Reporting DOWN servers which are not in config...?


Sorry if this ahs been asked, I used the search feature and did not see it.

We had an old Nagios 2 install, and turned it off a while back (old boss didnt like it). I recently turned it back on, and while it was down one of our monitored servers was removed from the network. It was obviously reported DOWN. I removed it from the configs and restarted. Cool. A few hours later? Reported as DOWN again. Every few hours. Grep’ed every config file for that server, did not exist.

I notice Nagios 3 is out - ok lets do a fresh install. Nagios 2 was installed from a rpm. I backup my configs, rpm -e nagios to remove it, and then I download and install Nagios 3. I see the configs are ser up very differently and like the new setup, so instead of using the old configs I decide to start from scratch.

I add in 3-4 servers to try it out.

A few hours later? Host mail is DOWN!

What the hell?

A few horus after that? EVERY host that was in my old 2.0 config is being reported as DOWN, like the fact they should be checked was cached somewhere.

So I backup my 3.0 configs (for which I started fresh theres NO chance the old hosts were in there) and I do a find / grepping for nagios and I rm -rf EVERY nagios directory and file I see.

Should clear it all up right?

I reinstall nagios, put my near-empty 3.0 config file back.

A few horus later? 20 DOWN hosts being reported to me again.

Why is nagios reporting hosts that arent in the configs? Where is this cache being stored? I removed everything I could find and yet its still there. Anyone? Help? I cant use nagios if its going to report down hosts that dont exist evey 3 hours…


Nevermind - reading the forums on other issues I saw someone report multiple nagios services running - so I checked ours, and I had THREE copies of nagios running. One of which had been running for weeks apparently, was still 2.0, and using a config file I deleted 2 weeks ago.


All fixed now it seems!


lol… the ghost in the machine…