Prevent ITOM Cloud discovery to discover Retired machines
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
an hour ago - last edited an hour ago
Hi Experts,
There are many threads and explanation documentation , but opening a fresh one to understand as what would be the best way to stop discovering Non-operational / Retired servers from cloud.
We have discovery in place but
This consumes license and hence need to restrict .
Is scripting needed here to avoid insert of new CI (retired) or in CI CLASS MANAGER the identification rule resolved this problem.
New to RTE so if anyone can suggest
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
an hour ago
When discovering cloud resources, retired or non-operational servers can still get discovered because Discovery identifies whatever exists in the cloud provider inventory/API. The Identification Rules in CI Class Manager only help determine whether to update an existing CI or create a new one — they do not stop Discovery from finding retired resources.
To avoid consuming licenses unnecessarily, the best approach is usually to filter or exclude those resources before they are inserted/updated in CMDB. This can be handled in a few ways depending on your cloud discovery setup.
For most implementations, the preferred approach is:
- Use Cloud Discovery filters, schedules, or tags to exclude retired instances.
- Mark retired resources properly in the cloud provider (AWS/Azure/GCP) using tags or status.
- Configure Discovery patterns or identification logic to ignore instances with states like terminated, stopped, decommissioned, or retired.
If Discovery is still inserting retired CIs, then scripting/customization may be needed
However, scripting should be the last option because it increases maintenance effort during upgrades.
as per my knowledge Identification Rules in CI Class Manager will NOT fully solve license consumption issues because Discovery still processes the device before identification occurs. They mainly prevent duplicates, not discovery itself.
