes: > > Well, I've now caught it in the act again. :( I've got a fairly > large mailing list on this machine handled by majordomo. I just found > these in my syslog LOG_MAIL facility output: > > Jun 24 12:22:29 goof procmail[17218]: Unknown user "shy-k9s-approval" > Jun 24 12:22:29 goof sm[16074]: S.nfzEi3968: to=, delay=00:00:23, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] > Jun 24 12:22:29 goof procmail[17219]: Unknown user "postmaster" > Jun 24 12:22:29 goof sm[16334]: S.nfzEu3739: to=, delay=00:00:11, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] > Jun 24 12:22:30 goof procmail[17220]: Unknown user "shy-k9s-approval" > Jun 24 12:22:30 goof sm[16074]: S.nfzEu3840: to=, delay=00:00:12, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] > Jun 24 12:22:30 goof procmail[17221]: Unknown user "postmaster" > Jun 24 12:22:30 goof sm[16074]: S.nfzEs3776: to=, delay=00:00:14, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] > Jun 24 12:22:31 goof procmail[17222]: Unknown user "shy-k9s-approval" > Jun 24 12:22:31 goof sm[16334]: S.nfzEr3793: to=, delay=00:00:16, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] > Jun 24 12:22:32 goof procmail[17223]: Unknown user "postmaster" > Jun 24 12:22:32 goof sm[16074]: S.nfzEq3967: to=, delay=00:00:18, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "postmaster"[exit status 67 (addressee unknown)] > Jun 24 12:22:33 goof procmail[17224]: Unknown user "shy-k9s-approval" > Jun 24 12:22:33 goof sm[16334]: S.nfzEw3844: to=, delay=00:00:13, xdelay=00:00:00, mailer=local, stat=error2 procmail: Unknown user "shy-k9s-approval"[exit status 67 (addressee unknown)] > > > So clearly, aliases are not being expanded prior to getting to > procmail, at least in some circumstances. Anyone have any ideas? I've > gotta move back to sendmail for the time being to keep these lists > alive. Ok, I thought this might help - here's my aliases file: ----- # # @(#)aliases 5.3 (Berkeley) 5/24/90 # # Aliases in this file will NOT be expanded in the header from # Mail, but WILL be visible over networks or from /bin/mail. # # >>>>>>>>>> The program "newaliases" must be run after # >> NOTE >> this file is updated for any changes to # >>>>>>>>>> show through to sendmail. # # Basic system aliases -- these MUST be present MAILER-DAEMON: postmaster postmaster: root # General redirections for pseudo accounts bin: root daemon: root games: root ingres: root system: root toor: root uucp: root news: root usenet: news postoffice: root zooz: root # Well-known aliases -- these should be filled in! root: mmead manager: root dumper: root operator: root # local definitions meads: mmead, rmead@clark.net, rmead@arinc.com, margiem@helix.org, nmead draiodoir: ppm@nrao.edu nobody: /dev/null www: http http: mmead,fbm ftp: mmead toon: ":include:/goof/goof/home/jlido/.toon-list" toon-devel: ":include:/goof/goof/home/jlido/.toon-devel-list" mike: mwilson # PNetrek.ORG aliases clue: ":include:/goof/goof/home/mmead/txt/clue.list" plc: ":include:/goof/goof/home/mmead/txt/plc.list" Dawn.Patrol: ":include:/goof/goof/home/mmead/txt/dawn.patrol.list" ICDFH: Dawn.Patrol Interstellar.Crack.Dealers.From.Hell: Dawn.Patrol plague: ":include:/goof/goof/home/mmead/txt/plague.list" team80s: ":include:/goof/goof/home/mmead/txt/team80s.list" team80s-subs: ":include:/goof/goof/home/mmead/txt/team80s-subs.list" # majordomo base stuff majordomo: "|/usr/local/majordomo/wrapper majordomo" majordom: majordomo majordodo: majordomo owner-majordomo: mmead, pmurphy majordomo-owner: owner-majordomo # For the "bounces" list. bounces: "|/usr/local/majordomo/wrapper resend -l bounces -h goof.com bounces-yumayumayumayuga0" bounces-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/bounces" owner-bounces: owner-majordomo owner-bounces-yumayumayumayuga0: owner-majordomo bounces-approval: owner-majordomo # shy-k9s stuff shy-k9s: "|/usr/local/majordomo/wrapper resend -l shy-k9s -h goof.com shy-k9s-yumayumayumayuga0" shy-k9s-digest: shy-k9s shy-k9s-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/shy-k9s.qmail", "|/usr/local/majordomo/wrapper digest -r -C -l shy-k9s-digest shy-k9s-digest-yumayumayumayuga0", "|/usr/local/majordomo/wrapper archive2.pl -a -m -f /usr/local/majordomo/archive/shy-k9s/shy-k9s" shy-k9s-digest-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/shy-k9s-digest.qmail" shy-k9s-request: "|/usr/local/majordomo/wrapper request-answer shy-k9s" shy-k9s-digest-request: "|/usr/local/majordomo/wrapper request-answer shy-k9s-digest" owner-shy-k9s: kim@k9.cv.nrao.edu owner-shy-k9s-yumayumayumayuga0: owner-shy-k9s owner-shy-k9s-digest: owner-shy-k9s owner-shy-k9s-digest-yumayumayumayuga0: owner-shy-k9s shy-k9s-approval: owner-shy-k9s shy-k9s-digest-approval: owner-shy-k9s # phoenix stuff phoenix: "|/usr/local/majordomo/wrapper resend -l phoenix -h goof.com phoenix-yumayumayumayuga0" phoenix-digest: phoenix phoenix-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/phoenix.qmail", "|/usr/local/majordomo/wrapper digest -r -C -l phoenix-digest phoenix-digest-yumayumayumayuga0", "|/usr/local/majordomo/wrapper archive2.pl -a -m -f /usr/local/majordomo/archive/phoenix/phoenix" phoenix-digest-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/phoenix-digest.qmail" phoenix-request: "|/usr/local/majordomo/wrapper request-answer phoenix" phoenix-digest-request: "|/usr/local/majordomo/wrapper request-answer phoenix-digest" owner-phoenix: mmead@goof.com owner-phoenix-yumayumayumayuga0: owner-phoenix owner-phoenix-digest: owner-phoenix owner-phoenix-digest-yumayumayumayuga0: owner-phoenix phoenix-approval: owner-phoenix phoenix-digest-approval: owner-phoenix # bible stuff bible: "|/usr/local/majordomo/wrapper resend -l bible -h goof.com bible-yumayumayumayuga0" bible-digest: bible bible-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/bible.qmail", "|/usr/local/majordomo/wrapper digest -r -C -l bible-digest bible-digest-yumayumayumayuga0", "|/usr/local/majordomo/wrapper archive2.pl -a -m -f /usr/local/majordomo/archive/bible/bible" bible-digest-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/bible-digest.qmail" bible-request: "|/usr/local/majordomo/wrapper request-answer bible" bible-digest-request: "|/usr/local/majordomo/wrapper request-answer bible-digest" owner-bible: mfender@mindspring.com owner-bible-yumayumayumayuga0: owner-bible owner-bible-digest: owner-bible owner-bible-digest-yumayumayumayuga0: owner-bible bible-approval: owner-bible bible-digest-approval: owner-bible # paradise-workers stuff paradise-workers: "|/usr/local/majordomo/wrapper resend -l paradise-workers -h pnetrek.org paradise-workers-yumayumayumayuga0" paradise-workers-digest: paradise-workers paradise-workers-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/paradise-workers.qmail", "|/usr/local/majordomo/wrapper digest -r -C -l paradise-workers-digest paradise-workers-digest-yumayumayumayuga0", "|/usr/local/majordomo/wrapper archive2.pl -a -m -f /usr/local/majordomo/archive/paradise-workers/paradise-workers" paradise-workers-digest-yumayumayumayuga0: ":include:/usr/local/majordomo/lists/paradise-workers-digest.qmail" paradise-workers-request: "|/usr/local/majordomo/wrapper request-answer paradise-workers" paradise-workers-digest-request: "|/usr/local/majordomo/wrapper request-answer paradise-workers-digest" owner-paradise-workers: mmead@goof.com owner-paradise-workers-yumayumayumayuga0: owner-paradise-workers owner-paradise-workers-digest: owner-paradise-workers owner-paradise-workers-digest-yumayumayumayuga0: owner-paradise-workers