Passive Checks and nagios core "agent"

[johndoe]

Raising a question here about how everyone deals with passive checks.

We use mostly 99% passive checks in our infrastructure, the way we've been going about collecting the information on each node is to install and configure nagios core on each machine.

You may say... well nagios core is quite lightweight so why all the fuss? The idea here is simple, if we're managing all through nagiosXI it doesn't make much sense to have Apache installed as a dependency on all machines when all we need is just "agent" functionality that checks the local machine and forwards it to NagiosXI.

Due to the nature of our machines we must be strict and run only what is strictly needed, the core depending on Apache just makes it bloated and opens another port (HTTP 80) which although we could close or change just seems like an unnecessary step as I believe many would benefit from a lightweight agent.

Am i correct on my assumptions that this is how people have been dealing with collecting passive results? Or what's the best way of going about this? All we need is just the plugins and a wrapper to ensure they are ran at specific times and the results sent to naigosXI. Pretty much just an "agent" functionality.

[scottwilkerson]

You might want to look at NRDS
http://library.nagios.com/library/produ ... s-tutorial

This is a passive agent designed to deliver the needed plugins, run them on interval and return the results to your XI server.

This is pre-installed in Nagios XI 2012 under Admin -> NRDS Config Manager

[johndoe]

Very good, exactly what I was looking for, need to try this out