I have an internal lab I need to monitor, but I want to make sure that I'm monitoring the right stuff (this is my first crack at operational management). Right now I have the following servers to monitor:
- Web Servers (IIS)
- Database server (SQL - OLTP)
- Data warehouse server (SQL - OLAP)
Plus 3 servers that actually generate load during load tests.
Currently I'm just monitoring for disk space usage and am getting alerts if any logical disk falls below 10% free space. None of these machines are production, they're all in a testing lab, so 24/7 uptime isn't required and we don't have someone on call to fix the machines. Mainly we just want a way to know if a disk is about to fail, if we're running out of space during the day (due to database bloat during a test or whatever), and anything I'm not really considering (do I need to monitor network traffic, for example?)
For the purposes of this question, assume I'm just running Perfmon and am picking out the counters manually.