cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (CASSANDRA-5073) Node movement results in excessive streaming
Date Tue, 18 Dec 2012 04:08:14 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-5073?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jonathan Ellis resolved CASSANDRA-5073.
---------------------------------------

    Resolution: Won't Fix

vnodes makes the old node movement code mostly obsolete.  Not worth rewriting to handle a
corner case at this point.
                
> Node movement results in excessive streaming
> --------------------------------------------
>
>                 Key: CASSANDRA-5073
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5073
>             Project: Cassandra
>          Issue Type: Bug
>    Affects Versions: 1.0.8
>            Reporter: Ishaaq Chandy
>
> I have a 3-node cluster where almost all the keyspaces (other than Cassandra's system
and OpsCenter keyspaces) have a replication factor=3, this implies that all nodes have all
the data.
> I needed to add a new node, so I proceeded to rearrange the tokens to make room for it.
My expectation was that node movement would be cheap before I added the new node. Turns out
it wasn't so, despite the fact that each node has all the data, there was a ton of data being
streamed

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message