2016-12-22 - production
14:51	<elukey>	restarting the yarn node manager java daemons on all the Hadoop worker nodes due to suspect memory leak	[production]