Issue #2074 has been updated by Kristian Van Der Vliet.
Likewise we've now run into this problem on a whole bunch of machines (~200).
They are all VMs under VMWare, and running a mix of CentOS 5 (5.2, 5.3 & 5.4)
and Debian Lenny. The CentOS machines are running 0.25.4 while the Debian
machines are running 0.25.1.
>From one of our CentOS machines:
# cat /etc/redhat-release
CentOS release 5.2 (Final)
# /usr/sbin/puppetd --version
0.25.4
# ps aux | grep defunct
root 512 0.0 0.0 0 0 ? Z Apr28 0:00 [sh] <defunct>
root 514 0.0 0.0 0 0 ? Z Apr28 0:00 [sh] <defunct>
root 516 0.0 0.0 0 0 ? Z Apr28 0:00 [sh] <defunct>
root 1396 0.0 0.0 0 0 ? Z Apr28 0:00 [sh] <defunct>
...
root 25217 0.0 0.0 0 0 ? Z 09:24 0:00 [sh] <defunct>
root 26056 0.0 0.0 0 0 ? Z 09:54 0:00 [sh] <defunct>
root 26058 0.0 0.0 0 0 ? Z 09:54 0:00 [sh] <defunct>
root 26060 0.0 0.0 0 0 ? Z 09:54 0:00 [sh] <defunct>
root 27023 0.0 0.0 0 0 ? Z 10:24 0:00 [sh] <defunct>
root 27025 0.0 0.0 0 0 ? Z 10:24 0:00 [sh] <defunct>
root 27027 0.0 0.0 0 0 ? Z 10:24 0:00 [sh] <defunct>
# ps aux | grep defunct | wc -l
139
# /etc/init.d/puppet restart
# ps aux | grep defunct | wc -l
4
On a Debian machine:
# cat /etc/debian_version
5.0.2
# puppetd --version
0.25.1
# ps aux | grep defunct
root 2675 0.0 0.0 0 0 ? Z Apr27 0:00 [sh] <defunct>
root 2712 0.0 0.0 0 0 ? Z Apr27 0:00 [sh] <defunct>
root 2723 0.0 0.0 0 0 ? Z Apr27 0:00 [sh] <defunct>
...
root 29710 0.0 0.0 0 0 ? Z 10:28 0:00 [sh] <defunct>
root 30007 0.0 0.0 0 0 ? Z 10:59 0:00 [sh] <defunct>
root 30015 0.0 0.0 0 0 ? Z 10:59 0:00 [sh] <defunct>
root 30017 0.0 0.0 0 0 ? Z 10:59 0:00 [sh] <defunct>
# ps aux | grep defunct | wc -l
220
# /etc/init.d/puppet restart
# ps aux | grep defunct | wc -l
4
These machines are all running the "same" quite small manifest. On the CentOS
machines it will have 2 exec{}s and the Debian machines it will run 3, so that
doesn't directly correlate to the number of defunct processes Puppet is
producing.
For a little background, we were already using Puppet to build out new
configurations for ~60 machines and do not have this problem on any of the
machines we've built up "from scratch" with Puppet, but we've now installed
Puppet on 200 "legacy" machines in order to control things like SSH & local
users passwords, and on those machines we've now run into this problem with
defunct processes. On the legacy machines we're using a slimmed-down manifest
which does the bare minimum we require, while on the "new build" machines we
have a much larger manifest that we're running: in other words, we're only
seeing this problem on machines running this particular manifest.
----------------------------------------
Bug #2074: Lots of Defunct Processes left behind by puppetd's Execs
http://projects.puppetlabs.com/issues/2074
Author: Hari Sekhon
Status: Needs more information
Priority: Normal
Assigned to:
Category:
Target version:
Affected version: 0.24.8
Keywords:
Branch:
I'm running CentOS 5.2 and have lots of <defunct> processes in the process
table dating back a month.
The processes are all different programs, so I cannot believe that it is the
fault of any of the programs involved considering they are shells (sh) and
common programs.
Restarting puppetd clears them. Surely puppetd should have closed them?
--
You have received this notification because you have either subscribed to it,
or are involved in it.
To change your notification preferences, please click here:
http://projects.puppetlabs.com/my/account
--
You received this message because you are subscribed to the Google Groups
"Puppet Bugs" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/puppet-bugs?hl=en.
