2015-05-31 - production
17:12	<gwicke>	performed a rolling restart of RESTBase Cassandra nodes to address elevated request error rates apparently related to schema disagreement	[production]