Just to check whats going on... You have a dead node 'A', it's turned off and still showing up in the nodetool ring results for other nodes but the token field is empty. And you've tried running nodetool removetoken on any other node in the cluster. Is that correct ?
Can you include the the output of the nodetool ring command and the version you are using ? Thanks Aaron On 29 Jan 2011, at 14:43, buddhasystem wrote: > > It does remove tokens, and the "ring" shows that the problematic node owns 0 > tokens, which is OK. However, it's still there, listed. > > It's not a bug but kind of like a feature -- you can move that node back in > two days later and "move" tokens in same or different way. > > What I wish happened was that API allowed for the nodetool to issue a > command: > > nodetool --host foobar removeempty > > Which would then really scratch the node with zero tokens from the ring, no > questions asked. Even if the flaky node physically disappeared. > > -- > View this message in context: > http://cassandra-user-incubator-apache-org.3065146.n2.nabble.com/Node-going-down-when-streaming-data-what-next-tp5962944p5971851.html > Sent from the cassandra-u...@incubator.apache.org mailing list archive at > Nabble.com.
