Advice on configuring desired behaviour

#1

Hi, I’m running Nagios 2 from a small server at home to monitor a couple of hosts that I have a contract to troubleshoot for.

The problem is that my home router and ISP are not the most reliable, so there’s usually a small window each day when they don’t work and I get a lot of unnecessary notifications (false positives) generated.

I have the “parents” clause of each of the remote hosts I am monitoring set to my ISP gateway host, and the ISP gateway’s parent is set to my router. There’s only a small number of services (mail, web, SSH) on each of the remote hosts that I am monitoring.

What I would like is, to NOT receive any service notifications for the remote hosts if either my router or ISP gateway goes down, but I am unsure how to go about this.

Currently, when either my router or ISP gateway goes down, I receive a critical alert for every service on the remote hosts as well as host down alerts for the remote hosts. So far none of the remote hosts or their services have actually failed, but all these false positives are driving me nuts. I thought the “parents” parameter would fix things, but it doesn’t appear to do anything.

How can I configure Nagios to get the behaviour I want?

Thanks!

P.S. one other small question, how do I increase the ping latency tolerance for the PING service?

#2

I’ll start with the PS:
you’ll have to modify the “check-host-alive” command definition which is in the “checkcommands.cfg” file. From there, it should be pretty obvious.

For your problem, there may be several ways to do it, but the easiest and most logical (well, imho :)) would be to use service dependencies and host dependencies: they should work as the “parents” you defined should have.

I’ll let you have a look by yourself, cause it’s friday evening and writting more on this forum would delay my precious week end :wink: