durable_writes complaint

gistfile1.txt

(06:13:50 PM) Fredderf: Might have found a bug in 2.1.3 and same behaviour in 2.1.5 but have only debugged 2.1.3. Setting durable_writes on a keyspace is only applied after a restart. When mutations for a column family is about to be applied, the cached instance of the keyspace metadata is read. But the schema mutation for durable_writes hasn't been applied to this cached instance. Anyone knows if this is a known issue?
(06:15:07 PM) snair: Adding a DC cayused Cannot achiecve conistency level one
(06:15:10 PM) snair: eon all nodes
(06:15:12 PM) snair: oplease helpo
(06:15:15 PM) snair: production
(06:15:29 PM) snair: stopped the 3 node cassandra cluster
(06:23:08 PM) weiz2: rcoli: thobbs: according to this http://docs.datastax.com/en/cassandra/2.0/cassandra/tools/toolsDecommission.html, Descriptionhttp://docs.datastax.com/en/cassandra/2.0/cassandra/tools/toolsDecommission.html?scroll=toolsDecommission__description_unique_6
(06:23:08 PM) weiz2: Causes a live node to decommission itself, streaming its data to the next node on the ring.
(06:23:38 PM) weiz2: Decommission will cause the data to be streamed to satisfy the RF
(06:24:09 PM) weiz2: I was referring to your discussion an hour back
(06:25:42 PM) thobbs: Fredderf: that doesn't sound familiar, can you open a jira ticket to look into it?
(06:26:04 PM) thobbs: weiz2: right, but the RF doesn't require streaming, assuming you have already altered the keyspace to remove the replicas in that DC
(06:26:31 PM) thobbs: snair: are your keyspaces using SimpleStrategy or NetworkTopologyStrategy?
(06:27:33 PM) weiz2: I see, your discussion is for decommission the entire DC
(06:27:55 PM) weiz2: for decommission the individual node, streaming will happen
(06:28:49 PM) Fredderf: thobbs: I'll open a ticket.
(06:30:09 PM) thobbs: thanks