Description of problem: during install the bootstrapIP is written to an annotation on the etcd-endpoints configmap[1]. This is intended to be used by master nodes so they can communicate with etcd on the temporary control-plane. The bootstrap teardown controller[2] will continuously fail because it does not have the conditions to remove the member as it is only member in the cluster. The solution is to omit the bootstrap annotation for single node during render and instead add it as an endpoint directly.[3] [1] https://github.com/openshift/cluster-etcd-operator/blob/release-4.8/bindata/bootkube/manifests/00_etcd-endpoints-cm.yaml [2] https://github.com/openshift/cluster-etcd-operator/blob/master/pkg/operator/bootstrapteardown/bootstrap_teardown_controller.go#L140 [3] https://github.com/openshift/cluster-etcd-operator/blob/master/pkg/operator/etcdendpointscontroller/etcdendpointscontroller.go#L136 Version-Release number of selected component (if applicable): How reproducible: 100% Steps to Reproduce: 1. install a single node cluster 2. 3. Actual results: bootstrap endpoint is never removed. Expected results: bootstrap endpoint is never added and instead endpoint configmap is properly seeded during render. Additional info:
*** Bug 1917115 has been marked as a duplicate of this bug. ***
*** Bug 1932206 has been marked as a duplicate of this bug. ***
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