Alliance - Fotolia
Dynamic thresholds are a powerful and effective way for tools like vRealize Operations Manager -- formerly vCenter Operations Manager -- to adjust automatically in response to a changing data center environment. But alerts and dynamic threshold calculations are only valid when components in the environment are otherwise operating properly. A machine learning tool must see what is "normal" to calculate what would be "abnormal."
Consider what happens when equipment is taken offline for regular maintenance or upgrades. A management tool might see the down server or unavailable application as a serious fault and generate alerts. This is fine if the downtime is unexpected, but it can be annoying when the downtime is part of a regular maintenance window. The problem is even worse with dynamic thresholds. If a tool attempts to calculate dynamic thresholds during a known downtime, the resulting calculations might generate unacceptable thresholds leading to excessive or improper alerts.
VMware's vRealize Operations Manager supports regular maintenance schedules and an on-demand maintenance mode that will inhibit alerts and metric collection involving systems that are offline or performing poorly due to maintenance. For example, you can tell vRealize Operations Manager that a server is usually offline for maintenance every Thursday from 1 a.m. to 4 a.m. Similarly, you can simply tell vRealize that the same server is offline anytime it needs attention.
You can create a new maintenance schedule by selecting Administration and Maintenance Schedules. Opt to add a new schedule, name it, specify the time and day details, save it, and then assign the schedule to the desired server or other resource which vRealize Operations Manager is monitoring. Once a maintenance schedule is created, it can be edited or removed.
There is little practical reason to forego regular maintenance schedules; every data center resource needs periodic attention for patches, upgrades, or an inspection and air filter cleaning. Remember that the goal of a maintenance schedule is to prevent false-positive alerts and erroneous dynamic threshold calculations during regular downtime. There is no penalty if the system doesn't come down during the maintenance period -- or the entire period -- and you can always tell vRealize Operations Manager to ignore the system ad-hoc if more maintenance is needed outside of the scheduled window.
Taming data center disturbances with vRealize Operations Manager
Test your vRealize Suite knowledge
What's new in vRealize Operations Manager 6.3
Dig Deeper on Using monitoring and performance tools with VMware
Related Q&A from Stephen J. Bigelow
Just because software passes functional tests doesn't mean it works. Dig into stress, load, endurance and other performance tests, and their ... Continue Reading
Don't neglect form factor as part of your data center server selection. Instead, figure out what type of environment you need and learn which server ... Continue Reading
Learn how load balancing in the cloud differs from a traditional network traffic distribution, and explore the different services available from AWS,... Continue Reading