Subj : Re: Routing Problem
To   : Dale Shipp
From : Mvan Le
Date : Wed Jan 09 2008 06:46 am

 -=> On 01-08-08  05:58,  Mvan Le <=-
 -=> spoke to Dale Shipp about Re: Routing Problem <=-

 DS>   I am ready to give up and resign myself to manually routing each
 DS>   netmail -- there really are not many.

 ML> Ok.

 ML> Do this, 

 ML> Route Hold 1:123/500 World
 ML> #End Route.Cfg

 DS>   We are getting some where!   Your suggestion resulted in a netmail
 DS>   addressed to you getting packed out addressed to (presumably routed
 DS>   through) 1:123/500, but because of the HOLD flag, it did not get sent
 DS>   -- even when I polled 1:123/500.

That shouldn't happen. When you polled 1:123/500, your mailer should have 
looked for a file named 007B01F4.HLO and renamed it to *.flo and send it off to 
1:123/500, or create a *.ilo file in the outbound directory to trigger the 
poll.

Maybe your uplink is not advertising 1:123/500 as an alias ... (?) Check your 
mailer's session log.
             
 DS>   I tried changing that to
 DS>   Route Normal 1:123/500 World
 DS>   and the result was that Argus sent the mail on its way.   Hopefully,
 DS>   the BBS you are using will deliver it to you by the time you read this
 DS>   message.

Dunno. I suggested Hold instead of Normal, to allow time for you to inspect 
your outbound directory. I also use Hold myself on my system because I observe 
ZMH (04:00 AEST) :)

 DS>   The question that now wonders me is why did
 DS>   Route Normal 1:123/500 World
 DS>   work when the previous command of
 DS>   Route Normal 1:123/500 1:All 2:All 3:All 4:All 5:All 6:All
 DS>   not work?

Well, the #:All syntax should work.

My guess is your Route.Cfg was set up like this, 

   1: Send  Normal World
   2: Route normal 1:123/500 1:All 2:All 3:All 4:All 5:All 6:ALL
   3: Send  Crash  1:140/1
   4: Send  Crash  1:261/1
   5: Send  Crash  1:261/38
   6: Send  normal 1:123/500

The problem is that the first line scans for all normal flavoured, uncompressed 
mail from any host, and then processes the mail according to your specified 
rule (ie. compresses them as Normal).

When the second line runs, it finds no mail to process because Squish looks for 
Normal *Uncompressed* mail to process. Since all mail has been compressed by 
the first rule, nothing beneath it will be processed properly.

Refer to "The Route Command" section of Squish.doc, 

  <nodes> is the optional list of network addresses for which
  Squish should readdress mail.  Squish will look for normal-
  flavoured, uncompressed packets for these nodes, and then route
  them through the specified target.

Try reordering your route.cfg to run the Route command first. 

As better practice, apply specific node rules before sweeping nodes with 
widlcards; like how you would set up a DENY policy firewall Eg.,

   1: Send  Crash  1:140/1
   2: Send  Crash  1:261/38
   3: Send  Crash  1:261/1
   4: Send  normal 1:123/500
   5: Route normal 1:123/500 1:All 2:All 3:All 4:All 5:All 6:ALL
   6: Send  Normal World

albeit the last line will do nothing unless you're catching mail from othernets 
(ie. outside zones 1-6). 

 DS>   Thanks much for your continued patience and perserverance.  I think
 DS>   that you have managed to solve the problem, although as I said above,
 DS>   I don't understand the difference and why what I had before did not
 DS>   work.

When you attach a crash flag to a message, it no longer is a Normal flavoured 
mail so is not processed by Squish hence typically handled by the mailer. So 
only give priviledged users Crash mail access on your BBS.

--- Maximus/2 3.01
 * Origin: Top Hat 2 BBS (1:343/41)