Cluster processes shut down and node fenced after "CMAN: too many transition restarts - will die" in RHCS 4

Issue

• A node's cluster processes shut down after "Inconsistent cluster view" while a node was rejoining the cluster

Feb 25 12:02:15 node1 kernel: CMAN: node node2 rejoining
Feb 25 12:02:15 node1 kernel: CMAN: Initiating transition, generation 15
Feb 25 12:02:30 node1 kernel: CMAN: Initiating transition, generation 16
Feb 25 12:02:45 node1 kernel: CMAN: Initiating transition, generation 17
Feb 25 12:03:00 node1 kernel: CMAN: Initiating transition, generation 18
Feb 25 12:03:15 node1 kernel: CMAN: Initiating transition, generation 19
Feb 25 12:03:30 node1 kernel: CMAN: Initiating transition, generation 20
Feb 25 12:03:45 node1 kernel: CMAN: Initiating transition, generation 21
Feb 25 12:04:00 node1 kernel: CMAN: Initiating transition, generation 22
Feb 25 12:04:15 node1 kernel: CMAN: Initiating transition, generation 23
Feb 25 12:04:30 node1 kernel: CMAN: Initiating transition, generation 24
Feb 25 12:04:45 node1 kernel: CMAN: Initiating transition, generation 25
Feb 25 12:05:00 node1 kernel: CMAN: too many transition restarts - will die
Feb 25 12:05:00 node1 kernel: CMAN: we are leaving the cluster. Inconsistent cluster view
Feb 25 12:05:00 node1 kernel: WARNING: dlm_emergency_shutdown
Feb 25 12:05:00 node1 kernel: dlm: dlm_lock: no lockspace

• My cluster node was rebooted after this error:

kernel: CMAN: removing node1 from the cluster : Inconsistent cluster view