2017-02-01 - production
16:20	<elukey>	restarting Yarn Node Manager daemons on all the Hadoop nodes to bandaid a memory leak causing OOMs	[production]