paradise-workers-approval: owner-paradise-workers paradise-workers-digest-approval: owner-paradise-workers ------ -- Matthew C. Mead Technology Support Staff, Mathematics Department, mmead@math.vt.edu Virginia Tech (540)231-2643 From zmailer-owner@nic.funet.fi Wed Jun 25 11:47:56 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 25077 [193.87.100.12]) by nic.funet.fi with ESMTP id <2781-4901>; Wed, 25 Jun 1997 11:47:13 +0300 Received: by ccduck.ke.sanet.sk id <16239-2>; Wed, 25 Jun 1997 10:46:41 +0200 Subject: 48+patch6 From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Wed, 25 Jun 1997 10:46:35 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 818 Message-Id: <97Jun25.104641+0200met_dst.16239-2@ccduck.ke.sanet.sk> Return-Path: Hello, again problems with compiling after applying patch 6 to zmailer-2.99.48 cd ../libident; make libident.a gcc -O2 -DHAVE_CONFIG_H -I../include -I.. -c identuser.c identuser.c:55: conflicting types for dent_tcpport' identuser.h:12: previous declaration of dent_tcpport' identuser.c:149: conflicting types for dent_sockuser2' identuser.h:15: previous declaration of dent_sockuser2' identuser.c:247: conflicting types for dent_tcpuser9' identuser.h:17: previous declaration of dent_tcpuser9' *** Error code 1 too bad. Q: what does that checking in 49 ? does it check who can mail to who ? -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Wed Jun 25 12:55:59 1997 Received: from NO-IDENT-SERVICE@ccduck.ke.sanet.sk (port 49909 [193.87.100.12]) by nic.funet.fi with ESMTP id <15372-18837>; Wed, 25 Jun 1997 12:54:51 +0300 Received: by ccduck.ke.sanet.sk id <16252-3>; Wed, 25 Jun 1997 11:54:05 +0200 Subject: dropping connections From: Matus Uhlar To: zmailer@nic.funet.fi (zmailer) Date: Wed, 25 Jun 1997 11:53:53 +0200 (MET DST) Reply-To: Matus.Uhlar@tuke.sk X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 770 Message-Id: <97Jun25.115405+0200met_dst.16252-3@ccduck.ke.sanet.sk> Return-Path: HEllo, do you thing it's bad idea to drop connection from machine which provides wrong 'hello machine.name' ? or just ignore everything provided until machine provides acceptable machine name ? acceptable name can be get from DNS by trying lookups on IP -> hostname (and probably back), given hostname etc etc... I've heard RFC (don't know which one) says mailer shouldn't drop connection when provided hostname is wrong. I even think DNS should work and if not, that's the problem. Problems with spam are too big now. (or should I write new RFC ?) -- Matus "fantomas" Uhlar, sysadmin of one server at SANET Kosice, Slovakia E-mail: Matus.Uhlar@tuke.sk ; WWW: http://ccsun.tuke.sk/users/uhlar And don't forget: I work for SANET, but talk for myself, dude... ;-) From zmailer-owner@nic.funet.fi Wed Jun 25 14:37:59 1997 Received: from root@lupinella.troll.no (port 81 [195.0.254.19]) by nic.funet.fi with ESMTP id <2968-4901>; Wed, 25 Jun 1997 14:37:22 +0300 Received: by troll.no id <79711-4814>; Wed, 25 Jun 1997 13:37:00 +0200 Sender: agulbra@troll.no To: Matus.Uhlar@tuke.sk Cc: zmailer@nic.funet.fi (zmailer) Subject: Re: 48+patch6 References: <97Jun25.104641+0200met_dst.16239-2@ccduck.ke.sanet.sk> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 25 Jun 1997 13:36:55 +0200 In-Reply-To: Matus Uhlar's message of Wed, 25 Jun 1997 10:46:35 +0200 (MET DST) Message-ID: Lines: 10 Return-Path: Matus Uhlar > too bad. Q: what does that checking in 49 ? does it check who can mail to > who ? Yes. I built it yesterday, and probably will switch my production box from .48 to .49 tomorrow. It looks rather good, I think. (I had to add -g to the smtpserver options in order to check that it worked, though. Perhaps the docs should recommend "smtpserver -g -i".) --Arnt From zmailer-owner@nic.funet.fi Wed Jun 25 14:44:46 1997 Received: from root@lupinella.troll.no (port 36695 [195.0.254.19]) by nic.funet.fi with ESMTP id <2438-18837>; Wed, 25 Jun 1997 14:43:58 +0300 Received: by troll.no id <79711-4815>; Wed, 25 Jun 1997 13:43:31 +0200 Sender: agulbra@troll.no To: Matus.Uhlar@tuke.sk Cc: zmailer@nic.funet.fi (zmailer) Subject: Re: dropping connections References: <97Jun25.115405+0200met_dst.16252-3@ccduck.ke.sanet.sk> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 25 Jun 1997 13:43:26 +0200 In-Reply-To: Matus Uhlar's message of Wed, 25 Jun 1997 11:53:53 +0200 (MET DST) Message-ID: Lines: 25 Return-Path: Matus Uhlar > do you thing it's bad idea to drop connection from machine which provides > wrong 'hello machine.name' ? Yes. I looked at doing that, and it's rather a bad idea. Too much legitimate mail would be bounced. The HELO name has never mattered, so people who get it wrong don't fix it. Mine may wrong at the moment, I think :) > or just ignore everything provided until > machine provides acceptable machine name ? acceptable name can be get from > DNS by trying lookups on IP -> hostname (and probably back), given hostname > etc etc... The smtpserver ignores the HELO name unless you say -g, I just learned. The default is to use a DNS PTR lookup. > I've heard RFC (don't know which one) says mailer shouldn't drop connection > when provided hostname is wrong. I even think DNS should work and if not, > that's the problem. Problems with spam are too big now. People provide CNAMEs and whatnot. --Arnt From zmailer-owner@nic.funet.fi Sat Jun 28 21:05:01 1997 Received: from root@lupinella.troll.no (port 20775 [195.0.254.19]) by nic.funet.fi with ESMTP id <16241-25472>; Sat, 28 Jun 1997 21:04:13 +0300 Received: by troll.no id <79570-8370>; Sat, 28 Jun 1997 20:04:02 +0200 Sender: agulbra@troll.no To: zmailer@nic.funet.fi Subject: .49, spam Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 28 Jun 1997 20:03:57 +0200 Message-ID: Lines: 33 Return-Path: ZMailer 2.99.49 looks very good (thanks, Matti), but I'm feeling a little down at the moment. The fourth message coming in after I upgraded my real mail server to 2.99.49p2 was spam. 1, 2, 3, spam. Didn't even take a minute. The message came from something like <1fvdf@ejfvsd.afhdjalf3.vmx>, and I don't see any way to reject things like that. Could smtpserver do an MX/A lookup of the envelope sender domain? Would that be a bad idea? I know about -sf, but that seems like policy ("we ignore mail from such and such domain"), not a DNS query. Btw, this patch ought to make policy-builder.sh a little better. --- utils/policy-builder.sh Sat Jun 28 16:10:22 1997 +++ /local/zmailer/bin/policy-builder.sh Sat Jun 28 16:50:02 1997 @@ -55,7 +55,7 @@ # (We use file from "http://www.webeasy.com:8080/spam/spam_download_table" # which is intended for QMAIL, and thus needs to be edited..) if [ -f smtp-policy.spam ] ; then - cat smtp-policy.spam | \ + cat smtp-policy.spam | tr '[A-Z]' '[a-a]' | grep -v '[^-/\.a-zA-Z0-9@]' \ awk '/^@/{ # Domain, but ZMailer processes them without '@' printf "%s = _bulk_mail\n", substr($1,2); next; The tr handles things like cyber-s-i-q.com vs. cyber-S-I-Q.com, the grep (untested, I did that by hand) should grep away stuff zmailer won't digest, "bhaffle,wuyffle.com" for example (note comma). --Arnt From zmailer-owner@nic.funet.fi Sun Jun 29 04:52:27 1997 Received: by nic.funet.fi id <16347-29341>; Sun, 29 Jun 1997 04:51:08 +0300 Subject: Re: .49, spam From: Matti Aarnio To: agulbra@troll.no (Arnt Gulbrandsen) Date: Sun, 29 Jun 1997 04:50:56 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Arnt Gulbrandsen" at Jun 28, 97 08:03:57 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 4390 Message-Id: <19970629015108Z16347-29341+95@nic.funet.fi> Return-Path: > ZMailer 2.99.49 looks very good (thanks, Matti), but I'm feeling a > little down at the moment. > > The fourth message coming in after I upgraded my real mail server to > 2.99.49p2 was spam. 1, 2, 3, spam. Didn't even take a minute. I didn't feel very good either when at last weekend somebody at Santa Ana, California dialed in and did relay-hijack spam of marketing stuff all around with claiming source addresses of: asdfg@utu.fi, 536533@utu.fi, and some such bogus ones. Of course and got hate-mail.. The mechanisms we have are able to limit/block relay mailer hijack, but now that SPAMmers appear to have started individual message delivery straight off their machines to the destination addresses as per MX/A resolving, much isn't doable. One needs near instantaneous knowledge of spam sources, or a real AI system for message content analysis/spam filtering, but ... (if you can find this latter, I would adapt it into an auto-secretary for me ;-) I have been dreaming of it since 1986 when I go to EARN/BITNET, and had 'mail flood' of 20 messages per day... ) Of course we could consider similar approaces to the problem as virus scanner makers are using. Just scan the message body for magic identification factors, and bingo... but what will happen, if somebody forwards a message with matching string, but with legitimate content ? Like when sending such a mail to postmaster, or abuse, or ... > The message came from something like <1fvdf@ejfvsd.afhdjalf3.vmx>, and > I don't see any way to reject things like that. Could smtpserver do > an MX/A lookup of the envelope sender domain? Would that be a bad > idea? Some sites already do, AOL.COM, for example. They seem to run massive amounts (10-20) parallel input systems all running sendmails to verify incoming MAIL FROM:<...> addresses per MX/A data, however ... ... I have spent a day at fighting incoming FLOOD of error messages from some SMAP/sendmail pair sending same messages over and over again. So far vger.rutgers.edu has received some 26 000 of those messages, several per second. The Vger did have several very high load bursts -- up to 170! with these incoming messages, until I realized that doing ident lookup (the '-a' option at the smtpserver) delays the server too much! Also apparently the feeding system is a lot bigger one, than Vger. Vger is a sparc-classic w/ 32 MB RAM! That has teached to me that MX/A resolving - while nice idea - is not necessarily something you can do without multiple external input front-end machines, or you soon run out of resources. By the way, we solved the problem to a large extent by defining the error receiving address ( owner-some-list@ver.rurgers.edu ) to have policy attribute: 'freeze +' (and $POSTOFFICE/freezer/ directory had to be created there too) The smtpserver shunts all the most volumous errors into a freezer box, and leaves the system to process far fewer legitimate msgs. I had to patch the smtpserver/policytest.c to also test the full target address, and not just the target host. Noticing at Vger often even as much as 20 incoming smtpserver sessions from the SAME source IP address gets me wondering, if there would be sense to have a feature at the smtpserver, that it could refuse to server the client when there are already more than some [smtpserver.conf configurable] number of clients from the same IP address. ... and it is clear that Gabor Kissigs original plan of running smtpserver, and router in tandem would have been terrible disaster at these loads... > I know about -sf, but that seems like policy ("we ignore mail from > such and such domain"), not a DNS query. > > Btw, this patch ought to make policy-builder.sh a little better. ... > The tr handles things like cyber-s-i-q.com vs. cyber-S-I-Q.com, the > grep (untested, I did that by hand) should grep away stuff zmailer > won't digest, "bhaffle,wuyffle.com" for example (note comma). Yeah, qmail allows any junk, while ZMailer preselects :-) Anyway, the "makendbm -p ..." does translate all of the key strings into lowercase, and then complain duplicates, like in your cyber-s-i-q.com case above. Also, the db gets visited at places of interest only, in which case a moderate amount of junk around the corners should not matter. > --Arnt From zmailer-owner@nic.funet.fi Sun Jun 29 09:46:58 1997 Received: from TIMEDOUT@alpha.Xerox.COM (port 33295 [13.1.64.93]) by nic.funet.fi with SMTP id <870-5877>; Sun, 29 Jun 1997 09:45:20 +0300 Received: from Jaws.PARC.Xerox.xns by alpha.xerox.com via XNS id <15382(4)>; Sat, 28 Jun 1997 23:45:07 PDT X-NS-Transport-ID: 0800200AF21F374DD67D Date: Sat, 28 Jun 1997 23:44:39 PDT From: Nicholas_Briggs.PARC@xerox.com Subject: Re: .49, spam In-Reply-to: "mea@nic.funet:fi:Xerox's message of Sat, 28 Jun 1997 18:50:56 PDT" To: mea@nic.funet.FI cc: agulbra@troll.NO, zmailer@nic.funet.FI Message-Id: <97Jun28.234507pdt."15382(4)"@alpha.xerox.com> Return-Path: With 65000 messages a day I can run a router process for every smtpserver on a slow SPARC 5, and verify every From address, and filter based on message-ids also. This is using zmailer 2.2e10. Point to note: (a) I have the sticky bits set on the router and smtpserver, so it's a little quicker to start them, (b) I have a prestoserve disk accelerator in the machine, so 90% of the writes hits the RAM cache and aren't passed on to the disk. Points on the down side -- because the regexp matching doesn't always work correctly, for some of the matching rules I have to fork a "grep" against the address to catch, for example [0-9][0-9]*@(hotmail\.com|mindspring\.com|...). \nick From zmailer-owner@nic.funet.fi Tue Jul 1 16:23:26 1997 Received: from TIMEDOUT@eekholt.NL.net (port 41735 [193.78.241.1]) by nic.funet.fi with ESMTP id <164-8822>; Tue, 1 Jul 1997 16:20:39 +0300 Received: (from alexis@localhost) by eekholt.NL.net (8.8.5/8.8.5) id PAA09390 for zmailer@nic.funet.fi; Tue, 1 Jul 1997 15:20:19 +0200 (MET DST) From: Alexis Yushin Message-Id: <199707011320.PAA09390@eekholt.NL.net> Subject: using zmailer for serious production To: zmailer@nic.funet.fi Date: Tue, 1 Jul 1997 15:20:19 +0200 (MET DST) Reply-To: alexis@NL.net (Alexis Yushin) X-Office-Phone: +31 (20) 495-2795 X-NIC-Handle: AY23 X-RIPE-Handle: AY6-RIPE X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Return-Path: Greetings, I'm following the development of zmailer since the time Rayan was its author. The design is very amazing. As for now I work for a large internet provider in Holland and we relay/handle a couple hundreds of thousands emails a day. This is a very tough task for sendmail so I am really considering switching our major operations to zmailer. I looked into current version of it and it was quite a pleasant seen although neither properly documented nor commented still looks very promising and solid. I would like to hear some success stories/comments/opinions/advices regarding running zmailer in production environment. Needless to say our email operations is almost the most critical part in our business so should it fail a company like ours could be out of business in a couple of days. Sorry for being a bit noisy and thanks for your reactions in advance, alexis -- Yesterday dont matter when its gone From zmailer-owner@nic.funet.fi Tue Jul 1 18:52:54 1997 Received: by nic.funet.fi id <71-8822>; Tue, 1 Jul 1997 18:50:53 +0300 Subject: Re: using zmailer for serious production From: Matti Aarnio To: alexis@NL.net Date: Tue, 1 Jul 1997 18:50:45 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199707011320.PAA09390@eekholt.NL.net> from "Alexis Yushin" at Jul 1, 97 03:20:19 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3708 Message-Id: <19970701155053Z71-8822+273@nic.funet.fi> Return-Path: > Greetings, > > I'm following the development of zmailer since the time > Rayan was its author. The design is very amazing. As for now I > work for a large internet provider in Holland and we relay/handle > a couple hundreds of thousands emails a day. This is a very tough > task for sendmail so I am really considering switching our major > operations to zmailer. I looked into current version of it and it > was quite a pleasant seen although neither properly documented nor > commented still looks very promising and solid. The documentation will improve quite fast. Telecom Finland did hire a company specialized at technical documentation writing to do it, and first draft should be available for public review around 21st-23rd of July. (Before that I have had two iteration rounds.) I am working on all kinds of little problem people report to me to get the current version to stabilize -- right now I have done a few patches to the lattest public version (2.99.49p2), and should propably make them available as "p3" soon. At Telecom Finland we use ZMailer as the firewall frontend in between our internal mailers, and the Internet, and the traffic flows here are not that great -- on average 20-30k messages per day. At nic.funet.fi the daily traffic at the transport is between 140-280k addresses (deliveries and attempts) per day. In there, for 28-30th of June, my quick oneliner egrep/awk/cut/whatnot throwaway script gave daily routing statistics as: msgs=3320 addrs=137991 msgs=1135 addrs= 66862 msgs=3146 addrs=158407 This all while nic.funet.fi is doing its main job of feeding 20-30 GB per day via anonymous FTP to customers. The machine itself is oldish DEC 3000/900 uniprocessor (EV4) Alpha running at 275 MHz. This kind of recipient count/message ratio is abnormal, and is one of the reasons we see occasionally a bit slow delivery thru the system -- delays of 15-30 minutes in the router input queue. (My future main interest will be to get the router faster at its script execution, and data manipulation.) During this month we should do a larger MTA bencmarking run with all kinds of systems (X.400, and non-X.400 ones). Then I can perhaps tell how much ZMailer really can push, and compare it (at least) against sendmail 8.8.x. > I would like to > hear some success stories/comments/opinions/advices regarding > running zmailer in production environment. Needless to say our > email operations is almost the most critical part in our business > so should it fail a company like ours could be out of business in > a couple of days. Sorry for being a bit noisy and thanks for your > reactions in advance, Our (Telecom Finland) ISP service group is using sendmails at the mailbox delivery stage, because the original system was done with them, but that too might change. Occasionally some users have complained that ZMailer is "slow" to deliver their email -- like when it gets rapidly large number of messages, or when its queue is empty, and the router (or scheduler) is in their 20 second queue-recheck delay phase.. Nevertheless, the router can churn happily 200k-800k messages per day per process, and the scheduler can push out even more. Sure you can have faster delivery with sendmail at light load, but try to saturate your system. Sendmail kills it, just like Ethernet can't give full 10/100 Bps for multiple users, and FDDI can. These are just fundamentally different technologies. ... huh, I begin to sound like a marketroid ... > alexis > -- > Yesterday dont matter when its gone /Matti Aarnio (FUNET archive admin) (working at Telecom Finland) From zmailer-owner@nic.funet.fi Tue Jul 1 18:55:52 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 31402 [194.67.3.135]) by nic.funet.fi with ESMTP id <738-20648>; Tue, 1 Jul 1997 18:53:21 +0300 Received: from NO-IDENT-SERVICE@ariel (port 34044 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68046-27020>; Tue, 1 Jul 1997 19:53:03 +0400 Date: Tue, 1 Jul 1997 19:53:00 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: using zmailer for serious production To: Alexis Yushin cc: zmailer@nic.funet.fi In-Reply-To: <199707011320.PAA09390@eekholt.NL.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: > work for a large internet provider in Holland and we relay/handle > a couple hundreds of thousands emails a day. This is a very tough > task for sendmail so I am really considering switching our major > operations to zmailer. I looked into current version of it and it Here, we are running Zmailer for 2+years. Recently, mail traffic is somewhere between 10,000 and 20,000 messages/day. Zmailer runs on an SS20 (together with POP/IMAP from UWashington) and handles this traffic without visible effort. Typical load average is 0.4, and this is mostly due to numerous POP processes. Typical latency when delivering within local network is less then a minute. Eugene From zmailer-owner@nic.funet.fi Tue Jul 1 20:48:00 1997 Received: from NO-IDENT-SERVICE@chronos.sovam.com (port 63402 [194.67.3.135]) by nic.funet.fi with ESMTP id <15452-20648>; Tue, 1 Jul 1997 20:45:02 +0300 Received: from NO-IDENT-SERVICE@ariel (port 34075 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68046-21439>; Tue, 1 Jul 1997 21:44:56 +0400 Date: Tue, 1 Jul 1997 21:44:42 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: policy checking in smtpserver To: zmailer@nic.funet.fi Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: I just started playing with policy in smtpserver. Noticed something that is probably a (minor) bug: 553-chronos - You are on our reject-IP-address -list, GO AWAY! 553 If you feel we mistreat you, do contact us. 553 Ask HELP for our contact information. note that there is no dash in the second line while it should be there. a general note: the syntax is, hmm, interesting. I've read the comments in the smtp-policy.src twice and I still do not fully understand it. Just a dream: to have in zsh style: sift "$fromdomain" in .*\.nodomain\..* return 0 ;; sovam.com return 1 ;; tfis I am not sure if it is practical to compile zshell into smtpserver, but why not? On systems with dynamic loading libsh could be built shared... Eugene From zmailer-owner@nic.funet.fi Wed Jul 2 00:18:59 1997 Received: from root@lupinella.troll.no (port 50046 [195.0.254.19]) by nic.funet.fi with ESMTP id <2203-30738>; Wed, 2 Jul 1997 00:17:50 +0300 Received: by troll.no id <79584-17381>; Tue, 1 Jul 1997 23:17:30 +0200 Sender: agulbra@troll.no To: Matti Aarnio Cc: zmailer@nic.funet.fi Subject: Re: .49, spam References: <19970629015108Z16347-29341+95@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 01 Jul 1997 23:17:27 +0200 In-Reply-To: Matti Aarnio's message of Sun, 29 Jun 1997 04:50:56 +0300 (EET DST) Message-ID: Lines: 31 Return-Path: Matti Aarnio > One needs near instantaneous knowledge of spam sources, or a real > AI system for message content analysis/spam filtering, but ... > (if you can find this latter, I would adapt it into an auto-secretary > for me ;-) I do that in my mailing-list processor. AI it's not, but perl regexps make a decent substitute :) $spamchance += 10 if ( $body =~ /(?:(?:like)|(?:instructions)) to be removed from our.