Kube-apiserver CO degraded due to failed conditional check (ConfigObservationDegraded)

Issue

• The kube-apiserver CO is degraded:

$ oc get co kube-apiserver
NAME             VERSION   AVAILABLE   PROGRESSING   DEGRADED   SINCE
kube-apiserver   4.8.x     True        False         True       30h
===========================================================                                             
  Conditions:
    Last Transition Time:  2021-08-25T04:24:23Z
    Message:               ConfigObservationDegraded: configmaps openshift-etcd/etcd-endpoints: no etcd endpoint addresses found
    Reason:                ConfigObservation_Error
    Status:                True
    Type:                  Degraded

• The kube-apiserver-operator pod logs throws the following error, in spite of the fact that etcd-endpoints config map is well populated with the master node IPs.

E0830 04:06:28.282295       1 base_controller.go:266] ConfigObserver reconciliation failed: configmaps openshift-etcd/etcd-endpoints: no etcd endpoint addresses found