When SharedLocalYieldingClusterLockManager#lock fails to get a lock it does not update the timeout correctly, so it continues to loop longer than it should. This is from BZ 993041, where it removed an exception so it will retry, but failed to reset the timeout in that case (which wasn't needed when it threw an exception).
Verified with EAP 6.4.11.CP.CR1
Retroactively bulk-closing issues from released EAP 6.4 cummulative patches.