{1,15}list/six ); > Of course we could consider similar approaces to the problem > as virus scanner makers are using. Just scan the message body > for magic identification factors, and bingo... but what will > happen, if somebody forwards a message with matching string, > but with legitimate content ? Like when sending such a mail to > postmaster, or abuse, or ... That's why I haven't transferred my code into zmailer. I wrote: > > The message came from something like <1fvdf@ejfvsd.afhdjalf3.vmx>, and > > I don't see any way to reject things like that. Could smtpserver do > > an MX/A lookup of the envelope sender domain? Would that be a bad > > idea? On closer reflection, I see that it's not likely to help against spam. The spammers will just pick a random valid domain. --Arnt From zmailer-owner@nic.funet.fi Wed Jul 2 00:24:36 1997 Received: from root@lupinella.troll.no (port 60165 [195.0.254.19]) by nic.funet.fi with ESMTP id <17-19265>; Wed, 2 Jul 1997 00:22:54 +0300 Received: by troll.no id <79583-17382>; Tue, 1 Jul 1997 23:22:39 +0200 Sender: agulbra@troll.no To: Matti Aarnio Cc: alexis@NL.net, zmailer@nic.funet.fi Subject: Re: using zmailer for serious production References: <19970701155053Z71-8822+273@nic.funet.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII From: Arnt Gulbrandsen Date: 01 Jul 1997 23:22:35 +0200 In-Reply-To: Matti Aarnio's message of Tue, 1 Jul 1997 18:50:45 +0300 (EET DST) Message-ID: Lines: 30 Return-Path: Matti Aarnio > Occasionally some users have complained that ZMailer is "slow" > to deliver their email -- like when it gets rapidly large > number of messages, or when its queue is empty, and the router > (or scheduler) is in their 20 second queue-recheck delay phase.. > Nevertheless, the router can churn happily 200k-800k messages > per day per process, and the scheduler can push out even more. FWIW, I've seen some slowness with .49 that I'm not used to. Normally, I think zmailer is beatifully fast: Give it a pile of work and it'll sort out what to do PDQ then do it in an effective way. So it takes five, ten, twenty seconds to do local deliver. No problem, as long as it never takes much more, even during the ugly load spikes. But in .49 the router seems to do DNS lookups (tail -f shows the router taking quite some time to process certain odd domains) despite my toplevels setting, so if I have more mailing-list messages than routers, all the routers can block for a while. > Sure you can have faster delivery with sendmail at light load, > but try to saturate your system. Sendmail kills it, just like > Ethernet can't give full 10/100 Bps for multiple users, and > FDDI can. These are just fundamentally different technologies. I once had sendmail stop emptying its queue because of just eleven queued message. I had "* * * * * /usr/sbin/sendmail -q" in cron, but these eleven nastily undeliverable messages still made sendmail choke. --Arnt From zmailer-owner@nic.funet.fi Wed Jul 2 00:37:26 1997 Received: from NO-IDENT-SERVICE@misery.sdf.com (port 40720 [204.244.210.193]) by nic.funet.fi with SMTP id <319-19265>; Wed, 2 Jul 1997 00:36:13 +0300 Received: from tom by misery.sdf.com with smtp (Exim 1.62 #1) id 0wjAVf-0000as-00; Tue, 1 Jul 1997 14:30:51 -0700 Date: Tue, 1 Jul 1997 14:30:50 -0700 (PDT) From: Tom Samplonius To: Arnt Gulbrandsen cc: Matti Aarnio , alexis@NL.net, zmailer@nic.funet.fi Subject: Re: using zmailer for serious production In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: On 1 Jul 1997, Arnt Gulbrandsen wrote: > I once had sendmail stop emptying its queue because of just eleven > queued message. I had "* * * * * /usr/sbin/sendmail -q" in cron, but > these eleven nastily undeliverable messages still made sendmail choke. That cron line will cause bad things to happen. It will launch a queueing running sendmail every minute. As stated in the documentation, the correct way to keep the queue empty is by running sendmail as "sendmail -bd -q10m", which run the queue every 10 minutes. Remember that in sendmail, messages only go into the queue, if initial delivery fails. > --Arnt > > Tom From zmailer-owner@nic.funet.fi Wed Jul 2 11:40:16 1997 Received: from TIMEDOUT@eekholt.NL.net (port 7946 [193.78.241.1]) by nic.funet.fi with ESMTP id <15646-30738>; Wed, 2 Jul 1997 11:39:28 +0300 Received: (from alexis@localhost) by eekholt.NL.net (8.8.5/8.8.5) id KAA02403 for zmailer@nic.funet.fi; Wed, 2 Jul 1997 10:39:04 +0200 (MET DST) From: Alexis Yushin Message-Id: <199707020839.KAA02403@eekholt.NL.net> Subject: a) sender rewriting b) separate queues/manual queue runs To: zmailer@nic.funet.fi Date: Wed, 2 Jul 1997 10:39:03 +0200 (MET DST) Reply-To: alexis@NL.net (Alexis Yushin) X-Office-Phone: +31 (20) 495-2795 X-NIC-Handle: AY23 X-RIPE-Handle: AY6-RIPE X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Return-Path: Morning (or whatever), Sorry for not solving this myself but I need to give Zmailer a try very quickly in order to decide whether to go on or not. Here are several questions, I couldnt find obvious answer for: I need to do sender rewriting, like I have a lot of virtual domains for one machine and when users send their mail outside of it the sender information has to point back to the virtual domain associated with this sender. So we have relation between accounts names and fqdn addresses, the question is what part of configuration I have to do the rewriting in. I tried rrouter and crossbar but that gave me no result. On other hand I am able to rewrite all the addresses (sender AND recipient) in internet (rewriting procedure) but cannot tell whether I rewrite sender or recepient address there. Any advice for the newbie? :-) Next, what I need to do is to hold all the mail for certain customers until they establish connection and request to run the queue. I realized I can hold the queue with the hold command in scheduler.conf but how do I initiate queue run for certain domain manually? ETRN is not a solution for all of the customers we have. I feel this posting is really inappropriate here but I cannot think of any other place to ask these questions at the moment so thanks in advance, alexis -- Yesterday dont matter when its gone From zmailer-owner@nic.funet.fi Wed Jul 2 14:12:00 1997 Received: from TIMEDOUT@eekholt.NL.net (port 34063 [193.78.241.1]) by nic.funet.fi with ESMTP id <15393-19265>; Wed, 2 Jul 1997 14:10:41 +0300 Received: (from alexis@localhost) by eekholt.NL.net (8.8.5/8.8.5) id NAA09913 for zmailer@nic.funet.fi; Wed, 2 Jul 1997 13:10:31 +0200 (MET DST) From: Alexis Yushin Message-Id: <199707021110.NAA09913@eekholt.NL.net> Subject: bug in ordered To: zmailer@nic.funet.fi Date: Wed, 2 Jul 1997 13:10:30 +0200 (MET DST) Reply-To: alexis@NL.net (Alexis Yushin) X-Office-Phone: +31 (20) 495-2795 X-NIC-Handle: AY23 X-RIPE-Handle: AY6-RIPE X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Return-Path: Greetings, There is a minor bug in ordered.c: *** ordered.c.orig Wed Jul 2 13:09:09 1997 --- ordered.c Wed Jul 2 13:09:47 1997 *************** *** 130,136 **** if (*buf == '\n') ++linecnt; buf = fm->membuf; ! fm->offsets = (long*)emalloc(sizeof(long)*(linecnt+1)); linecnt = 0; while (buf < eof) { fm->offsets[linecnt] = (buf - (const char*)fm->membuf); --- 130,136 ---- if (*buf == '\n') ++linecnt; buf = fm->membuf; ! fm->offsets = (off_t*)emalloc(sizeof(off_t)*(linecnt+1)); linecnt = 0; while (buf < eof) { fm->offsets[linecnt] = (buf - (const char*)fm->membuf); alexis PS I solved that things with sender rewriting -- Yesterday dont matter when its gone From zmailer-owner@nic.funet.fi Wed Jul 2 14:20:29 1997 Received: by nic.funet.fi id <2933-19265>; Wed, 2 Jul 1997 14:19:07 +0300 Subject: Re: a) sender rewriting b) separate queues/manual queue runs From: Matti Aarnio