Description of problem: OLM emits the `csv_succeeded` and `csv_abnormal` metrics, that are used in various telemeter dashboards to answer numerous questions using data aggregated from all live openshift clusters. When the olm pod restarts, the `csv_succeeded` and `csv_abnormal` metrics are not persisted, which means the data being displayed across the org is inaccurate. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Install an operator 2. Inspect metrics emitted by olm operator pod to see the `csv_succeeded` (or csv_abnormal metric for the installed operator 3. Delete the olm pod and let a new pod get created 4. Inspect metrics emitted by the new olm operator pod Actual results: No `csv_succeeded` or `csv_abnormal` metrics Expected results: Metrics related to CSVs are persisted across pod restart. Additional info:
Looks like this issue was already reported in 1952576. Closing this as duplicate *** This bug has been marked as a duplicate of bug 1952576 ***