On Wed, Aug 12, 1998 at 02:42:25PM +0100, Philip Hazel wrote:
> > This is where the problem lies: for a message forwarded from user1's
> > .forward file, the $original_domain will still be a.com, and so this message
> > fails to get handled by the filter for b.com. Other than entering a
> > rewriting rule in the main config file, does anyone have a solution for
> > this problem?
>
> Why do you need $original_domain? Why not just $domain?
Because $original_domain will have a.com whereas $domain will have b.com
I'm filtering based on what the original domain was. Basically, the way I'm
handling virtual domains here is that all mail for a specific domain is
handed off to a local user, eg *@anand.org is handled by local user arb.
This generates addresses of the form:
arb-info@??? for info@???
The .forward file for arb then takes care of stuffing *all* the messages
into arb's Maildir, and then when arb connects, I use serialmail to send
off all this mail to their mail server (store-and-forward). I'm short on IP
addresses, so ETRN (which needs a static IP for each customer) is not
workable. Serialmail works happily under dynamic IP's.
Now, for some domains, I have multiple Maildirs in the local user's home
directory, which each handle a sub domain, eg:
arb-info@??? for info@???
arb-info@??? for info@???
The only way I can distinguish between these 2 "identical" recipients is by
looking at the original_domain in the filter, and then deliver mail into
either Maildir-one or Maildir-two
So going back to my original problem, while this scenario works quite well
most times, it fails when a mail has come via a .forward file whose
original domain is not the same as the virtual domain.
> > PS. Philip: exim 2.02 on my server crashed once last week with signal 11 -
> > still seems there might be some bugs in it. It's doing well otherwise.
>
> Sorry to hear that, but of course I can't do much about it without more
> details.
OK. I don't have much to report. As before, there was a core file in exim's
spool directory. I ran gdb, and it reported that exim had exited on signal
11. This was happening previously with local deliveries, so I went back to
the log file at the time the core was dumped. All local deliveries at that
time had taken place. So this must be something else.
Regards,
--
Anand
"I haven't lost my mind; I know exactly where I left it."
--
*** Exim information can be found at
http://www.exim.org/ ***
