I recently setup Nagios for the first time and I love it so far. Figuring out how to better organize all those config files took a while, but once I had a better understanding of everything I'm really enjoying this.
I have a current server setup to monitor all our EC2 server instances such as processor load, disk space, memory usage, and so on. I'm using the standard SNMP as well as NRPE to get things like the processor loads from remote machines.
That being said, I'd really like to monitor the bandwidth usage of each machine. I've found a number of plugins and guides, but they all seem to be geard towards network switches and not linux machines.
Are there any plugins or simple scripts I can use in order to see how much bandwidth our machines are consuming?
Best Answer
I now use OMD of all my monitoring setups. It's a packaged version of Nagios + check_mk + pnp4nagios + other things too.
If you're not using check_mk already, you really should try it. It's takes so much of the pain out of nagios configurations with auto-inventory (i.e. no more editing sanza after sanza of config across multiple files).
Tied with pnp4nagios you get beautiful graphs for everything your monitoring.
In answer to your question, the check_mk agent will report on all your NICs by default (bps, pps, errors, etc). Should be exactly what you want, plus you can tell the check if you want warning/critical through-put alarms.