yes, we have configured the SA to 20 children

/usr/bin/spamd -v -u vpopmail -m 20 -x -q -s stderr -r /var/run/spamd/spamd.pid 
\
        -i 172.16.10.9 -A 172.16.10.0/24 2>&1 | \
        /usr/local/bin/setuidgid qmaill \
        /usr/local/bin/multilog t !spamdappend /var/log/qmail/spamd &
        echo "spamd started"
        ;;

memory rises after the spamd stopped working


> Subject: Re: FW: SpamAssassin error Interrupted system call
> From: guent...@rudersport.de
> To: users@spamassassin.apache.org
> Date: Thu, 4 Jun 2009 23:50:10 +0200
> 
> On Thu, 2009-06-04 at 17:18 +0000, Luis campo wrote:
> >              total       used       free     shared    buffers     cached
> > Mem:       1033468    1012956      20512          0      71440     270720
> > -/+ buffers/cache:     670796     362672
> > Swap:      2031608          0    2031608
> 
> OK, no swap usage initially.
> 
> > works for a few minutes then stops again
> 
> A few minutes. When originally reporting the issue, you mentioned 20
> minutes. So, did the operational time decrease, since you doubled the
> spamd children to 20?
> 
> > @400000004a27f6b60c9922c4 simscan:[26843]:CLEAN (0.00/3.00):112.9283s:: 
> > @400000004a27f6b60ea04d04 simscan:[23571]:CLEAN (0.00/3.00):221.3648s:: 
> 
> That's *much* more time than you reported before. Both might hint you
> actually are hitting swap.
> 
> 
> >              total       used       free     shared    buffers     cached
> > Mem:       1033468     531860     501608          0      26700     147200
> > -/+ buffers/cache:     357960     675508
> > Swap:      2031608      13684    2017924
> 
> Hmm, these after figures are slightly odd. I take it you got that after
> killing spamd?
> 
> Yeah, there you are using swap. Not much, but then again lots of your
> physical memory has been freed, too. So that probably could just be a
> timing issue -- numbers /while/ spamd turns unresponsive would be more
> revealing.
> 
> Anyway, yes -- I agree it looks like a swap problem. Bringing up 10
> additional spamd children with a Gig of memory seriously didn't help at
> all. I'd try as Bowie suggested.
> 
> 
> Also, some questions remain un-answered.  (a) Do you scan *all*
> messages, regardless of their size? Don't do that, but skip scanning for
> messages larger than about 500 kByte. Scanning large messages consumes
> lots of RAM, and will amplify your problem.  (b) Do you have any third-
> party rule-sets or plugins enabled?
> 
> 
> -- 
> char *t="\10pse\0r\0dtu...@ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4";
> main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1:
> (c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}
> 

_________________________________________________________________
Connect to the next generation of MSN Messenger 
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline

Reply via email to