I have a Nagios server which was set up as a demo but somehow made its way into production. (How many times has that happened ) It has since become a critical support tool but running on an old pc so I'm looking at the most cost effective way (as in no cost) to migrate it to more resilient hardware in a clustered environment. I have two nice DL380's under utilised running a few Virtual MS servers so I was thinking of running two virtual machines with CentOS with Linux HA drdb and mon and all that other good stuff. Would not be my first choice to be running it under windows but seems to be a cheap solution requiring no hardware etc. Anyone think this a viable solution?
Clustering Nagios might be a bit of overkill. There's a method of running 2 near identical instances of Nagios on seperate machines, with one machine monitoring the other. That machine has alerts turned off. If the main nagios service or machine dies, the second one enables alerts.
I know there is a method to run several instances of Nagios across different machines (assuming it's Nagios 3+) but I can't find the doc at the moment
Edit: http://nagios.sourceforge.net/docs/3_0/distributed.html That's the link I was looking for
Thanks for the link! Yep I'm running 3.06 I wasn't aware that you could run it that way. I agree it may be overkill but I want to brush up on HA and drbd so thought I may as well kill two birds with one stone. Good to have the option anyway so I'll take a look at Distributed Monitoring.