I’ve been heavily focused with getting Nagios implemented in my new environment at IHOP. It was surprising to learn that an overall monitoring and alerting system was not in place yet, given the complexity and maturity of the infrastructure. I had some prior exposure to Nagios through testing a VMA, but this was my first full installation.
One of the best aspects of Nagios compared to proprietary platforms is that there is a very active open-source community, and often plugins have already been developed for your particular need. For instance, I deployed one of these SNMP plugins (check_snmp_env) to monitor our Cisco switches and devices, so now we receive alert notifications in the event of any Cisco hardware failure.
There is an issue I discovered with installing the otherwise excellent NSClient++ on Windows hosts. Be careful with installing on SQL Server 2005 servers, sometimes the client install causes MS DTC Service and then SQL Server to shut down. Other than that, the NSClient++ has been stable on our Windows 2003 and 2008 servers, both 32 bit and 64 bit machines, and VMware VM’s as well as physical servers.
One more thing, if you haven’t yet implemented SNMP in your network, you’ll probably want to plan on getting at least SNMPv2 set up on the Nagios server(s), network devices and, ideally, Windows and Linux hosts.
I should mention there is an extremely useful book titled “Nagios, 2nd Edition” which will help provide additional documentation and get you up and running quickly.