-
Bug
-
Resolution: Done
-
Major
-
None
-
4.18.z
-
Quality / Stability / Reliability
-
False
-
-
None
-
Important
-
None
-
None
-
None
-
CNF RAN Sprint 275
-
1
-
Done
-
Bug Fix
-
-
None
-
None
-
None
-
None
Description of problem:
After the sidecar process cloud-event-proxy being killed, the notification API remaining as clockClass = 0 even when the pod is recovered(recreated).
Version-Release number of selected component (if applicable):
4.18
How reproducible:
Always
Steps to Reproduce:
1. Install SNO with RAN profile configured 2. Create a bounday clock and enable the fast notification API 3. Check PTP clockClass with Event Fast Notification AP/metrics API/pmc command 4. Kill the cloud-event-proxy process in sidecar container of pod linuxptp-daemon, note that don't kill the conmon process. 5. Continously check PTP clockClass with Event Fast Notification API/metrics API/pmc command
Actual results:
3. clockClass is 6 for all the 3 check points. 5. clockClass is 6 in pmc output and metrics API, but in Notification API it reamained as 0(event-not-found) forever, same for lock-state
Expected results:
3. clockClass is 6 for all the 3 check points. 5. clockClass is 6 for all the 3 check points.
Additional info: