I already had a problem like this on some servers. It seems it was a problem with a (purposely?) malformed mime in a message. This caused clam to go totally nuts an push the load average to an end (got about 5000.0 on a server). Now I reject every message with a "medium" mime error, and I never got the same problem again. So it seems to me it's a clam problem, not really an exim one.
drop
condition = ${if > {$mime_anomaly_level}{1}{true}{false}}
message = This message contains a MIME error
($mime_anomaly_text)
log_message = MIME Error ($mime_anomaly_text)
John Oxley wrote:
> I'm running FreeBSD 5.4-RELEASE, exim 4.62 and clam 0.88.2 on a P4 2.8
> with a gig of ram.
>
> What started happening today is clam would suddenly start using 50% of
> the processor. Doing a /usr/local/etc/rc.d/clamav-clamd.sh stop sits
> and waits for the PID of clam to stop. When I did a kill -9 of clam's
> PID, the machine went completely bananas and very unresponsive:
> # w
>
> 6:08PM up 4:49, 3 users, load averages: 605.85, 749.77, 573.68
>
> On killing exim by doing /usr/local/etc/rc.d/exim.sh stop then killall
> -9 exim-4.62-0 restores the machine to normal.
>
> I started exim and clam again, and after a couple of hours, the same
> thing happened. I can't find anything in either exim logs or clam logs
> slightly out of the ordinary.
>
> This machine has been running with no hickups for months. Has anyone
> else had even vaguely similar problems?
>
> -John
>
--
Nikademus
http://www.octools.com
.O.
..O
OOO
PGP key: http://www.llorien.org/gnupg/key.pub
signature.asc
Description: OpenPGP digital signature
-- ## List details at http://www.exim.org/mailman/listinfo/exim-users ## Exim details at http://www.exim.org/ ## Please use the Wiki with this list - http://www.exim.org/eximwiki/
