by Jonas Lenntun, on May 19, 2020 1:15:00 PM
In some scenarios, you might want to have the possibility to get notified if all of a sudden, a large volume of alerts get raised inside Operations Manager.
It could be that a large amount of alerts gets raised with low severity that usually isn't a problem on an individual occurrence. Maybe a VMware datastore latency alert is configured as a warning because they might fire off during backup windows.
But what if all of the datastores all of a sudden indicates there is a latency issue and raises an alert. Maybe that might suggest that there is a more significant underlying issue in your infrastructure that needs more urgent attention.
We at OpsLogix have addressed this issue and created a simple yet powerful free management pack to solve this.
The OpsMgr Alert Storm Monitoring Management Pack periodically checks and counts the number of alerts in Operations Manager during the last period.
If the alert rate exceeds a defined threshold, an alert gets raised. The purpose of this is to detect situations when an abnormal number of alerts gets generated in Operations Manager, and immediately notify administrators.
Two thresholds can be defined: one for all type of alerts (regardless of source/workflow), and one for multiple alerts (identical workflow alerts from different sources).
Alerts of all levels (Information, Warning and Critical) are included when checking the alert rate. It is also possible to detect and include repeated alerts.