You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by "Dan Mahoney, System Admin" <da...@prime.gushi.org> on 2004/09/22 18:46:28 UTC
timeout waiting for input from local during Draining Input
Hey all,
Since the upgrade to 3.0.0, my load average has shot up quite a bit:
last pid: 130; load averages: 9.65, 9.79, 7.00
up 11+14:33:24 12:45:36
868 processes: 9 running, 857 sleeping, 2 zombie
CPU states: 40.5% user, 5.2% nice, 44.6% system, 9.7% interrupt, 0.0%
idle
Mem: 632M Active, 71M Inact, 223M Wired, 39M Cache, 108M Buf, 5544K Free
Swap: 2048M Total, 227M Used, 1821M Free, 11% Inuse
PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU
COMMAND
93826 root 64 0 23456K 22896K RUN 1:30 10.89% 10.89%
perl
93828 root 64 0 28328K 27760K RUN 1:17 10.84% 10.84%
perl
93827 root 63 0 24956K 24348K RUN 1:21 10.30% 10.30%
perl
93824 root 64 0 26820K 26224K RUN 1:24 10.25% 10.25%
perl
93825 root 63 0 24852K 24304K RUN 1:27 9.77% 9.77%
perl
9
And I'm seeing a lot of entries like this in my log:
Sep 22 12:40:35 prime sm-mta[97463]: i8MGeP9Y097463:
from=<wp...@modello-svizzero.ch>, size=760, class=0, nrcpts=1,
msgid=<34...@cica.es>, proto=SMTP, daemon=MTA,
relay=[220.234.225.135]
Sep 22 12:42:36 prime sm-mta[97568]: i8MGeP9Y097463: timeout waiting for
input from local during Draining Input
Any ideas what could be causing this?
-Dan
--
--------Dan Mahoney--------
Techie, Sysadmin, WebGeek
Gushi on efnet/undernet IRC
ICQ: 13735144 AIM: LarpGM
Site: http://www.gushi.org
---------------------------
Re: timeout waiting for input from local during Draining Input
Posted by Theo Van Dinter <fe...@kluge.net>.
On Wed, Sep 22, 2004 at 01:17:55PM -0400, Dan Mahoney, System Admin wrote:
> Hrmmm, a good feature in a future version would be the ability to grow
> this number dynamically up to some hard-and-fast high limit.
Yeah. The first step was going from the previous "fork() per incoming
message" model to the 3.0 "pre-fork() children" model. I was thinking
about emulating the httpd method of prefork, dynamically spawn/kill,
etc, but that was a bit much to bite off for 3.0.
--
Randomly Generated Tagline:
"See, that's the advantage of running OS/2 ... Viruses don't support it."
- Theo talking to Rob
Re: timeout waiting for input from local during Draining Input
Posted by Alex S Moore <as...@edge.net>.
On Wed, 2004-09-22 at 12:17, Dan Mahoney, System Admin wrote:
> Hrmmm, a good feature in a future version would be the ability to grow
> this number dynamically up to some hard-and-fast high limit.
This is the way that I am seeing mimedefang-multiplexor work. Have you
looked at MIMEDefang?
Alex
Re: timeout waiting for input from local during Draining Input
Posted by "Dan Mahoney, System Admin" <da...@prime.gushi.org>.
On Wed, 22 Sep 2004, Theo Van Dinter wrote:
> On Wed, Sep 22, 2004 at 12:55:44PM -0400, Dan Mahoney, System Admin wrote:
>> Additionally, Im seeing PAGES of spamc when I do a ps aux|grep spam:
>
> That can definitely happen if you don't have enough spamd processes for
> the number of requests (or if the spamd processes running are all doing
> something intensive like a bayes db upgrade or expire they'll be tied up and
> can't handle requests).
Hrmmm, a good feature in a future version would be the ability to grow
this number dynamically up to some hard-and-fast high limit.
On a related note, has anyone had any success in generating graphs of
spamd logs? i.e. amount of time taken per request, average scores,
average spam counts, etc?
-Dan
>
> --
> Randomly Generated Tagline:
> First Law of System Requirements:
> "Anything is possible if you don't know what you're talking about..."
