Putting UpcomingSprint just in case 4.5 backport PR does not get merged EOW
I was able to reproduce the issue (or a variation ) on 4.5.0-0.nightly-2020-09-14-124053 which should have the fix.
In a 50 worker AWS cluster, 1 minute after running oc adm must-gather, the following alert fired:
Pod openshift-must-gather-gbggn/must-gather-sd5n7 has been in a non-ready state for longer than 15 minutes.
It had not been 15 minutes since I started the must-gather. See pod info below at the time the alert fired.
# oc get pods --all-namespaces | grep must-gather
openshift-must-gather-gbggn must-gather-sd5n7 0/1 Init:0/1 0 86s
> openshift-must-gather-gbggn must-gather-sd5n7 0/1 Init:0/1 0 86s
After applying the fix, must-gather pod has no longer the init container. Can you run `oc get -o yaml` over the must-gather pod?
Moving back ON_QA to try later build.
4.5.0-0.nightly-2020-09-14-124053 is the latest 4.5 nightly and it does not have the fix. Not sure why automation moved it to ON_QA. Moving it back to MODIFIED while awaiting a nightly with the PR