The most effective Monitoring Solutions are comprised of multiple Monitoring Applications. For example, Mobieus™ has three operational tiers: 1) Real-Time Presence 2) Active Intelligence and 3) Data Analytics and each tier is powered by multiple monitoring applications. They work in concert to deliver the Mobieus™ Monitoring Network. I am the first to say it would be so much easier with a single monitoring application that could just do it all but in today’s heterogeneous environments that’s just not realistic. There are some terrific blogs out recently on this specific subject. For example check out this one from LogicMonitor: 5 Common Mistakes in Application/IT Infrastructure Monitoring. It got me thinking and so I wanted to share my thoughts on the subject. The following five characteristics will go a long way to ensuring your Monitoring Solution develops into and remains a valuable IT asset:
- Automatically Identify Changes in Health State – Monitoring applications are best known for alerting on problems and providing the health (Critical, Warning, and Healthy) of individual objects. For example, a logical disk object that is inaccessible will go from a healthy state to a critical state and an Alert will be generated. A good monitoring application will be able to tell when the disk becomes accessible again, automatically reset the health state and close the alert without any user intervention. This ensures the problem has been fixed (temporarily or permanently – that is an entirely different discussion) and frees up the operator from having to manually close the alert.
- Don’t Forget the Hypervisor Host – Visualization is everywhere but it is not uncommon to overlook the monitoring of the host infrastructure. Your monitoring solution for visualization is only ½ complete if only focusing on guests. The hosts and preferably both storage and networking should all be included.
- Don’t Forget the Hardware – Server hardware continues to become substantially more sophisticated. Environmental, Intrusion Detection and physical attributes of Disks, CPU, Controllers and Memory should also all be included in your monitoring solution.
- High Availability – Monitoring solutions are expected to reliable watch over mission-critical applications so it is only seems reasonable that they are configured for high availability and are part of a tested DR plan. Consider a disaster or a cascading service disruption – you will be able to confidently rely on your monitoring solution to verify all systems have returned to normal.
- Proactive – In today’s highly competitive marketplace, you’ve already fallen behind if your monitoring solution is primarily reactionary. Proactive monitoring is required in order to achieve the highest possible availability for mission-critical applications, deliver the fastest possible performance and most importantly, keep your customers happy – or at least away from the Help Desk.