Cluster node reboots unexpectedly with LLT INFO messages in /var/log/messages on another node at the same time

Issue

In a cluster of Red Hat Enterprise Linux 5 or 6 systems, one cluster node is rebooting unexpectedly a couple of times each day. Messages on a healthy node during the reboot do not show a successful fence event. Instead, there are several message similar to the following:

Nov 15 11:41:25 xyz kernel: LLT INFO V-14-1-10205 link 1 (eth5) node 0 in trouble
Nov 15 11:41:30 xyz kernel: LLT INFO V-14-1-10205 link 2 (eth1) node 0 in trouble
Nov 15 11:41:30 xyz kernel: LLT INFO V-14-1-10032 link 1 (eth5) node 0 inactive 8 sec (13316333)
Nov 15 11:41:31 xyz kernel: LLT INFO V-14-1-10032 link 1 (eth5) node 0 inactive 9 sec (13316333)
Nov 15 11:41:32 xyz kernel: LLT INFO V-14-1-10032 link 1 (eth5) node 0 inactive 10 sec (13316334)