On Tue, Jul 11, 2000 at 11:20:37AM -0400, Paul Farber wrote:
> I grepped the maillog and no exiting was found.
Then I'm not certain qmail-send actually exited.
> As for tcpserver logs, I don't log that info to speed up the server.... I
> only get qmail messages.
Fair enough, that's your choice. Though I would suggest looking at
multilog (from daemontools). I find it very efficient.
> I'm not sure about the triggers, but no one was logged in (that I could
> find) and the only thing that I saw abnormal was the remote concurrency go
> from 0-96 (first due to CNAME failures), then remote slowly declined to 0,
> then it ramped back up to 96 with what appeared as remotes going through,
> then it trickled down to 0, then back up to 96 with more remote
> deliveries.
I think it's quite likely you're seeing qmail merely retrying messages
it's had to defer because of DNS problems. Are the 96 messages destined
for addresses within the same domain? Or perhaps in domains served by
the same nameservers?
The scenario could have been something like:
1. qmail receives ~90 messages to be sent to domain X.
2. Domain X has dodgy DNS, messages all get deferred.
3. qmail waits a bit, and tries (all) the deferred messages again.
4. Repeat 2 and 3 until DNS problems get sorted.
> I never see concurrency remotes that high.. generally 20 is a big number..
> that's why I think that has *something* to do with it.
DNS problems, especially those that manifest themselves as time-outs,
will lengthen how long a qmail-remote runs for each delivery
attempt. When they run for longer more of them are likely to be
running concurrently.
Regards,
james
--
James Raftery (JBR54) - Programmer Hostmaster - IE TLD Hostmaster
IE Domain Registry - www.domainregistry.ie - (+353 1) 706 2375
"Managing 4000 customer domains with BIND has been a lot like
herding cats." - Mike Batchelor, on [EMAIL PROTECTED]
