incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ian Danforth <idanfo...@numenta.com>
Subject CASSANDRA-2863
Date Tue, 15 Nov 2011 22:12:35 GMT
Re: https://issues.apache.org/jira/browse/CASSANDRA-2863

All,

 I encountered this NPE during a node move. Everything seemed to complete
(no streaming, no further compactions, no further errors in system.log)
however the move command never returned. After several hours I then
cancelled that command, drained the node, stopped it and restarted it,
hoping it would rejoin the ring at the new Token. It didn't, it picked up
the old token.

 My question is, should I forget about 'move' entirely and go back to
decommission -> manually assign token -> restart node?

Thanks in advance,

Ian

Mime
View raw message