Multiple Alerts Checking For Severity Change Not Working
-
Wednesday, February 20, 2013 6:24 PM
Ok, I have created 2 subscriptions, both checking for "low free space" on the HDDs of our servers. The difference between the subscriptions is the severity. I want an alert directed to email for "warnings" and another alert directed to our pagers for "critical" alerts. When a server moves from Healthy to Warning, I receive the first email, but when the server moves from Warning to Critical I get nothing. The only way I was able to trip this was to manually flip the Resolution State.
Any ideas?
I initially had the subscriptions with 3 criteria; monitor, severity, and resolution state (set to "New"). I tried removing resolution state, but that didn't seem to change anything.
Mark
All Replies
-
Thursday, February 28, 2013 5:03 PM
Hi Mark,
When an alert Changes from Warning to Critical state, Alert GUID will not change and it will be same. If you try to check using SQL Query, you will not see two different alert for the issue in OperationsManager DB. Hence I believe it will not send an email if the Alert State changes from Warning to Critical.
Solution:
---------
Create two different Monitors; one for Warning and another for Critical. By using this method, you will receive two alerts if the issue changes from Warning to Critical and also you will get two email notifications.
Regards, Suresh
-
Thursday, February 28, 2013 5:09 PM
Thank you for the reply. It has been hard to find information on this.
Actually what you suggested doesn't work. That was actually the first thing I tried, since I wanted a different set of subscribers configured for the different alerts. Apparently when the ticket is created, the workflow processes the subscriptions, applies anything that matches, and ignores what doesn't match. When an alert changes from warning to critical, it doesn't see a state change, and does not re-asses the subscription workflows, and thus will not apply the "critical" subscription.
Mark
-
Friday, March 01, 2013 8:11 PM
Hi, But if you really need it there are some workarounds...here some examples
1. Create two different monitors..one for warning and one for critical ..each with its own range
warning when disk space range from 20% to 30%
critical when disk space <20%
the same for free space in MBs
2. You can take advantage of MS Orchestrator with its IP for Operations Manager that has Monitor Alert activity to intercept severity change for alert. It's trigger Updated alerts. Then update alert in order to be processed by SCOM Notification in future..all you need is to build the right logic ..avoiding looping execution
- Marked As Answer by Yog LiMicrosoft Contingent Staff, Moderator Wednesday, March 06, 2013 9:04 AM

