Description of problem:

After upgrading the cluster from version 4.18.9 to 4.18.27, the MCH became stuck in the Pending state because the search-postgres pod was stuck in the ContainerCreating state, while the other pods in the open-cluster-management namespace were in the Running state.

The pod is unable to start due to a multi-attach issue. Since the PVC access mode is ReadWriteOnce (RWO), the volume can be mounted by only one pod at a time.

This issue occurs consistently whenever a cluster upgrade is performed.