[
https://issues.apache.org/jira/browse/CASSANDRA-2129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12991638#comment-12991638
]
Jonathan Ellis commented on CASSANDRA-2129:
-------------------------------------------
Does removetoken force clear up the second problem?
We should allow reducing the nodes in the cluster below the RF count, just as
we allow creating a keyspace with RF greater than the node count. (In both
cases, writes will be rejected until more nodes are added or RF is reduced.)
> removetoken after removetoken rf error fails to work
> ----------------------------------------------------
>
> Key: CASSANDRA-2129
> URL: https://issues.apache.org/jira/browse/CASSANDRA-2129
> Project: Cassandra
> Issue Type: Bug
> Affects Versions: 0.7.0
> Reporter: Mike Bulman
>
> 2 node cluster, a keyspace existed with rf=2. Tried removetoken and got:
> mbulman@ripcord-maverick1:/usr/src/cassandra/tags/cassandra-0.7.0$
> bin/nodetool -h localhost removetoken 159559397954378837828954138596956659794
> Exception in thread "main" java.lang.IllegalStateException: replication
> factor (2) exceeds number of endpoints (1)
> Deleted the keyspace, and tried again:
> mbulman@ripcord-maverick1:/usr/src/cassandra/tags/cassandra-0.7.0$
> bin/nodetool -h localhost removetoken 159559397954378837828954138596956659794
> Exception in thread "main" java.lang.UnsupportedOperationException: This node
> is already processing a removal. Wait for it to complete.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
