nodetool repair consuming large disk when it ran after increasing replication factor on live cluster ----------------------------------------------------------------------------------------------------
Key: CASSANDRA-3109 URL: https://issues.apache.org/jira/browse/CASSANDRA-3109 Project: Cassandra Issue Type: Bug Components: Tools Affects Versions: 0.7.4 Environment: Centos 5.5 Reporter: Sunil Kumar Hi all, I have two node cassandra cluster initial replication factor for this cluster is 1. When i changed replication factor to 2 on this live cluster.From cassandra wiki(http://wiki.apache.org/cassandra/Operations#Replication) it is recommended to run nodetool repair command on each node of the cluster. But when i run nodetool command on both node disk usage on the both node got increased to 50 times of the initial disk usage. Initially in my cluster data with one replication factor was 1 GB on each node now its 50GB on each node and still nodetool process is going on. Can any body help me in this problem or is it desired behavior of cassandra??? -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira