Nagios Support Forum

Here is the screenshot you asked for from "Schedule a forced immediate check". But to confirm... This is not happening with other NCPA checks at all, just this one host with this one check. wanted to make sure that was still clear.

We were able to correct the issue by deleting the host and its services and recreating everything using the NCPA wizard. Now everything returns OK, including the SVN logical disk check.

I would still like to understand how this could happen where one new check goes off the rails.

If it was the check_ncpa.py script that would impact all, not just one check on one host. Why did deleting everything and recreating fix the issue for the new check?

Is it possible that MySQL and other dat files got out of sync somehow, like retention.dat and object.cache? Just spitballing here as I can't seem to come up with a reasonable answer.

Also we noticed after recreating the host and services, the performance history data was still available. Does this mean the RRD data does not get removed when deleting services and hosts, or is it just handled at a different time via some clean up routine?

Kris

I would still like to understand how this could happen where one new check goes off the rails.

It is hard to say after the fact. Initially I thought that your command (in the config) was different, compared to the one that you were using to test from the CLI, but obviously, this was not the case. This could have been a case of a "ghost" service but I could not be sure. The important thing is that the issue is resolved now. Keep on eye on this service and let us know if you run into more issues with it in the future.

Does this mean the RRD data does not get removed when deleting services and hosts, or is it just handled at a different time via some clean up routine?

The RRDs and XMLs never get deleted, when hosts/services are removed as most companies want to keep the historical data. If you want to delete these files, you will need to do this manually.