Nagios Support Forum

What version of nagios are you running and can you post your nagios.cfg file?
Are you seeing any errors on the switches?
If the switches are getting polled to quickly and not responding, there should be some type of error on it.
Have you adjusted the timeout value on the check and if so, what value did you use?

It appears to have been an issue with the object (host) .cfg files of the cisco nexus switches.

We resolved the issue by removing the cisco nexus host/object .cfg files ... restarting nagios without them ... then we re-added the same exact nexus .cfg files back to the same directory and restarted nagios. Nagios has been clean for nearly 2 weeks issuing no false bandwidth alerts. Bandwidth stats are also polling properly on the Nagios graphs.

The strange thing is that the nexus .cfg files are exactly the same from when we removed and re-added them back to the directory.

The issue started to happen when we modified the nexus .cfg files (adding/removing various interface bandwidth monitors on the switches) using a script.

Our test environment is also running clean, but I will need to play with the test environment to see if I can recreate the issue and apply the same fix again. May be an issue with the script that was editing our object .cfg files when we remove or add an interface bandwidth monitor on the cisco switch.

Thanks for all of your help, you may close this thread.

Much appreciated.