crm_resource --restart fails to restart a clone instance with a "resource not running on node" error in a Pacemaker cluster

Issue

• The crm_resource --restart --resource <rsc> --node <node> command successfully restarts an instance of a clone resource on one node. But for the other nodes, the command fails with the error shown below.

# crm_resource --restart --resource stateful --node node2
stateful is not running on node2 and so cannot be restarted