Major Description of problem:
Able to see duplicated pods and in crashloopbackoff in the openshift-marketplace
$ oc get pods -n openshift-marketplace NAME READY STATUS RESTARTS AGE 5799beea233b33b6f3b550ce9d4386be1ce0eaf18538c6e51a4da28265vhkst 0/1 Completed 0 1d certified-operators-flklx 0/1 CrashLoopBackOff 8 34m certified-operators-zlpxd 0/1 CrashLoopBackOff 8 34m community-operators-cqbjn 0/1 CrashLoopBackOff 8 33m community-operators-s7gxf 0/1 CrashLoopBackOff 8 33m marketplace-operator-5fd8df5986-8tszf 1/1 Running 0 33m redhat-marketplace-2fmk5 0/1 CrashLoopBackOff 6 23m redhat-marketplace-8xkbg 0/1 CrashLoopBackOff 8 33m redhat-operators-9lddb 0/1 CrashLoopBackOff 8 33m redhat-operators-blbcf 0/1 CrashLoopBackOff 8 34m
Also, the `catalog-operator` pod is restarting after giving an stacktrace with the error concurrent map writes
$ oc get pods -n openshift-operator-lifecycle-manager NAME READY STATUS RESTARTS AGE catalog-operator-f85c49c5-m6b4d 1/1 Running 277 11d $ oc logs catalog-operator-f85c49c5-m6b4d -p -n openshift-operator-lifecycle-manager ... 2023-09-25T07:25:36.269842917Z time="2023-09-25T07:25:36Z" level=info msg=syncing event=update reconciling="*v1alpha1.Subscription" selflink= 2023-09-25T07:25:36.270101523Z fatal error: concurrent map writes 2023-09-25T07:25:36.273353697Z 2023-09-25T07:25:36.273353697Z goroutine 638 [running]: 2023-09-25T07:25:36.273353697Z runtime.throw({2023-09-25T07:25:36.273407198Z 0x1ee9827, 0xc003a645a8}) 2023-09-25T07:25:36.273407198Z /usr/lib/golang/src/runtime/panic.go:1198 +0x71 fp=0xc003a64550 sp=0xc003a64520 pc=0x43bcd1 2023-09-25T07:25:36.273407198Z runtime.mapassign_faststr(0x1be0e40, 0xc006651d40, {0x1f17e43, 0x2e2023-09-25T07:25:36.
Version-Release number of selected component (if applicable):
$ oc get clusterversion NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.10.41 True False 11d Cluster version is 4.10.41
How reproducible:
No able to reproduce
Steps to Reproduce:
N/A
Actual results:
Two pods for the certified-operators/community-operators/redhat-marketplace/redhat-operators in the project openshift-marketplace exists and also the catalog-operator pod is giving an stacktrace
Expected results:
Only a pod exists for the certified-operators/community-operators/redhat-marketplace/redhat-operators and the catalog-operator doesn't give an stacktrace
Additional info:
Even when this is an OCP 4.10 out of Maintenance support, this bug is filled before starting the upgrade activity, then, the idea is to be able to move further to be able to upgrade to OCP 4.11.
