I am having a very unusual problem with Nagios. I am running a few
instances of Nagios 1.0b6 on a Linux RedHat7.2 box, one of which has
about 700+ services, most of which are polled every 5 minutes.
After a very short amount of time, there are hundred on nagios processes
hanging around for this instance. I know that you are supposed to see
some nagios processes because of the plugin forking, but this is well
beyond what there should be. It's around the 200-300 range. The box
starts dipping into swap and nagios doesn't get all of the plugins done
in time, even though the load on the box is still minimal. Soon the
plugins are almost 1/2 hour behind.
Now here is the confusing part. I turned on obsess_over_services for
the instance and created a command that just logs the results of every
check to a file. I wanted to try and see in more detail what was
happening. I stopped and started the instance, and now the problem is
gone. All the checks are completed at the right time, there is plenty
of free memory (no swapping) and the process count stays very low.
Is this some form of bug? What is special with the obsess_over_services
setting that gets rid of the problem? If anyone has any idea what could
be going on, please respond. Thanks.
-Russell Scibetti
--
Russell Scibetti
Quadrix Solutions, Inc.
http://www.quadrix.com
(732) 235-2335, ext. 7038
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: russell@quadrix.com