Problems with exim

[Philipp Marek]

> Over the last 24 hours, <root at common-lisp…> has received a number of
> messages from exim complaining about null message ids.  This seems to
> have ended with a message about an exim4 panic.

This seems to be because I put a "TaskMax" limit to exim4 (via
/etc/systemd/system/exim4.service.d/cpu.conf and
/sys/fs/cgroup/pids/system.slice/exim4.service) - the monitoring server
croaked about too many processes active (and exim4 alone had about 200
or so, IIRC)

I guess that wasn't the smoothest way, but we (EH+me) couldn't see
another (clean) way.


Furthermore, a few things have CPU quotas now; see
/etc/systemd/system/*/cpu.conf for details.


> I notice that clam-av seems not to be running as well:
> 
> Jun 30 06:33:35 common-lisp amavis[23388]: (23388-13) (!)connect to
> /var/run/clamav/clamd.ctl failed, attempt #1: Can't connect to a UNIX
> socket /var/run/clamav/clamd.ctl: No such file or directory
> 
> I don’t have time to look into things at the moment, but will have
> time in about six hours or so.
> 
> If anyone has a chance to poke around to figure out what is wrong
> before then, please report to this thread.

This message was already in the logs two weeks ago, so it's not my fault 
;)

It seems that amavis tries to contact a clamd service and, failing that,
just starts clamav each time. So "only" some more load, not a mail
classification problem, AFAICS.