Nagios Support Forum

Hi,

We are using Nagios Core version 3.2.3. I recently added a new machine to be monitored for the "http" test by Nagios. The only test I had Nagios performing for this machine was the "http" test. We immediately starting getting alerts that this host was down. I was able to get rid of these alerts by clicking on "Disable active checks of this host", choosing this machine and hitting "committ".

What test or tests does Nagios use for the "active checks of this host"? I don't believe it is a ping test because our Nagios GUI shows that we have a test called "ping" and I doubt we have a redundant test.

Thanks in advanced for any help.

I'm not completely sure what you are asking... disabling active checking does at it says in it's description, it disables scheduling of active checks on that object. Is that what you are looking to get answered?

Hi slansing,

Thanks for repsponding to my question.

I am not sure how Nagios is checking that the server I added is up. Basically, what test is it using? The only service test we have specified for that server is the “http” test but choosing the “Disable active checks of this host” would not stop this service test from being run. The only way to stop the http service test from being run is to choose the “Disable checks of all services on this host” option. And, I can see that the “http” service test is still being run.

A simplistic way to ask this is:
Before I clicked on “Disable active checks of this host”, how was Nagios checking if the “Host Status” was up or down? I don't believe it tries to verify this with a ping test, so how is it checking the “Host Status”?

Thanks again for any insight

You need to make the distinction between hosts and services.

A host is usually something you can walk up to and touch. A router, a web server, a switch, even a VoIP phone are all considered hosts. Services are processes that run on those hosts, such as HTTP, DNS, and SMTP, or they can be metrics like CPU usage, load, and disk usage. A single host has only one check run against it, but you can attach as many services to that host as you want.

When you choose to "disable active checks of this host" then you are only disabling the one check that is used for the host itself, usually ping. When you choose to "disable checks of all services on this host" then you are disabling checks for all the attached services.

Does this make sense?

Hi tmcdonald,

That does make sense. I thought it would be ping but we have a "ping" service test for our servers so it looks like we are using a redundant test on some servers. Thanks.

If that's all I will be closing the thread. Feel free to open another if you need anything in the future.