A family of System Center products that provide infrastructure monitoring, help ensure the predictable performance and availability of vital applications, and offer comprehensive monitoring for datacenters and cloud, both private and public.
Hi Saiyad,
From your description, I would not try to solve this by keeping one broad Telephony class in the subscription and then trying to exclude Test later. The cleaner approach is to split the monitored instances into two non-overlapping groups and scope the subscriptions to those groups. In SCOM terms, that usually means an instance group for Telephony Services - Test and another for Telephony Services - Prod. Instance groups can contain the actual monitored objects, which is the better fit when the alerts are raised by service instances rather than just by the hosting computers.
If the distinction between Test and Prod is really a server/environment distinction, then populate those two service-instance groups dynamically based on where the service is hosted, for example:
- Telephony service instances hosted by servers in the Test computer group
- Telephony service instances hosted by servers in the Prod computer group
That "contained by / hosted by another group" pattern is a common way to get very granular scoping for notifications, overrides, and views.
Then create:
- one subscription scoped to Telephony Services - Test → Team A
- one subscription scoped to Telephony Services - Prod → Team B
The important part is to make those groups mutually exclusive. I would not scope one subscription to the full Telephony class and the other to Test, because then you create overlap and duplicate notifications. Microsoft’s SCOM notification guidance also calls out that the Scope tab does not support a native "not equal to" or "not a member of" style exclusion, so trying to solve this as an exclusion problem is usually the wrong direction.
If this Test/Prod split is permanent and important beyond notifications, then the more structured long-term design is to create two separate discoveries / classes instead of one shared class, for example Telephony Service - Test and Telephony Service - Prod. That gives you cleaner targeting for monitors, overrides, views, and subscriptions later on. Kevin has also shown an approach for discovering class instances based on computer name when the server split is static and known.
So my recommendation would be: keep the current service class if you want, but introduce two dynamic instance groups of that class, based on the hosting Test vs Prod servers, and then scope the two subscriptions to those groups. That is usually the cleanest and most supportable way to do it.
Hope that helps you out.
Best Regards
Stoyan Chalakov
"If my response was useful, please consider marking it as the answer. It keeps the forum clean, structured, and more helpful for everyone. Thank you for supporting the community."