So in our cluster of 10 nodes we have 2 bad nodes that are in disagreement with the others. On one of the two bad nodes I tried to move the Schema* and Migration* files out of the system data directory as is listed in the http://wiki.apache.org/cassandra/FAQ#schema_disagreement FAQ . When the node started back up it threw thousands of exceptions and it still did not agree with the 8 nodes with the correct schema. Instead it was listed in "UNREACHABLE NODES". Nodetool ring from the bad node and any other node indicated that it was up. I restarted Cassandra on the bad node again and when it came back up it once again had the bad schema but was no longer listed as unreachable.
Tips? All I can really think of right now is to replace the two bad nodes with two totally new ones though I'd definitely prefer to avoid that. -- View this message in context: http://cassandra-user-incubator-apache-org.3065146.n2.nabble.com/Schema-disagreement-in-1-0-2-tp7098609p7098609.html Sent from the cassandra-u...@incubator.apache.org mailing list archive at Nabble.com.
