On Monday 18 of May 2009, Rafał Kupka wrote:
> On Sun, May 17, 2009 at 02:49:50PM +0200, Arkadiusz Miskiewicz wrote:
> Hello,
>
> > exim 4.69 + few patches from cvs (buffer overflow fixes etc).
> >
> > It's configured to be run in queue only mode + queues are run once per
> > 15s.
> >
> > Unfortunately after some time queue is no longer run, new mails are
> > accepted but not delivered.
> >
> > Restarting exim fixes the problem for a while and after some time (few
> > minutes-few days) the problem appears again.
> >
> > Any ideas on how to debug why queues are no longer run?
>
> Maybe it's problem with signal delivery. Exim uses alarm(2) to run queue
> runners.
>
> Try kill -SIGALRM <pid_of_exim>. It should spawn queue runner from main
> exim process.
>
> What's exim signal masks (ps s -p <exim_pid>)?
UID PID PENDING BLOCKED IGNORED CAUGHT STAT TTY TIME COMMAND
79 24618 0000000000000000 0000000000000000 0000000000001000 0000000180012201 SNs ? 0:00 /usr/bin/exim -q2m
Signals are delivered if I try to send ALRM by hand.
So looks like ALRM isn't send. How's that organized in exim
- is there one process that sends ALRM periodically?
--
Arkadiusz Miśkiewicz PLD/Linux Team
arekm / maven.pl http://ftp.pld-linux.org/
