Nagios is an old, stable, strangely configured package that is simultaneously one of the most loved and most hated server-based monitoring applications. It is loved because of it’s hierarchical structures allowing lots of flexibility and power, and it’s very stable and reliable once you get it set up, it’s loathed by users who want something that “just works” when you install it.
A single Nagios server running on fairly low powered hardware can monitor hundreds of servers and other network doo-dads with ease and tell you when something bad happened, or is about to. You can tell it to notify one guy (or a list) if it’s during the day and of a level of severity, and another guy (or list) after hours, weekends, whatever. This is why it’s powerful. It can also tell you something granular like how many e-mails are in a mail queue on a remote server when they get over a threshold, something cool like that, but roll up your sleeves if you want that to happen, you have to understand “Nagios logic” (and install remote reporting clients too and have your network allow them to talk). Here’s a pretty simple setup howto, if you can survive this, you may be able to figure out how to survive more complex setups, but most documentation uses odd Nagios syntax, so you still might have to translate what they’re saying in order to make your setup work 🙂
Nagios server setup howto
apt-get install nagios3 nagios-plugins nagios-nrpe-plugin (WORKGROUP -> just hit 'OK')