On 6/7/2006 8:09 PM, Arias Hung wrote:
On Tue, 06 Jun 2006, Daryl C. W. O'Shea delivered in simple text monotype:
How long are messages (that are logged) taking to be scanned by
SpamAssassin when/before this happens. What timeout are you using
with spamc? You are using spamc, right, and not spamassassin?
<---snip--->
Yes, i'm using spamc. I didn't set timeout on spamc, so I'm assuming
it's at its default level 300 seconds. Should it be longer?
The default should be more than sufficient.
I'm also noticing a lot of copy_config timeouts, and a few normal
timeouts that exceed the 300 seconds. I will try upping the timeout
value ... are you aware if the copy_config be resolved this way as well,
or can upping the # of children help?
For the "normal timeouts", it sounds like you might be consistently
having a problem with bayes expiry. Although, such a problem isn't
normally consistent AND long (time wise) when using spamd. You could
try running an "sa-learn --force-expire" to see if it helps.
As for the copy_config timeouts... what kind of system load are you
seeing. 10, 50, 500, or higher? The current 20 seconds alarm is twice
the original alarm timeout, but if you've got a high enough load it
could still be a problem. You could increase this value to something
practically unusable, like 300, but I'd be really surprised (and would
like to about) if the timeout isn't being caused by insane load or
excessive swapping.
So... how much memory do you have in this machine, how much is free, and
how much (hopefully none or little) swap is being used. If swap is
being used, how much of the spamd processes are being swapped out (check
will the system is idle after it's been busy for a bit).
Since you suggested that this might be a personal workstation you're
running this on, there's a good chance that 4 children might actually be
too many.
BTW... is this Linux, or BSD, or something else?
Daryl