>
--
Amerikanskaya firma Transceptor Technology pristupila k poizvodstu komputerov "Personal'ni Sputnik"
--Snap, "The Power"
--------Dan Mahoney--------
Techie, Sysadmin, WebGeek
Gushi on efnet/undernet IRC
ICQ: 13735144 AIM: LarpGM
Site: http://www.gushi.org
---------------------------
Re: timeout waiting for input from local during Draining Input
Posted by Theo Van Dinter <fe...@kluge.net>.
On Wed, Sep 22, 2004 at 12:55:44PM -0400, Dan Mahoney, System Admin wrote:
> Additionally, Im seeing PAGES of spamc when I do a ps aux|grep spam:
That can definitely happen if you don't have enough spamd processes for
the number of requests (or if the spamd processes running are all doing
something intensive like a bayes db upgrade or expire they'll be tied up and
can't handle requests).
--
Randomly Generated Tagline:
First Law of System Requirements:
"Anything is possible if you don't know what you're talking about..."
Re: timeout waiting for input from local during Draining Input
Posted by "Dan Mahoney, System Admin" <da...@prime.gushi.org>.
On Wed, 22 Sep 2004, Dan Mahoney, System Admin wrote:
More on this:
I'm seeing an oddness in that each message seemingly has two procmail
processes atached to it (snippets from ps aux):
kimry 2543 0.0 0.1 1208 832 ?? I 12:50PM 0:00.00
procmail -f astonfrm@yahoo.com -Y -a kimmy-aht-eferfa.net -d
kimry 2526 0.0 0.1 1208 832 ?? Is 12:50PM 0:00.02
procmail -f astonfrm@yahoo.com -Y -a kimmy-aht-eferfa.net -d
root 2525 0.0 0.3 5488 2728 ?? I 12:50PM 0:00.02
sendmail: ./i8MGoYJ0002486 from queue (sendmail)
root 2512 0.0 0.3 5740 3032 ?? Is 12:50PM 0:00.00
sendmail: ./i8MGoUOm002506 mx2.hotmail.com.: user open (sendm
fstick 2504 0.0 0.0 1432 480 ?? I 12:50PM 0:00.00
/usr/local/bin/spamc
fstick 2503 0.0 0.1 1248 788 ?? I 12:50PM 0:00.00
procmail -f sez5uox7dest84-tqjglx6xi3hfe@fendley7.org -Y -a -
fstick 2489 0.0 0.1 1248 788 ?? Is 12:50PM 0:00.02
procmail -f sez5uox7dest84-tqjglx6xi3hfe@fendley7.org -Y -a -
root 2488 0.0 0.3 5456 2684 ?? I 12:50PM 0:00.02
sendmail: ./i8MGoGbl002378 from queue (sendmail)
max 2485 0.0 0.0 1436 488 ?? I 12:50PM 0:00.00
/usr/local/bin/spamc
max 2484 0.0 0.1 1248 808 ?? I 12:50PM 0:00.00
procmail -f MJSMOEZP@prontomail.com -Y -a cyrus-aht-bbi.com -
max 2467 0.0 0.1 1248 808 ?? Is 12:50PM 0:00.02
procmail -f MJSMOEZP@prontomail.com -Y -a cyrus-aht-bbi.com -
root 2466 0.0 0.3 5472 2708 ?? I 12:50PM 0:00.02
sendmail: ./i8MGoLkU002410 from queue (sendmail)
lmerge 2460 0.0 0.1 1432 512 ?? I 12:50PM 0:00.00
/usr/local/bin/spamc
lmerge 2459 0.0 0.1 1236 784 ?? I 12:50PM 0:00.00
procmail -f %FROMUSER@hushmail.com -Y -a rusba-aht-magesfire.
lmerge 2445 0.0 0.1 1236 784 ?? Is 12:50PM 0:00.02
procmail -f %FROMUSER@hushmail.com -Y -a rusba-aht-magesfire.
