Filesystem resources restart when a node re-joins in a RHEL 6 or 7 High Availability cluster with pacemaker

Issue

Cluster is configured with multiple filesystem resources. When node A is fenced and re-joins the cluster again the filesystem resources get restarted on node B where it was already activated despite stickiness set (so resources should stay on node B untouched) :

Feb 15 13:03:09 nodeB corosync[2645]: [TOTEM ] A new membership (10.20.30.11:124) was formed. Members joined: 1
Feb 15 13:03:09 nodeB corosync[2645]: [QUORUM] Members[2]: 1 2
Feb 15 13:03:09 nodeB corosync[2645]: [MAIN  ] Completed service synchronization, ready to provide service.
Feb 15 13:03:09 nodeB crmd[4279]:  notice: pcmk_quorum_notification: Node nodeA[1] - state is now member (was lost)
Feb 15 13:03:09 nodeB pacemakerd[4273]:  notice: pcmk_quorum_notification: Node nodeA[1] - state is now member (was lost)
.
.
Feb 15 13:04:17 nodeB pengine[4278]:  notice: Restart FS_resource1#011(Started nodeB)
Feb 15 13:04:17 nodeB pengine[4278]:  notice: Restart FS_resource2#011(Started nodeB)
Feb 15 13:04:17 nodeB pengine[4278]:  notice: Restart FS_resource3(Started nodeB)
Feb 15 13:04:17 nodeB pengine[4278]:  notice: Restart FS_resource4#011(Started nodeB)