Checking if server is alive and taking action if it isn't

Question

Possible Duplicate:
What tool do you use to monitor your servers?

I am looking for a monitoring solution for a server farm. It should resolve following issues:

1. Detect if the host has failed or alive and if possible detect whether the host really failed or if it is not reachable due to networking failure

2. If host fails, run a script (which does nothing but run some commands on other host(s) in the farm)

Answer 1

Nagios does what you want. Network failures can be handled by the dependency system, scripts on other hosts by the event infrastructure.

Answer 2

Although it sounds like what you want to do would be covered by Nagios, as SvenW mentioned, given that your tags include "high-availability" and "clustering", you might want to look into Pacemaker, which is a full-featured and extremely complex solution for clustering and high-availability that is capable of doing just about anything (once you work out how).