To: alexis@NL.net Date: Wed, 2 Jul 1997 14:19:03 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199707020839.KAA02403@eekholt.NL.net> from "Alexis Yushin" at Jul 2, 97 10:39:03 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 3963 Message-Id: <19970702111907Z2933-19265+305@nic.funet.fi> Return-Path: > Morning (or whatever), > > Sorry for not solving this myself but I need to give Zmailer > a try very quickly in order to decide whether to go on or not. Here > are several questions, I couldnt find obvious answer for: > > I need to do sender rewriting, like I have a lot of virtual > domains for one machine and when users send their mail outside of > it the sender information has to point back to the virtual domain > associated with this sender. So we have relation between accounts > names and fqdn addresses, the question is what part of configuration > I have to do the rewriting in. I tried rrouter and crossbar but that > gave me no result. On other hand I am able to rewrite all the > addresses (sender AND recipient) in internet (rewriting procedure) > but cannot tell whether I rewrite sender or recepient address > there. Any advice for the newbie? :-) You want to rewrite the VISIBLE headers ? Not only the transport envelope ? For the rewrite functions there should be additional parameter for the call. ... and actually there even is. I made a small change to the script at crossbar.cf, just by adding second parameter. The rest I leave for you. (Same can be done with all rewrite functions.) # This is usually the default message-header address rewriting function. # It is responsible for hostname hiding and qualification. internet (address, hname) { # 'address' is the address we are rewriting # 'hname' is the header whose address(es) we are rewriting; # value here is usually: 'From', 'To', etc. #echo "rewrite: internet: $hname $address" >> /tmp/hdr.log address="$(canonicalize "$address")" # Canonicalize does local # hostname hiding... sift "$address" in (.*)<@(.+)>(.*) #if [ $(deliver \2) ]; then # hostname hiding # address="\1@${mydomain}\3" # break #fi address="\1@\2\3" # No hostname hiding... ;; (.*)<(.+)>(.*) address="\1\2\3" ;; # defocus [^@]+ # This is a local part address w/o any domains! address="$(condquote "$address")" address="$address@$mydomain" # add our hostname ;; tfis return "$address" } # end of internet > Next, what I need to do is to hold all the mail for certain > customers until they establish connection and request to run the > queue. I realized I can hold the queue with the hold command in > scheduler.conf but how do I initiate queue run for certain domain > manually? ETRN is not a solution for all of the customers we have. Don't use the hold channel, unless you absolutely have to. There are no good uses for it aside of recovering from system malfunctions. The way the ETRN works is to send a special message (file) to the scheduler so that it triggers the queue run for the given target domain. Right now it supports only queue starts with exactly matching hostnames. That is, you can not do: ETRN *.foo.nl.net to start queues to hosts: bar.foo.nl.net fii.foo.nl.net etc.. all at once. You have to tell all target host names. For the scheduler I would put special selectors for the long delay customers, like: smtp/*.foo.nl.net interval=1m retries="10 30 300" expiry=14d ... That way, it will at first try immediately, then after 10 minutes, then 30 minutes after that, and finally every 300 minutes. Having a thousand messages there arriving every few minutes will start the timer over and over again, but in quiscent state it will quickly back off, and in any case it needs only resources of ONE process when it actively attempts sends. If you do not want this "instant retry at new arrival", perhaps I can add an option to control it. > I feel this posting is really inappropriate here but I > cannot think of any other place to ask these questions at the > moment so thanks in advance, On the contrary, this is (IMO) exactly what this list is for. > alexis > -- > Yesterday dont matter when its gone /Matti Aarnio From zmailer-owner@nic.funet.fi Wed Jul 2 14:28:10 1997 Received: from TIMEDOUT@eekholt.NL.net (port 53775 [193.78.241.1]) by nic.funet.fi with ESMTP id <15891-30738>; Wed, 2 Jul 1997 14:26:24 +0300 Received: (from alexis@localhost) by eekholt.NL.net (8.8.5/8.8.5) id NAA10445; Wed, 2 Jul 1997 13:26:13 +0200 (MET DST) From: Alexis Yushin Message-Id: <199707021126.NAA10445@eekholt.NL.net> Subject: Re: a) sender rewriting b) separate queues/manual queue runs To: mea@nic.funet.fi (Matti Aarnio) Date: Wed, 2 Jul 1997 13:26:13 +0200 (MET DST) Cc: alexis@NL.net, zmailer@nic.funet.fi In-Reply-To: <19970702111907Z2933-19265+305@nic.funet.fi> from "Matti Aarnio" at Jul 2, 97 02:19:03 pm Reply-To: alexis@NL.net (Alexis Yushin) X-Office-Phone: +31 (20) 495-2795 X-NIC-Handle: AY23 X-RIPE-Handle: AY6-RIPE X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Return-Path: Once Matti Aarnio wrote: > For the rewrite functions there should be additional parameter > for the call. ... and actually there even is. > > I made a small change to the script at crossbar.cf, just by > adding second parameter. The rest I leave for you. > (Same can be done with all rewrite functions.) > ># This is usually the default message-header address rewriting function. ># It is responsible for hostname hiding and qualification. > >internet (address, hname) { > # 'address' is the address we are rewriting > # 'hname' is the header whose address(es) we are rewriting; > # value here is usually: 'From', 'To', etc. > > #echo "rewrite: internet: $hname $address" >> /tmp/hdr.log Hmm, actually I've done it different way: $(sender) && { tmp=$(masquerades $address) && address=$tmp } > That way, it will at first try immediately, then after 10 > minutes, then 30 minutes after that, and finally every 300 > minutes. This helps. Thank you! alexis -- Yesterday dont matter when its gone From zmailer-owner@nic.funet.fi Wed Jul 2 14:30:15 1997 Received: by nic.funet.fi id <15404-30738>; Wed, 2 Jul 1997 14:28:10 +0300 Subject: Re: bug in ordered From: Matti Aarnio To: alexis@NL.net Date: Wed, 2 Jul 1997 14:27:58 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: <199707021110.NAA09913@eekholt.NL.net> from "Alexis Yushin" at Jul 2, 97 01:10:30 pm MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 326 Message-Id: <19970702112810Z15404-30738+288@nic.funet.fi> Return-Path: > Greetings, > > There is a minor bug in ordered.c: Right. You have a system with 64 bit off_t, and 32 bit long ? At my systems I have always: sizeof(off_t) <= sizeof(long) > alexis > PS I solved that things with sender rewriting > -- > Yesterday dont matter when its gone /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jul 3 10:51:02 1997 Received: from root@cam053212.student.utwente.nl (port 323 [130.89.226.142]) by nic.funet.fi with ESMTP id <2370-23894>; Thu, 3 Jul 1997 10:49:21 +0300 Received: by cam053212.student.utwente.nl id <51358-502>; Thu, 3 Jul 1997 09:45:23 +0200 Date: Thu, 3 Jul 1997 09:45:20 +0200 (CEST) From: Remco van de Meent X-Sender: remco@cam053212.student.utwente.nl To: zmailer@nic.funet.fi Subject: bulkmail Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-Path: Hey When handling relatively large mailinglists (say over a couple of hundred subscribers) with sendmail, you can get big improvements regarding speed, by supplying sendmail with a sorted list (using bulkmailer or something like that). My question: will this help with ZMailer too? When