On 05/03/2008 5:44 AM, Johann Spies wrote:
> On Thu, Feb 28, 2008 at 02:44:02PM +0200, Johann Spies wrote:
>> On a new mailserver with 8Gb ram and 2xdual-core CPU's we get regular
>> messages in the log:
>>
>> Feb 28 12:52:43 mail2 spamd[32558]: prefork: child states: BIBBB
>> Feb 28 12:52:44 mail2 spamd[459]: rules: failed to run TVD_STOCK1 test, 
>> skipping:
>> Feb 28 12:52:44 mail2 spamd[459]:  (child processing timeout at 
>> /usr/sbin/spamd line 1246.
>> Feb 28 12:52:44 mail2 spamd[459]: )
>>
>> And every time it involves TVD_STOCK1.

The rule doesn't look particular bad.  Have you been able to capture a
sample email that causes this?  Perhaps its an issue with a large
text/plain body with no line breaks.

>> Is this a bug in Spamassassin or in the rule? How do I fix it?
>>
>> Version:  3.2.3-0.volatile1 (on Debian Stable).
>>
>> Defaults: OPTIONS="--create-prefs --max-children 15 --helper-home-dir"
> 
> I have seen no reaction to the message quoted here.
> 
> After a "score TVD_STOCK1 0" the "child processing timeout" messages
> stopped, but exim is still complaining from time to time (11 times so
> far today and 239 times yesterday): 
> 
> error reading from spamd socket: Connection timed out
> 
> In /var/log/mail.info the message "prefork: server reached
> --max-children setting, consider raising it" appeared 188 times
> yesterday and 27 times today in the logs on the new server
> (spamassassin version 3.2.3-0.volatile1) with the --max-children 15
> setting.  On the older server (--max-children 5 and version
> 3.0.3-2sarge1) that has handled about double the numer of emails
> during the past 24 hours no such problem was reported either by exim
> or spamassassin.

3.2.3 changes the way DNS timeouts are calculated (SA used to time out
its second round of DNS lookups way too early).  Is the machine (or
specifically the spamd children) actually busy, or is everything sitting
rather idle.

You only say how many error messages you've received but not actually
how many messages you're trying to scan (and what your peak volume is),
so I can't tell how frequent this really occurs or recommend how many
children you may need.

Additionally, your mail (spam) volume may have gone up enough recently
that you'd be seeing the same messages with 3.0.3.  I've been seeing the
normal rapid annual increase in volume (between November and March) on
nearly every system I have access to.

> Any idea on what is going on here?

You could run spamd in debug mode to see where time is being spent doing
whatever it is doing.

Daryl

Reply via email to