Roland Dworschak <[EMAIL PROTECTED]> wrote:
>
> cct.at mail exchanger = 10 mail.cct.at.
> cct.at mail exchanger = 20 mail.defense.at.
>
> mail.defense.at has 'cct.at' in the /var/qmail/control/rcpthosts so it
> relays the mails directed to @cct.at to the mail.cct.at server when it
> gets back up. However, when cct.at is down and I try to send a mail to
> [EMAIL PROTECTED], mails get redirected to mail.defense.at (2nd mx)
They're not "redirected". They just happen to be going to a
lower-priority MX because the highest-priority one is unavailable.
> and then tries to send the mail to the 1st mx (mail.cct.at). Because
> the 1st mx isn't avaiable it gets delivered to the 2nd mx
> (mail.defense.at) and so it tries to resend the same message to itself
> every second and stops with 'too many hops':
This shouldn't happen. When qmail on mail.defense.at tries to send the
mail to cct.at and fails, it should automatically be able to determine
that the next-highest priority MX is itself, and therefore not attempt
to deliver to it.
Unless, of course, you've created a mail loop using smtproutes. Show us
the contents of /var/qmail/control/smtproutes. You've probably
configured it in a loop.
> Any ideas what's wrong here?
Show us the qmail-send logs of these deliveries occurring as well.
Charles
--
-----------------------------------------------------------------------
Charles Cazabon <[EMAIL PROTECTED]>
GPL'ed software available at: http://www.qcc.sk.ca/~charlesc/software/
-----------------------------------------------------------------------
