spamass-milt-list
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: spamass-milter stops working under heavy load


From: Kees Monshouwer
Subject: Re: spamass-milter stops working under heavy load
Date: Thu, 17 Jun 2004 19:41:27 +0200

----- Original Message ----- 
From: <address@hidden>
To: <address@hidden>
Sent: Wednesday, June 16, 2004 10:41 PM
Subject: spamass-milter stops working under heavy load


> Our two sendmail/spamassmilter-0.2.0-3 relays are working fine when
idling. But
> under more heavy load (20K/day) the process stops working after 1 to 2
hours.
> Then we got only this messages in the logs:
>
> May  9 07:06:23 rdh-relay sendmail[23731]: i4972NC4023731:
> mail.mydomain.com [1.3.9.9] did not issue MAIL/EXPN/VRFY/ETRN
> during connection to MTA
> May  9 07:06:26 rdh-relay sendmail[23733]: i4972QC4023733: Milter
> (spamassassin): timeout before data read
> May  9 07:06:26 rdh-relay sendmail[23733]: i4972QC4023733: Milter
> (spamassassin): to error state
> May  9 07:06:26 rdh-relay sendmail[23733]: i4972QC4023733: Milter
> (spamassassin): init failed to open
> May  9 07:06:26 rdh-relay sendmail[23733]: i4972QC4023733: Milter
> (spamassassin): to error state
>
> We have scrubbed to external Linux Sendmail specialists and both ran out
of ideas.
> They both confirmed our configuration is fine (all the relay prevention
stuff,
> white/blacklists, DNS MX record checking, spamlist checking, ClamAV check
et
> cetera).
> hardware: HP Proliant DL360 (single 2,8Ghz Xeon, 1GB RAM, 2x 72GB U320
SCSI discs)
> software:
>       -Suse 9.0 professional
>       -sendmail-8.12.10-7
>       -spamassassin-2.63/perl-spamassassin-2.63
>       -perl-5.8.1-81
>       -amavisd-new-20030616p5-23
>       -bind dns 9.2.2
>       -clamav-0.70-1
>       -spamass-milter-0.2.0-3
>
> Any idea?
>   Regards
>     Bernd
>
>
> _______________________________________________
> Spamass-milt-list mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/spamass-milt-list
>

I think te problem is the number of alowed spamd child processes. Increase
this number with the -m option of spamd.

from man spamd

       -m number, --max-children=number
           Specify a maximum number of children to spawn. Spamd will wait
           until another child finishes before forking again. Meanwhile,
           incoming connections will be queued.

           Please note that there is a OS specific maximum of connections
that
           can be queued (Try "perl -MSocket -e'print SOMAXCONN'" to find
this
           maximum). Also, this option causes spamd to create an extra pipe
           for each child.

(Little sugestion for Dan... What hapens when the OS limit is reached????
Crash????)


---
Kees Monshouwer








reply via email to

[Prev in Thread] Current Thread [Next in Thread]