As usual we have to cut branches for the upstream repositories that we maintain and bump every downstream fork to the latest stable version (unless we've identified a blocking regression).

The work can be tracked in https://github.com/openshift/cluster-monitoring-operator/issues/1107.