delivering to those large lists, I see that the local deliveries are very very earlier than the remote deliveries. Maybe that's caused by ZMailer sorting the list? Thanks for your information, Remco. -- // Remco van de Meent // email: remco@oloon.student.utwente.nl // www: http://oloon.student.utwente.nl // " Never make any mistaeks. " From zmailer-owner@nic.funet.fi Thu Jul 3 11:05:55 1997 Received: by nic.funet.fi id <15668-9954>; Thu, 3 Jul 1997 11:04:30 +0300 Subject: Re: bulkmail From: Matti Aarnio To: remco@oloon.student.utwente.nl (Remco van de Meent) Date: Thu, 3 Jul 1997 11:04:24 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Remco van de Meent" at Jul 3, 97 09:45:20 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1343 Message-Id: <19970703080430Z15668-9954+259@nic.funet.fi> Return-Path: > Hey > > > When handling relatively large mailinglists (say over a couple of hundred > subscribers) with sendmail, you can get big improvements regarding speed, > by supplying sendmail with a sorted list (using bulkmailer or something > like that). > My question: will this help with ZMailer too? When delivering to those large > lists, I see that the local deliveries are very very earlier than the remote > deliveries. Maybe that's caused by ZMailer sorting the list? For the routeing phase ? No. It COULD make sense to do two things at the same time: - pre-sort the recipient domains - send messages in 50-100 recipient batches, and make sure that multiple recipients for any host will not get split between multiple messages. That way the router gets smaller memory footprints for its processing of things, and it will be faster. The reason you see this vast delivery speed difference is that the scheduler is essentially parallel processing system. When a message reaches scheduler, ALL recipient addresses in it are analyzed for running transport agents in parallel. > Thanks for your information, > > Remco. > > -- > // Remco van de Meent > // email: remco@oloon.student.utwente.nl > // www: http://oloon.student.utwente.nl > // " Never make any mistaeks. " /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jul 3 12:59:27 1997 Received: by nic.funet.fi id <2881-9954>; Thu, 3 Jul 1997 12:57:26 +0300 Subject: smtp-server, and toplevel domain verification.. From: Matti Aarnio To: zmailer@nic.funet.fi Date: Thu, 3 Jul 1997 12:57:21 +0300 (EET DST) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 939 Message-Id: <19970703095726Z2881-9954+292@nic.funet.fi> Return-Path: Hello, I saw recently somebody sending mail thru smtp-server with following (quite mad) address: This prompted me to go thru smtpserver's rfc-821 scanner, and alter the treatment of the local part ("plspgme!!!!") to consider non-quoted bangs, and percents as specials. Now the server treas this address as: mail from: 501-5.5.2 ^ 501 5.5.2 Path data: After a '!', a non alphanumeric element Should similar scanners be added also into "sendmail" program for analysis of envelope addresses ? This did also prompt me to think of adding a quick first line defence of verifying that the address has valid top-level domain at the extreme right, or that it is one of the local system names. The question is, does this make sense ? Does it help ? Should I implement that test even without policytest facility being active ? /Matti Aarnio From zmailer-owner@nic.funet.fi Thu Jul 3 13:19:44 1997 Received: from TIMEDOUT@eekholt.NL.net (port 17679 [193.78.241.1]) by nic.funet.fi with ESMTP id <16139-9954>; Thu, 3 Jul 1997 13:16:58 +0300 Received: (from alexis@localhost) by eekholt.NL.net (8.8.5/8.8.5) id MAA18317 for zmailer@nic.funet.fi; Thu, 3 Jul 1997 12:16:46 +0200 (MET DST) From: Alexis Yushin Message-Id: <199707031016.MAA18317@eekholt.NL.net> Subject: neat configuration? To: zmailer@nic.funet.fi Date: Thu, 3 Jul 1997 12:16:46 +0200 (MET DST) Reply-To: alexis@NL.net (Alexis Yushin) X-Office-Phone: +31 (20) 495-2795 X-NIC-Handle: AY23 X-RIPE-Handle: AY6-RIPE X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 8bit Return-Path: Greetings, It looks like the present sample/proto configuration of ZMailer is somewhat a mess, I mean its not what you would call a prototype. Does anyone work on making this more organized? Perhaps some more examples for different cases should be available and a place where people could contribute their configuration solutions would be desired. Please let me know because I would like to contribute some efforts into this. Regards, alexis -- Yesterday dont matter when its gone From zmailer-owner@nic.funet.fi Fri Jul 4 00:02:50 1997 Received: by nic.funet.fi id <2212-5117>; Thu, 3 Jul 1997 23:59:54 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49 patch3 -- only full dump Message-Id: <19970703205954Z2212-5117+32@nic.funet.fi> Date: Thu, 3 Jul 1997 23:58:53 +0300 Return-Path: Hello, I have finally made 2.99.49 patch3 available. It has a few new files in it; utils/Makefile, for example, and more importantly, a couple performance and security fixes. Security by nulling routers/schedulers starters supplementary groups list. ... plus a couple performance things that stem from experiments at vger.rutgers.edu, and reading one SunSoft book about Solaris performance tuning. This source dump has been at the ftp.funet.fi for a few hours, but it took me a while before I got around to test install at other than Linux systems, and of course that first dump had an error in "utils/Makefile.in".. (An extra "-c" at install.) /Matti Aarnio Sun Jun 29 21:36:43 1997 Matti Aarnio * Makefile.in: Version 2.99.49p3 * configure.in, config.h.in, transports/libta/ctlopen.c, transports/mailbox/mailbox.c, transports/sm/sm.c, transports/smtp/smtp.c: Configuration option: --with-ta-mmap which uses mmap() facility to map in pages of the message data. At least at Solaris 2.x it is not advisable to use if you want to have highest possible performance, because the munmap() throws away ALL pages of the file from the system memory, even when other users have those pages in use! For that reason it is much better to rely on plain simple read()/write() behaviour that let buffer cache to exist. * configure.in: Configuration option: --without-fsync Option that turns off (globally) the use fsync() anywhere in the ZMailer. System configurers can decide what is best for their systems. It does give definite edge at buffered IO by not forceing every file modification to be commited up to the disk. * smtpserver/rfc821scn.c: Alteration at the treatment of "!" and "%" characters. Now they are special, and are analyzed in more profound manner within localpart analysis. * smtpserver/smtpchild.c: Register, and analyze properly how many SMTP sessions are currently active (non-reaped) from any given source IP address. * transports/smtp/smtp.c: Correction into notary-status-code pickup for ENHANCEDSTATUSCODES mode. * libsh/testeval.c: (+ several of proto/cf/*.cf files) New operators into the builtin 'test' function: -nt, -ot, -ef (file age comparators, and file equality comparator) * SiteConfig.in, router/rtsyslog.c, smtpserver/smtpserver.c, transports/libta/tasyslog.c: New ZENV entry: SYSLOGFLG= which can be used to turn off syslog() logging activity at the smtp-server/router/transport-agents * config.h.in, configure.in, router/router.c, scheduler/scheduler.c: Check of setgroups() routine, and at both the router, and the scheduler, use both to clean out the set of supplementary groups the processes have. * include/hostenv.h: Concise USE_ALLOCA/HAVE_ALLOCA defines * libident/identuser.c: Be a bit more tolerant at what comes back as response for the ident query. * libsh/interpret.c, router/functions.c: A bit more smarts into quote containing string processing; Especially when dequoteing string that contains only one char: ' or " ... * proto/cf/i-routes.cf: Accept incore database of form: $MAILVAR/db/routes.zmsh * router/libdb/headers.c: Allow "Reply-To:" and "Resent-Reply-To:" to have value: "<>" (sigh...) * scheduler/msgerror.c, transports/errormail/errormail.c, transports/fuzzyalias/fuzzyalias.c, transports/mailbox/mailbox.c: Stronger MIME-report boundary generator which isn't merely stocastic, but very likely unique at the creation time. * smtpserver/policytest.c, smtpserver/policytest.h: Move policy-database open out from database define routine. * smtpserver/policytest.c: - Study "rejectsource" at HELO/EHLO parameter check - For the RCPT TO:<..