Nagios: Sysadmin’s Best Friend

Nagios is one of those “I can’t believe it’s not butter” discoveries of mine. Except replace butter with commercial and add “and that I didn’t find it sooner”. It’s opensource server\service monitoring software with notification features and plugins and it’s flippin’ awesome. I slept better last night than I have in a long time…

Server Monitoring

I’m a reluctant sysadmin for a linux server I can’t really afford to host. I’ve learned a lot in the last few months, and have it to a point where I suffer almost zero downtime. But I constantly worry about its health and happiness and spend a lot of time checking up on things manually. I’ve coded some homemade server alarms and log scanners, but nothing elaborate. Nagios has provided me with a web interface for quickly determining what’s functioning, any problems it’s had since the last time I looked, and notifying me by email, phone or through a Firefox plugin that something is wrong.

Plugins

In addition to basic server health, you can get plugins for checking the uptime of any http server. I have a few sites on a shared host that goes through bouts of early-morning downtime which I like to leverage for discounts. This gives me a great map of downtime, to the minute, and I no longer need to run programs like Server Siren on a local machine.

I also run my own IMAP server, which I’ve just recently figured out how to make dependable. I can monitor it for downtime and have Nagios notify my phone if anything goes wrong. But here’s the great part: I can run an event handler first, in this case one that restarts the service, before I even have to worry about it. I can have it check 3 or 4 times over a 5 minute period and if it still hasn’t come back up, then give me a call (well, a text).

There’s a plugin API with lots of hooks. It’s easy to set up on most servers. It’s opensource. I love this thing. [Link]