CrashLoopBackOff of certified-operators & community-operators pods after cluster upgrade

Issue

• After cluster upgrade to 4.4.x, community and certified operators pods are continuosly crashing with Liveness and/or Readiness probes failures:

NAME                                    READY   STATUS             RESTARTS   AGE
certified-operators-5bcb56768c-64fxs    0/1     CrashLoopBackOff   27         118m
certified-operators-cddd74b58-k86fv     0/1     Running            6          14m
community-operators-698654bb96-zd4s6    0/1     CrashLoopBackOff   13         51m
community-operators-786f694c8d-gl7bj    0/1     Running            6          14m
marketplace-operator-7c4959c648-fwmn7   1/1     Running            0          15m
redhat-marketplace-5874897f8f-527hz     1/1     Running            0          14m
redhat-operators-7d877d5977-jp8wz       1/1     Running            0          14m

Events:
  Type     Reason     Age                    From                                                                     Message
  ----     ------     ----                   ----                                                                     -------
  Normal   Scheduled  51m                    default-scheduler                                                        Successfully assigned openshift-marketplace/community-operators-698654bb96-zd4s6 to node01.example.com
  Normal   Started    51m                    kubelet, node01.example.com  Started container community-operators
  Warning  Unhealthy  49m (x9 over 51m)      kubelet, node01.example.com  Readiness probe failed: timeout: failed to connect service "localhost:50051" within 1s
  Normal   Created    49m (x2 over 51m)      kubelet, node01.example.com  Created container community-operators
  Normal   Killing    49m                    kubelet, node01.example.com  Container community-operators failed liveness probe, will be restarted
  Normal   Pulled     26m (x9 over 51m)      kubelet, node01.example.com  Container image "quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:821853c24977f49986d51cf2a3756dc3d067fc3122c27ef60db9445f67d66c5c" already present on machine
  Warning  Unhealthy  6m46s (x125 over 51m)  kubelet, node01.example.com  Liveness probe failed: timeout: failed to connect service "localhost:50051" within 1s
  Warning  BackOff    106s (x98 over 36m)    kubelet, node01.example.com  Back-off restarting failed container

• The logs of the failing pods do not show errors, just packages being downloading.