max 2435 0.0 0.0 1436 484 ?? I 12:50PM 0:00.00
/usr/local/bin/spamc
max 2434 0.0 0.1 1204 800 ?? I 12:50PM 0:00.00
procmail -f architectural@email.uc.edu -Y -a bcm-aht-bbi.com
max 2420 0.0 0.1 1204 800 ?? Is 12:50PM 0:00.02
procmail -f architectural@email.uc.edu -Y -a bcm-aht-bbi.com
root 2419 0.0 0.3 5472 2708 ?? I 12:50PM 0:00.02
sendmail: ./i8MGnxVx002325 from queue (sendmail)
kimry 2408 0.0 0.1 1432 520 ?? S 12:50PM 0:00.00
/usr/local/bin/spamc
kimry 2407 0.0 0.1 1208 836 ?? I 12:50PM 0:00.00
procmail -f astonfrm@yahoo.com -Y -a kimmy-aht-eferfa.net -d
kimry 2389 0.0 0.1 1208 836 ?? Is 12:50PM 0:00.02
procmail -f astonfrm@yahoo.com -Y -a kimmy-aht-eferfa.net -d
root 2388 0.0 0.3 5488 2728 ?? I 12:50PM 0:00.02
sendmail: ./i8MGoAL3002370 from queue (sendmail)
lmerge 2368 0.0 0.0 1432 484 ?? I 12:50PM 0:00.00
/usr/local/bin/spamc
lmerge 2367 0.0 0.1 1248 808 ?? I 12:50PM 0:00.00
procmail -f %FROMUSER@runbox.com -Y -a stanfunicelli-aht-mage
lmerge 2350 0.0 0.1 1248 808 ?? Is 12:50PM 0:00.02
procmail -f %FROMUSER@runbox.com -Y -a stanfunicelli-aht-mage
fstick 2316 0.0 0.0 1432 488 ?? I 12:49PM 0:00.00
/usr/local/bin/spamc
fstick 2315 0.0 0.1 1244 836 ?? I 12:49PM 0:00.00
procmail -f mailer-fpc10-return-272-junk=darkdaemon.com@click
fstick 2298 0.0 0.1 1244 836 ?? Is 12:49PM 0:00.02
procmail -f mailer-fpc10-return-272-junk=darkdaemon.com@click
root 2297 0.0 0.3 5456 2692 ?? I 12:49PM 0:00.02
sendmail: ./i8MGnq07002231 from queue (sendmail)
web_virus 2294 0.0 0.0 1432 488 ?? I 12:49PM 0:00.00
/usr/local/bin/spamc
web_virus 2293 0.0 0.1 1244 836 ?? I 12:49PM 0:00.00
procmail -f bounce-zvmanntsmiwnn@ernmtthbn.pharmwarehouse.com
web_virus 2276 0.0 0.1 1244 836 ?? Is 12:49PM 0:00.02
procmail -f bounce-zvmanntsmiwnn@ernmtthbn.pharmwarehouse.com
Additionally, Im seeing PAGES of spamc when I do a ps aux|grep spam:
lmerge 1479 0.0 0.1 1432 572 ?? I 12:48PM 0:00.00
/usr/local/bin/spamc
max 1460 0.0 0.1 1436 576 ?? I 12:48PM 0:00.00
/usr/local/bin/spamc
max 1397 0.0 0.1 1436 520 ?? I 12:48PM 0:00.00
/usr/local/bin/spamc
rwmtcww 1351 0.0 0.1 1432 508 ?? I 12:48PM 0:00.00
/usr/local/bin/spamc
rbcorner 1331 0.0 0.1 1432 568 ?? I 12:48PM 0:00.00
/usr/local/bin/spamc
max 1286 0.0 0.1 1436 584 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
lmerge 1262 0.0 0.1 1432 568 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
max 1241 0.0 0.1 1436 528 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
fstick 1198 0.0 0.1 1432 572 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
lmerge 1165 0.0 0.1 1432 564 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
max 1095 0.0 0.1 1436 568 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
max 1030 0.0 0.1 1436 520 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
lmerge 874 0.0 0.1 1432 568 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
typhon 746 0.0 0.1 1432 512 ?? I 12:47PM 0:00.00
/usr/local/bin/spamc
lmerge 701 0.0 0.1 1432 508 ?? I 12:46PM 0:00.00
/usr/local/bin/spamc
lmerge 597 0.0 0.1 1432 512 ?? I 12:46PM 0:00.00
/usr/local/bin/spamc
fstick 570 0.0 0.1 1432 580 ?? I 12:46PM 0:00.00
/usr/local/bin/spamc
wrin 350 0.0 0.1 1432 512 ?? I 12:46PM 0:00.00
/usr/local/bin/spamc
max 324 0.0 0.1 1436 520 ?? I 12:46PM 0:00.00
/usr/local/bin/spamc
max 242 0.0 0.1 1436 576 ?? I 12:45PM 0:00.00
/usr/local/bin/spamc
