puppet server 3.7.1-1 running on RHEL 5.11 abrupt shutdown

Mathew Jose

unread,

Aug 6, 2019, 1:36:23 AM8/6/19

to Puppet Users

Hi Team,

I am having a legacy setup with the below config

cat /etc/redhat-release

Red Hat Enterprise Linux Server release 5.11 (Tikanga)

puppet --version

3.7.1

rpm -qi puppet-server-3.7.1-1.el5

Name : puppet-server Relocations: (not relocatable)

Version : 3.7.1 Vendor: Puppet Labs

Release : 1.el5 Build Date: Mon 15 Sep 2014 01:34:16 PM PDT

Install Date: Tue 16 Sep 2014 09:49:08 PM PDT Build Host: powell.delivery.puppetlabs.net

Group : System Environment/Base Source RPM: puppet-3.7.1-1.el5.src.rpm

Size : 10470 License: ASL 2.0

Signature : RSA/SHA1, Mon 15 Sep 2014 03:19:55 PM PDT, Key ID 1054b7a24bd6ec30

----

My puppet master stops abruptly with out logging ANYTHING in /var/log/messages.

# service puppetmaster status

puppet dead but pid file exists

# puppet agent --configprint pidfile

/var/run/puppet/agent.pid

# puppet master --configprint pidfile

/var/run/puppet/master.pid

# /usr/bin/puppet master --configprint rundir

/var/run/puppet

# netstat -tunlp | grep 8139

I am unable to identify the cause , if i restart the puppetmaster service , it will start running in without any problems , but after sometime ( mostly in two to three hours) I can find it as stopped .

Where should i start ? Is there a way to enable debug logs for puppet master ?

I see the server Resource utilisation is always under control .. any pointers would be appreciated ,

Thanks everyone

Mathew

jcbollinger

unread,

Aug 6, 2019, 9:04:16 AM8/6/19

to Puppet Users

On Tuesday, August 6, 2019 at 12:36:23 AM UTC-5, Mathew Jose wrote:

Where should i start ? Is there a way to enable debug logs for puppet master ?

There is a log_level configuration parameter, which you can set to "debug" (without quotes) in the [main] or [master] section of puppet.conf, or which, like any other config parameter, you can specify on the command line. That would indeed be a good place to start.

I see the server Resource utilisation is always under control .. any pointers would be appreciated ,

If the agent runs on your master then you should check its log, too, for one possibility is that the master is serving itself a catalog that ensures the puppetmaster service stopped. Even if that's not the issue, if the agent runs on the master then you could consider mitigating the problem by making sure that the master's catalog ensures the puppetmaster service running, and that the agent there is configured to use a cached catalog when it cannot retrieve a fresh one.

John

Mathew Jose

unread,

Aug 6, 2019, 9:20:38 AM8/6/19

to puppet...@googlegroups.com

Thanks John,

I will enable debug logs and update this thread.

--
You received this message because you are subscribed to a topic in the Google Groups "Puppet Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/puppet-users/p_Xqegan_aY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to puppet-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-users/b12f7b70-0344-4f47-8b7a-871ffc331a2e%40googlegroups.com.