Description of the problem: clusterrolebindings for open-cluster-management-agent keep being recreated every 5 minutes Release version: 2.6.1 Operator snapshot version: OCP version: 4.10.24 Browser Info: Steps to reproduce: 1. attach a cluster successfully to the 2.6.1 instance 2. confirm it completely imported 3. check klusterlet logs Actual results: the logs are full of ~~~ 2022-10-03T08:17:58.183873661Z I1003 08:17:58.183777 1 event.go:285] Event(v1.ObjectReference{Kind:"Namespace", Namespace:"open-cluster-management-agent", Name:"open-cluster-management-agent", UID:"", APIVersion:"v1", ResourceVers ion:"", FieldPath:""}): type: 'Normal' reason: 'ClusterRoleBindingCreated' Created ClusterRoleBinding.rbac.authorization.k8s.io/open-cluster-management:klusterlet-work:agent because it was missing 2022-10-03T08:17:59.665155494Z I1003 08:17:59.665044 1 event.go:285] Event(v1.ObjectReference{Kind:"Namespace", Namespace:"open-cluster-management-agent", Name:"open-cluster-management-agent", UID:"", APIVersion:"v1", ResourceVers ion:"", FieldPath:""}): type: 'Normal' reason: 'RoleBindingCreated' Created RoleBinding.rbac.authorization.k8s.io/open-cluster-management:management:klusterlet-registration:agent -n kube-system because it was missing ~~~ Expected results: information on if the clusterrolebinding has been created succesfully - object versions are stored in the etcd. That means a new record is created per object every 5 minutes unnecessarily Additional info: see private note added later
G2Bsync 1283546786 comment haoqing0110 Wed, 19 Oct 2022 07:20:35 UTC G2Bsync Fix in https://github.com/stolostron/registration-operator/pull/29, plz verify in ACM 2.6.3.
G2Bsync 1323822810 comment oafischer Tue, 22 Nov 2022 15:06:41 UTC G2Bsync Can someone provide a short summary of this fix? Thanks!
the issue was resolved when the upgrade to 2.6.3 was finished properly.