Over the last couple of years, and especially since the recent release of the System Center suite, I have noticed more and more companies implementing OpsMgr, which means more and more engineers are expected to not only interact with OpsMgr, but also to manage and maintain it. And often, this is an additional duty on top of an already overfull work schedule, and many of these engineers do not have a background in monitoring.
While OpsMgr is relatively easy to configure and manage, especially with the help of the great resources found online, one still needs to understand a couple of basic concepts around monitoring to ensure it works effectively for each environment.
When I talk about monitoring as a concept, I like to talk about the four legged creature, as I believe there are 4 components that make up effective monitoring, and if you take care of the four legs, you should find managing OpsMgr a lot easier as well.
The four legs as follows:
- Monitor: Can my system ‘see’ the device/application/component?
- Alert: Will my system raise an alarm of sorts if something happens? Typically, this alarm should be an alert visible in the console.
- Notify: Will my system let someone know if something happens? Typically, this is an email in someone’s inbox, a call logged at the helpdesk or even an SMS message on someone’s mobile.
- Report: Can I look at all the things that happened in my environment over a period of time?
This may sound like a simplistic approach, but I have found that this approach has allowed me to provide effective monitoring across many different environments, regardless of size or complexity, over the years.