Description of problem: Update the managementState of openshiftcontrollermanageroperatorconfig with any words will recreate the pods in project: openshift-controller-manager; Update the managementState of OpenShiftAPIServerOperatorConfig with any words will recreate the pods in project: openshift-apiserver. Version-Release number of selected component (if applicable): Cluster version is 4.0.0-0.nightly-2019-01-15-010905 How reproducible: Always Steps to Reproduce: 1. Update the managementState of openshiftcontrollermanageroperatorconfig with any words. 2. Update the managementState of OpenShiftAPIServerOperatorConfig with any words. Actual results: 1. Update succeed and recreate the pods in project: openshift-controller-manager; 2. Update succeed and recreate the pods in project : openshift-apiserver; Expected results: 1~2: Should restrict the value of spec.managementState with "Managed" and "Unmanaged" and should not recreate the pods in the object projects. Additional info:
1~2: Should restrict the value of spec.managementState with "Managed" and "Unmanaged" and should not recreate the pods in the object projects. This will require CR validation on API server side we don't have yet. We can validate on operator and report error condition when you set the field to unknown value.
https://github.com/openshift/library-go/pull/261 https://github.com/openshift/cluster-kube-controller-manager-operator/pull/168 https://github.com/openshift/cluster-kube-apiserver-operator/pull/272 https://github.com/openshift/cluster-openshift-apiserver-operator/pull/147 Moving to QE as these likely merge today.
https://github.com/openshift/cluster-kube-controller-manager-operator/pull/168 This PR is not merged , change the status to modified.
Confirmed with OCP: 4.0.0-0.nightly-2019-02-25-234632, all the master team operators have the event warning: [root@preserved-yinzhou-rhel-1 0221]# oc project openshift-kube-controller-manager-operator Now using project "openshift-kube-controller-manager-operator" on server "https://api.qe-yinzhou.qe.devcluster.openshift.com:6443". [root@preserved-yinzhou-rhel-1 0221]# oc get events LAST SEEN TYPE REASON KIND MESSAGE ... 5m36s Warning ManagementStateUnknown Deployment Unrecognized operator management state "Managed!@#" 46m Normal LeaderElection ConfigMap b573d757-396d-11e9-8835-0a580a820008 became leader [root@preserved-yinzhou-rhel-1 0221]# oc get po -n openshift-kube-controller-manager NAME READY STATUS RESTARTS AGE ... kube-controller-manager-ip-10-0-140-63.ap-northeast-1.compute.internal 1/1 Running 3 37m kube-controller-manager-ip-10-0-150-12.ap-northeast-1.compute.internal 1/1 Running 3 36m kube-controller-manager-ip-10-0-160-213.ap-northeast-1.compute.internal 1/1 Running 3 34m revision-pruner-2-ip-10-0-140-63.ap-northeast-1.compute.internal 0/1 Completed 0 45m revision-pruner-3-ip-10-0-140-63.ap-northeast-1.compute.internal 0/1 Completed 0 37m [root@preserved-yinzhou-rhel-1 0221]# oc get clusterversion NAME VERSION AVAILABLE PROGRESSING SINCE STATUS version 4.0.0-0.nightly-2019-02-25-234632 True False 26m Cluster version is 4.0.0-0.nightly-2019-02-25-234632
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:0758