Document URL: https://docs.openshift.com/container-platform/4.6/post_installation_configuration/cluster-tasks.html#dr-scenario-2-restoring-cluster-state_post-install-cluster-tasks Section Number and Name: 4c and 4e Describe the issue: In doc section 4c, when checking that respective etcd pod is properly stopped it is stated: "[core@ip-10-0-154-194 ~]$ sudo crictl ps | grep etcd The output of this command should be empty. If it is not empty, wait a few minutes and check again." This could give a false negative if the etcd-operator is running on that node. The ouput will not be empty it will contain etcd-operato. The same issue occurs in section 4e with "sudo crictl ps | grep kube-apiserver". Again if the kube-apiserver-operator pod is running on that node then grep will not return empty as expected. Suggestions for improvement: Instead use commmand: sudo crictl ps | grep 'etcd ' with the space in quotes. This will verify that etcd-operator does not match the grep command. Do the same with kube-apiserver Or you could just specify that the *-operator pods are not important here.
We are leaving this to the documentation team but it may need to be further looked at by the Multi-Arch team
Closing this old bug as I see the documentation has been updated to use the command "sudo crictl ps | grep etcd | grep -v operator". This fixes the problem. Thanks!