The mailing list for listmasters using Sympa

[Tornóci László <address@concealed>]

Hi,

I have been running sympa for a few years now without major problems 
(currently using 6.1.7 on RHEL 6.3). There was a blackout on July 25, 
the server suddenly lost electricity then later restarted. Since then 
task_manager.pl takes 98% of CPU, and consumes more and more RAM to the 
point that it consumes all, and the system will kill the process. This 
happens all the time I restart sympa. I have about 160 tasks in 
sympa/spool/tasks, their dates are mostly July 25. I tried to remove all 
tasks from sympa/spool/tasks (stopped sympa, moved all the files to 
another dir, then restarted sympa) but that didn't change anything. So I 
guess the problem is not the tasks themselves.
There is nothing special in the logs (/var/log/messages):

Jul 31 16:48:52 lists bounced[7867]: notice bounced exited normally due 
to signal
Jul 31 16:48:58 lists archived[7861]: notice archived exited normally 
due to signal
Jul 31 16:49:03 lists bulk[7854]: notice main::sigterm() signal TERM 
received, still processing current task
Jul 31 16:49:03 lists bulk[7854]: notice bulk.pl exited normally due to 
signal
Jul 31 16:49:07 lists sympa[7848]: notice main::sigterm() signal TERM 
received, still processing current task
Jul 31 16:49:07 lists sympa[7848]: notice Sympa exited normally due to 
signal
Jul 31 16:49:15 lists sympa[8173]: info Configuration file read, default 
log level 4
Jul 31 16:49:16 lists sympa[8173]: info Server started for all, pid 8175
Jul 31 16:49:16 lists sympa[8175]: notice Sympa 6.1.7 started
Jul 31 16:49:16 lists bulk[8179]: info Configuration file read, default 
log level 4
Jul 31 16:49:16 lists bulk[8179]: info Starting bulk master daemon, pid 8181
Jul 31 16:49:17 lists bulk[8181]: notice bulk.pl 6.1.7 Started
Jul 31 16:49:17 lists archived.pl: info Configuration file read, default 
log level 4
Jul 31 16:49:17 lists archived[8188]: notice archived 6.1.7 Started
Jul 31 16:49:18 lists bounced.pl: info Configuration file read, default 
log level 4
Jul 31 16:49:18 lists bounced[8194]: notice bounced Started
Jul 31 16:49:19 lists task_manager.pl: notice tools::write_pid() 
Previous process 17702 died suddenly ; notifying listmaster
Jul 31 16:49:41 lists rsyslogd-2177: imuxsock begins to drop messages 
from pid 8175 due to rate-limiting
Jul 31 16:49:46 lists rsyslogd-2177: imuxsock lost 860 messages from pid 
8175 due to rate-limiting
Jul 31 16:55:04 lists rsyslogd-2177: imuxsock lost 102 messages from pid 
6243 due to rate-limiting
Jul 31 16:55:05 lists wwsympa[6243]: info [robot 
lists.semmelweis-univ.hu] [session 4157943275349] [client 85.238.78.230] 
main::do_home() do_home
Jul 31 16:55:06 lists rsyslogd-2177: imuxsock begins to drop messages
...
Jul 31 17:01:34 lists kernel: task_manager.pl invoked oom-killer: 
gfp_mask=0x280da, order=0, oom_adj=0, oom_score_adj=0
Jul 31 17:01:34 lists kernel: task_manager.pl cpuset=/ mems_allowed=0
Jul 31 17:01:34 lists kernel: Pid: 8200, comm: task_manager.pl Not 
tainted 2.6.32-279.2.1.el6.x86_64 #1
Jul 31 17:01:34 lists kernel: Call Trace:
...
Jul 31 17:01:34 lists kernel: Out of memory: Kill process 8200 
(task_manager.pl) score 817 or sacrifice child
Jul 31 17:01:34 lists kernel: Killed process 8200, UID 0, 
(task_manager.pl) total-vm:5343504kB, anon-rss:3572544kB, file-rss:80kB


Process 8175 that the logger rate limits is /opt/sympa/bin/sympa.pl 
itself. /var/log/httpd/ssl_error_log is flooded with lines like this:
[Tue Jul 31 16:46:11 2012] [warn] [client 65.52.109.147] mod_fcgid: 
stderr: Use of uninitialized value $_[1] in sprintf at 
/usr/local/lib64/perl5/Sys/Syslog.pm line 400.
[Tue Jul 31 16:46:11 2012] [warn] [client 65.52.109.147] mod_fcgid: 
stderr: Use of uninitialized value $_[4] in sprintf at 
/usr/local/lib64/perl5/Sys/Syslog.pm line 400.
[Tue Jul 31 16:46:11 2012] [warn] [client 65.52.109.147] mod_fcgid: 
stderr: Use of uninitialized value $_[4] in sprintf at 
/usr/local/lib64/perl5/Sys/Syslog.pm line 400.

So the logging for sympa is practically useless. But that is not new 
here, it has been going like this for more than a year now. I complained 
about it on this list, but no one could help me.

But no being able to run task_manager.pl is a big problem. Do you have 
an idea what I can do about it?

                                        Yours: Laszlo