#109310 was a good start, but not quite enough to resolve https://github.com/openSUSE/qem-dashboard/issues/61. Incidents might be reused from one day to the next, so only deleting data that is 90 days old does not cover such cases.
- AC1: Whenever the
rr_numberof an incident changes, delete existing openQA jobs in the database.
- AC2: Ensure that there are no performance regressions
- Trigger the rr_number check from the same API that does the 90 day cleanup.
#9 Updated by kraih about 2 months ago
Added a log message to make this easier to verify: https://github.com/openSUSE/qem-dashboard/commit/46a110e2fa205b289f59e89fc4f71050b4123ae4
#10 Updated by kraih about 2 months ago
- Status changed from Feedback to Resolved
The feature is being used in production:
May 17 10:14:15 qam2 dashboard:  [i] Cleaning up old jobs for incident 24247, rr_number change: 272218 -> 272456
Considering there are no complaints yet, we can probably assume it works.