> study also the full addresses in addition to the domain part * smtpserver/smtpcmds.c, smtpserver/smtpserver.c: - Print out the incoming HELO/EHLO parameter as is into comment at "rcvdfrom" envelope header along with many other bits of info. - Change policy-rejection responses from: 553 5.1.8 ... to: 553 5.7.1 ... * ssl/ssl.c: Conflicting routine name within the ssl.c program: basename() ( Linux libc6 a.k.a. glibc-2.0.4 .. ) * support/vacation/vacation.sh.in: Modifications into 'vacation start' command behaviour. * transports/mailbox/mailbox.c: Found a couple off odd situations where creation of files into non-kosher location was allowed; partly corrected with the purge of supplementary groups at the scheduler, partly with a bit of new code. * transports/smtp/smtp.c: Recognize (unknown) smtp response value 571 to be a sign of policy based refusal. * utils/makendbm/ndbmlook.c: At Debian systems including causes surprises at parameters called 'errno' ... rename such parameters. From zmailer-owner@nic.funet.fi Fri Jul 4 14:19:01 1997 Received: from chronos.sovam.com ([194.67.3.135] EHLO chronos ident: NO-IDENT-SERVICE [port 37068]) by nic.funet.fi with ESMTP id <15633-31015>; Fri, 4 Jul 1997 14:17:24 +0300 Received: from NO-IDENT-SERVICE@ariel (port 34982 [194.67.3.216]) by chronos.sovam.com with ESMTP id <68046-21438>; Fri, 4 Jul 1997 15:17:11 +0400 Date: Fri, 4 Jul 1997 15:17:03 +0400 (MSD) From: Eugene Crosser Sender: crosser@ariel.sovam.com Reply-To: Eugene Crosser Subject: Re: 2.99.49 patch3 -- only full dump To: Matti Aarnio cc: zmailer@nic.funet.fi In-Reply-To: <19970703205954Z2212-5117+32@nic.funet.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Return-Path: X-Orcpt: rfc822;zmailer-log > * router/libdb/headers.c: > Allow "Reply-To:" and "Resent-Reply-To:" to have value: "<>" > (sigh...) Matti, I beleive that this form is *not* legal by RFC822. I examined this problem quite thoroughly when I needed to direct replies to nowhere. "Reply-To: <>" is illegal (while "Reply-To: Any comment :;" is legal). (Although, stupid sendmail will corrupt both forms beyond recognition :-( ) Eugene From zmailer-owner@nic.funet.fi Fri Jul 4 14:51:34 1997 Received: by nic.funet.fi id <15480-12174>; Fri, 4 Jul 1997 14:49:58 +0300 Subject: Re: 2.99.49 patch3 -- only full dump From: Matti Aarnio To: mark@cal026031.student.utwente.nl Date: Fri, 4 Jul 1997 14:49:56 +0300 (EET DST) Cc: zmailer@nic.funet.fi In-Reply-To: from "Mark Visser" at Jul 4, 97 11:39:31 am MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 896 Message-Id: <19970704114958Z15480-12174+33@nic.funet.fi> Return-Path: X-Orcpt: rfc822;zmailer-log > On Thu, 3 Jul 1997, Matti Aarnio wrote: > > > Hello, > > > > I have finally made 2.99.49 patch3 available. > > Matti, > > Could you also make a zmailer-2.99.49.patch3 available ? > I know a lot of people patching 49 with patch1 and then patch2... > > I only saw a 'full distri', or did i miss something ? The patch is a bit painfull to make, as there are "some" new files in it. I will give it a try. (5 minutes, and manual editing of the diff..) ... Huh! 700kB! Ok, test of the patch is at: ftp://mea.tmt.tele.fi/zmailer-2.99.49.patch3.gz GZIPed version: circa 150 kB > Thanks, > > Mark > Mark Visser | Student Civiele Technologie & Management > Calslaan 26 - 31 | E-mail: mark@cal026031.student.utwente.nl > 7522 MC Enschede | SNT-mail: mark@snt.student.utwente.nl > Telephone: 053-4895038 | /Matti Aarnio From zmailer-owner@nic.funet.fi Fri Jul 4 21:50:30 1997 Received: by nic.funet.fi id <15844-12174>; Fri, 4 Jul 1997 21:49:42 +0300 From: Matti Aarnio To: zmailer@nic.funet.fi Subject: 2.99.49 patch4 Message-Id: <19970704184942Z15844-12174+190@nic.funet.fi> Date: Fri, 4 Jul 1997 21:49:41 +0300 Return-Path: X-Orcpt: rfc822;zmailer-log Well... "when it rains, it pours.." I have been listening at people, and watching things around myself. Here are the results for today.. Even my collegues surprised me when one of them wrote a piece of code for smtp testing, and got the "DATA" command response a bit wrong.. She said: "But sendmail was happy with it!" The worst things I found is the bug at 'mailbox'. Sometimes smallish messages became scrambled in weird way. /Matti Aarnio Fri Jul 4 14:54:02 1997 Matti Aarnio * Makefile.in Version 2.99.49p4 * router/rfc822.c: Nuke headers: "Bcc:", "Return-Path:", "X-Orcpt:", "X-Envid:", and respective "Resent-*:" variants * router/libdb/bsdbtree.c: If the db can not be opened ( open_btree() returns NULL ), don't crash on it.. * scheduler/msgerror.c: In case the report we are composing does have "NOTIFY=NEVER" in effect, but no valid recipients exist, ignore any such NOTIFY value, and return the message in full -- to the system postmaster... * transports/mailbox/mailbox.c: Found a bug in let_buffer[] reuse -- it was an "automatic" variable, that is, stack reclaimed it in between message deliveries, and of course were able to corrupt it! I also enlarged the let_buffer[] size to be of reasonable size -- 16k to 64k depending on your BUFSIZ value. * transports/mailbox/mailbox.c: Add headers "X-Orcpt:" and "X-Envid:" to the message when storing it into a file. For a pipe the data is passed in envelope variables. (I got rather fed up with spams this morning, and decided to add a bit more trace as to whom the message has been sent originally..) * transports/smtp/smtp.c: Accept non rfc-821 behaviour of the remote server by treating non-conformant reply of form: "NNN" the same as conformant reply of form: "NNN". From zmailer-owner@nic.funet.fi Sun Jul 6 02:50:04 1997 Received: from cesium.clock.org ([140.174.97.8] EHLO cesium.clock.org ident: NO-IDENT-SERVICE [port 16911]) by nic.funet.fi with ESMTP id <16021-20287>; Sun, 6 Jul 1997 02:49:34 +0300 Received: from smd@localhost by cesium.clock.org id <77189-219>; Sat, 5 Jul 1997 16:49:26 -0700 To: Tom Samplonius Cc: Arnt Gulbrandsen , Matti Aarnio , alexis@NL.net, zmailer@nic.funet.fi Subject: Re: using zmailer for serious production References: Sender: "Sean M. Doran" From: smd@clock.org (Sean M. Doran) Date: 05 Jul 1997 19:49:20 -0400 In-Reply-To: Tom Samplonius's message of "Tue, 1 Jul 1997 17:30:50 -0400" Message-ID: Lines: 40 X-Mailer: Gnus v5.4.61/Emacs 19.34 Return-Path: X-Orcpt: rfc822;zmailer-log Tom Samplonius writes: > That cron line will cause bad things to happen. It will launch a > queueing running sendmail every minute. I think that was the point. > As stated in the documentation, the correct way to keep the queue empty > is by running sendmail as "sendmail -bd -q10m", which run the queue every > 10 minutes. The problem is that sendfail(8) tends to block on things, and when blocked, -q