Caught SIGSEGV, shutting down

[hemak88]

Script detects Nagios shutdown error in logs. So by that time error is detected, plugin will be stopped and nagios will be down, if in case a plugin is causing the issue.

However, I have enabled nagios alerts for processes and memory usage for nagios server itself. It has graphs available. These graphs has not shown any abrupt increase in memory or processes just before issue started.

[scottwilkerson]

However, I have enabled nagios alerts for processes and memory usage for nagios server itself. It has graphs available. These graphs has not shown any abrupt increase in memory or processes just before issue started.

It can be extremely hard deciphering these as a plugin with a memory leak can spike the memory in a fraction of a second until the memory killer stops the process.

Any hints in the syslog on the server at the time the memory spikes?

[hemak88]

In syslog, no memory errors or anomaly messages appeared. Plugins means, I am using only normal nagios plugins and few custom bash scripts as plugins. Any known issues with any plugins? If it was a plugin, then we should be able to see that in nagios logs and after that plugin check, the nagios service should crash. Am I thinking in the right direction?

[tgriep]

This link has some details about memory issues running the check_http plugin but is seems web site related.
https://github.com/nagios-plugins/nagio ... issues/235

The nagios.log file may not have enough information to determine what the issue is unless you enable debugging.
I would tale a look at the /var/log/messages file to see if you can find something in there.