http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5313
------- Additional Comments From [EMAIL PROTECTED] 2007-03-21 22:33 -------
This is not a memory resource problem as Daryl C. W. O'Shea keeps on trying to
suggest.
I personally have 5 boxes currently setup as an smtp gateway.
1 x FBSD 4.11 machine, 4Gigs of ram, dual proc.....
last pid: 65442; load averages: 3.10, 2.44, 3.37
210 processes: 2 running, 206 sleeping, 1 stopped, 1 zombie
CPU states: 44.5% user, 0.0% nice, 8.6% system, 0.2% interrupt, 46.8% idle
Mem: 537M Active, 848M Inact, 307M Wired, 79M Cache, 199M Buf, 238M Free
Swap: 1024M Total, 7164K Used, 1017M Free
This machine never ever runs into any problems. It's the other machines that
have the problems. They are:
4 x FBSD 6.2, 4Gigs of ram, dual proc.....
last pid: 27331; load averages: 1.33, 1.22, 0.96
125 processes: 2 running, 123 sleeping
CPU states: 8.8% user, 0.0% nice, 2.6% system, 0.1% interrupt, 88.5% idle
Mem: 231M Active, 1304M Inact, 243M Wired, 7160K Cache, 112M Buf, 1731M Free
Swap: 8192M Total, 8192M Free
Memory usage is light, and the problem happens regardless or load. Even with 60
concurrent processes, my setup actually happens to reach a disk limitation with
high amounts of I/O, rather than consuming too much memory and swapping. This
*problem* presents itself on any of the FBSD 6.2 machines, while the 4.11
machine runs without spamassassin needing to be restarted.
In fact, I caught the problem in action while writing this...FBSD 6.2....here is
the problem without me restarting anything...yet:
last pid: 30559; load averages: 0.14, 0.79, 0.97
227 processes: 1 running, 224 sleeping, 2 zombie
CPU states: 0.1% user, 0.0% nice, 0.0% system, 0.0% interrupt, 99.9% idle
Mem: 509M Active, 1174M Inact, 262M Wired, 92M Cache, 112M Buf, 1479M Free
Swap: 8192M Total, 20K Used, 8192M Free
PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
487 dnslog 1 -8 0 1208K 576K piperd 2 425:26 0.00% multilog
24560 dnscache 1 76 0 4508K 3952K select 2 348:18 0.00% dnscache
479 qmaill 1 -8 0 1208K 576K piperd 0 52:10 0.00% multilog
357 root 1 76 0 1300K 840K select 0 45:02 0.00% syslogd
551 mysql 8 20 0 56852K 25300K kserel 0 43:35 0.00% mysqld
58235 qmails 1 76 0 1696K 1100K select 0 37:14 0.00% qmail-send
480 qmaill 1 -8 0 1208K 576K piperd 0 19:49 0.00% multilog
58237 qmailr 1 76 0 1328K 768K select 0 10:19 0.00%
qmail-rspawn
575 easyagnt 1 4 0 6216K 5444K accept 0 6:49 0.00% perl
467 root 1 8 0 1228K 632K nanslp 0 3:35 0.00% svscan
58238 qmailq 1 -8 0 1196K 564K piperd 0 3:03 0.00%
qmail-clean
27670 qmails 1 4 0 1212K 712K accept 0 2:32 0.00% tcpserver
14275 spamd 1 78 0 59936K 55356K select 2 1:23 0.00% perl5.8.8
15536 spamd 1 80 0 58488K 53764K select 2 1:21 0.00% perl5.8.8
58236 root 1 76 0 1200K 576K select 3 1:17 0.00%
qmail-lspawn
552 root 1 4 0 3356K 1684K select 0 0:47 0.00% sshd
559 root 1 8 0 1328K 904K nanslp 0 0:28 0.00% cron
92332 root 1 8 0 46776K 43240K nanslp 0 0:22 0.00% perl5.8.8
2152 spamd 1 8 0 52908K 48440K exitho 3 0:16 0.00% perl5.8.8
28179 spamd 1 78 0 51600K 47704K select 2 0:13 0.00% perl5.8.8
2246 spamd 1 8 0 59632K 54208K exitho 3 0:12 0.00% perl5.8.8
45012 admin 1 76 0 6080K 2592K select 0 0:12 0.00% sshd
28125 spamd 1 80 0 51584K 47612K select 0 0:10 0.00% perl5.8.8
92361 root 1 76 0 2744K 1996K CPU0 0 0:09 0.00% top
457 root 1 76 0 1212K 696K select 0 0:08 0.00% usbd
28923 spamd 1 79 0 52096K 47984K select 2 0:06 0.00% perl5.8.8
29049 spamd 1 79 0 52344K 48316K select 2 0:06 0.00% perl5.8.8
2052 spamd 1 8 0 50372K 46548K exitho 2 0:04 0.00% perl5.8.8
29175 spamd 1 77 0 49700K 46052K select 2 0:03 0.00% perl5.8.8
1830 spamd 1 8 0 50200K 46116K exitho 0 0:02 0.00% perl5.8.8
29548 spamd 1 82 0 49136K 45512K select 3 0:02 0.00% perl5.8.8
2268 spamd 1 8 0 48920K 45204K exitho 2 0:01 0.00% perl5.8.8
1511 spamd 1 8 0 47624K 43928K exitho 1 0:01 0.00% perl5.8.8
381 spamd 1 8 0 48472K 44840K exitho 1 0:00 0.00% perl5.8.8
1674 spamd 1 8 0 50024K 45904K exitho 1 0:00 0.00% perl5.8.8
385 spamd 1 8 0 47372K 43688K exitho 1 0:00 0.00% perl5.8.8
1841 spamd 1 8 0 48468K 44780K exitho 3 0:00 0.00% perl5.8.8
1641 spamd 1 8 0 47328K 43648K exitho 1 0:00 0.00% perl5.8.8
and the current smtp connections:
smtp-proxy03# fgrep status /var/log/smtpd/current | tail -1
@400000004602147f2c00658c tcpserver: status: 18/120
Message from /var/log/messages:
Mar 22 00:26:42 smtp-proxy03 spamd[92332]: prefork: select returned -1!
recovering: Bad file descriptor
Mar 22 00:27:13 smtp-proxy03 last message repeated 31 times
Mar 22 00:29:14 smtp-proxy03 last message repeated 121 times
After a restart of spamd, everything is back to normal.
Again, take note that this happens on FBSD 6.2 only as far as I can tell and
doesn't affect FBSD 4.11. I haven't tested FBSD 5.x, cause FBSD 5.x is garbage.
I can provide any additional information as this happens up to something like 10
times a day. Just let me know whatever information I can provide so that this
can get fixed. This is a show stopper (almost) ;)
Thanks
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
