https://bugzilla.wikimedia.org/show_bug.cgi?id=72867

            Bug ID: 72867
           Summary: Multiple queue runners on tools-mail
           Product: Wikimedia Labs
           Version: unspecified
          Hardware: All
                OS: All
            Status: NEW
          Severity: normal
          Priority: Unprioritized
         Component: tools
          Assignee: [email protected]
          Reporter: [email protected]
                CC: [email protected], [email protected],
                    [email protected], [email protected]
       Web browser: ---
   Mobile Platform: ---

On tools-mail, there are often errors in exim's paniclog:

| root@tools-mail:~# cat /var/log/exim4/paniclog 
| 2014-10-30 16:29:09 1Xjsaz-0004TE-Hw re-exec of exim (/usr/sbin/exim4) with
-Mc failed: No such file or directory
| 2014-10-30 21:09:04 1Xjwxs-0002Vi-A9 re-exec of exim (/usr/sbin/exim4) with
-Mc failed: No such file or directory
| 2014-10-30 23:49:05 1XjzSj-0006Kv-Ej re-exec of exim (/usr/sbin/exim4) with
-Mc failed: No such file or directory
| 2014-11-01 07:48:50 1XkTQY-00047u-0m re-exec of exim (/usr/sbin/exim4) with
-Mc failed: No such file or directory
| 2014-11-01 17:50:07 1XkcoR-0007T8-1y re-exec of exim (/usr/sbin/exim4) with
-Mc failed: No such file or directory
| root@tools-mail:~#

These seem to stem from multiple exim queue runners running at the same time:

| root@tools-mail:~# ps aux | fgrep exim
| 107       1215  0.0  0.0 108852  1600 ?        Ss   18:48   0:00
/usr/sbin/exim4 -bd -q10m
| root      1653  0.0  0.0   8028   612 pts/1    S+   18:53   0:00 fgrep
--color=auto exim
| 107       9782  0.0  0.0 109876  1700 ?        S    Oct30   0:00
/usr/sbin/exim4 -bd -q10m
| 107      15924  0.0  0.0 109880  1708 ?        S    07:48   0:00
/usr/sbin/exim4 -bd -q10m
| 107      17323  0.0  0.0 109880  1720 ?        S    Oct30   0:00
/usr/sbin/exim4 -bd -q10m
| 107      24421  0.0  0.0 109876  1700 ?        S    Oct30   0:00
/usr/sbin/exim4 -bd -q10m
| 107      28751  0.0  0.0 109880  1704 ?        S    17:50   0:00
/usr/sbin/exim4 -bd -q10m
| root@tools-mail:~#

These extra queue runners could be spawned from a cron job or Puppet runs;
/etc/cron.daily/exim4-base looks innocent and would only account for one extra
queue runner per day.  However:

| root@tools-mail:~# find /proc -maxdepth 1 -user Debian-exim -ls
| 33257803    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Nov  1 18:48
/proc/1215
| 31720118    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Oct 30 21:09
/proc/9782
| 32898356    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Nov  1 07:49
/proc/15924
| 31558318    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Oct 30 16:29
/proc/17323
| 31816466    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Oct 30 23:49
/proc/24421
| 33226702    0 dr-xr-xr-x   8 Debian-exim Debian-exim        0 Nov  1 17:50
/proc/28751
| root@tools-mail:~#

shows that there are multiple instances per day and the time (:09/:29/:49)
correlates nicely with the Puppet runs that happen on this VM at :08/:28/:48.

But the Puppet logs (that do cover the period above) mention only the
exim4-daemon-light/exim4-daemon-heavy fight:

| root@tools-mail:~# zgrep -i exim /var/log/puppet.log /var/log/puppet.log.?.gz
| sort -u
| /var/log/puppet.log.1.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.1.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.2.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.2.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.3.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.3.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.4.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.4.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.5.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.5.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.6.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.6.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log.7.gz:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log.7.gz:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| /var/log/puppet.log:Notice:
/Stage[main]/Exim4/Package[exim4-daemon-light]/ensure: created
| /var/log/puppet.log:Notice:
/Stage[main]/Toollabs::Mailrelay/Package[exim4-daemon-heavy]/ensure: created
| root@tools-mail:~#

but not any "service restarted" or something similar.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
Wikibugs-l mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikibugs-l

Reply via email to