max 221 0.0 0.1 1436 516 ?? I 12:45PM 0:00.00
/usr/local/bin/spamc
max 194 0.0 0.1 1436 512 ?? I 12:45PM 0:00.00
/usr/local/bin/spamc
max 99935 0.0 0.1 1436 572 ?? I 12:45PM 0:00.00
/usr/local/bin/spamc
sadcat 99872 0.0 0.1 1432 576 ?? I 12:45PM 0:00.00
/usr/local/bin/spamc
max 99795 0.0 0.1 1436 568 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
icehacker 99724 0.0 0.1 1432 580 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
rbcorner 99680 0.0 0.1 1432 568 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
nduli 99657 0.0 0.1 1432 512 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
max 99615 0.0 0.1 1436 572 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
rbcorner 99594 0.0 0.1 1432 512 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
typhon 99567 0.0 0.1 1432 572 ?? I 12:44PM 0:00.00
/usr/local/bin/spamc
lmerge 99370 0.0 0.1 1432 568 ?? I 12:43PM 0:00.00
/usr/local/bin/spamc
max 99175 0.0 0.1 1436 564 ?? I 12:43PM 0:00.00
/usr/local/bin/spamc
max 98956 0.0 0.1 1436 572 ?? I 12:43PM 0:00.00
/usr/local/bin/spamc
root 93791 0.0 2.0 21548 20604 ?? Is 12:32PM 0:00.95
/usr/local/bin/spamd -c -d -r /var/run/spamd.pid (perl)
root 4162 0.0 0.0 392 212 ph DL+ 12:54PM 0:00.00 grep
spam
was1 4160 0.0 0.1 1432 504 ?? S 12:54PM 0:00.00
/usr/local/bin/spamc
max 4141 0.0 0.1 1436 500 ?? S 12:54PM 0:00.00
/usr/local/bin/spamc
danial 4123 0.0 0.1 1432 496 ?? S 12:54PM 0:00.00
/usr/local/bin/spamc
I'm not quite sure something is right here. Mail *is* coming through, but
it doesn't seem that spamassassin is catching it all.
-Dan
> Hey all,
>
> Since the upgrade to 3.0.0, my load average has shot up quite a bit:
>
> last pid: 130; load averages: 9.65, 9.79, 7.00 up 11+14:33:24 12:45:36
> 868 processes: 9 running, 857 sleeping, 2 zombie
> CPU states: 40.5% user, 5.2% nice, 44.6% system, 9.7% interrupt, 0.0% idle
> Mem: 632M Active, 71M Inact, 223M Wired, 39M Cache, 108M Buf, 5544K Free
> Swap: 2048M Total, 227M Used, 1821M Free, 11% Inuse
>
> PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU
> COMMAND
> 93826 root 64 0 23456K 22896K RUN 1:30 10.89% 10.89% perl
> 93828 root 64 0 28328K 27760K RUN 1:17 10.84% 10.84% perl
> 93827 root 63 0 24956K 24348K RUN 1:21 10.30% 10.30% perl
> 93824 root 64 0 26820K 26224K RUN 1:24 10.25% 10.25% perl
> 93825 root 63 0 24852K 24304K RUN 1:27 9.77% 9.77% perl
> 9
>
> And I'm seeing a lot of entries like this in my log:
>
> Sep 22 12:40:35 prime sm-mta[97463]: i8MGeP9Y097463:
> from=<wp...@modello-svizzero.ch>, size=760, class=0, nrcpts=1,
> msgid=<34...@cica.es>, proto=SMTP, daemon=MTA,
> relay=[220.234.225.135]
> Sep 22 12:42:36 prime sm-mta[97568]: i8MGeP9Y097463: timeout waiting for
> input from local during Draining Input
>
> Any ideas what could be causing this?
>
> -Dan
>
>
> --
>
> --------Dan Mahoney--------
> Techie, Sysadmin, WebGeek
> Gushi on efnet/undernet IRC
> ICQ: 13735144 AIM: LarpGM
> Site: http://www.gushi.org
> ---------------------------
>
--
"Hey, call me anything you like. I'm Dan to my friends, gushi to my
close friends, 'hey, you' to my girlfriend, 'mrrow?' to my cat, and 'why
the hell is the router on fire?' to my job.
-Dan Mahoney
12/2/02
--------Dan Mahoney--------
Techie, Sysadmin, WebGeek
Gushi on efnet/undernet IRC
ICQ: 13735144 AIM: LarpGM
Site: http://www.gushi.org
---------------------------