Description of problem: We have seen in CI that the same master always returns as leader when going down. In a recent CI run the time master was down for around 2 minutes: https://prow.ci.openshift.org/view/gcs/origin-ci-test/logs/periodic-ci-openshift-release-master-ci-4.8-upgrade-from-stable-4.7-e2e-gcp-ovn-upgrade/1375019219732664320 No other master took over leadership while the previous active leader was down. dcbw found this is because we are not releasing the lock when the ovnkube master exits.
Verified on 4.8.0-0.nightly-2021-04-08-005413 Leader changes after ovnkube-master killed. I0409 02:27:42.941540 1 ovnkube.go:121] Received signal terminated. Shutting down I0409 02:27:42.941694 1 services_controller.go:164] Shutting down controller ovn-lb-controller I0409 02:27:42.941794 1 reflector.go:225] Stopping reflector *v1.Service (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.941875 1 reflector.go:225] Stopping reflector *v1.Pod (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.941944 1 reflector.go:225] Stopping reflector *v1.EgressIP (0s) from github.com/openshift/ovn-kubernetes/go-controller/pkg/crd/egressip/v1/apis/informers/externalversions/factory.go:117 I0409 02:27:42.942193 1 reflector.go:225] Stopping reflector *v1beta1.CustomResourceDefinition (0s) from k8s.io/apiextensions-apiserver/pkg/client/informers/externalversions/factory.go:117 I0409 02:27:42.942240 1 reflector.go:225] Stopping reflector *v1.NetworkPolicy (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.942287 1 reflector.go:225] Stopping reflector *v1.Node (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.942326 1 reflector.go:225] Stopping reflector *v1.Service (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.942358 1 reflector.go:225] Stopping reflector *v1.Endpoints (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.942406 1 reflector.go:225] Stopping reflector *v1.Namespace (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.942445 1 reflector.go:225] Stopping reflector *v1beta1.EndpointSlice (0s) from k8s.io/client-go/informers/factory.go:134 I0409 02:27:42.959646 1 master.go:106] No longer leader; exiting
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 (Moderate: OpenShift Container Platform 4.8.2 bug fix and security update), 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/RHSA-2021:2438