[ https://issues.apache.org/jira/browse/CASSANDRA-1553?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12933992#action_12933992 ]
Brandon Williams commented on CASSANDRA-1553: --------------------------------------------- Weird output with 0.6: Address Status Load Owns Range Ring 102855431326809200097386204650499263481 10.179.64.227 Up 854 bytes -39.80% 35142115559466303407164061620519096874 |<--| 10.179.65.102 Up 854 bytes 17.72% 65295694165122757235402903158743316735 | | 10.179.111.137Up 854 bytes 22.08% 102855431326809200097386204650499263481 |-->| > warn of imbalanced ranges in nodetool ring > ------------------------------------------ > > Key: CASSANDRA-1553 > URL: https://issues.apache.org/jira/browse/CASSANDRA-1553 > Project: Cassandra > Issue Type: Improvement > Components: Tools > Reporter: Jonathan Ellis > Assignee: Jon Hermes > Priority: Minor > Fix For: 0.6.9, 0.7.0 > > Attachments: 1553-06-2.txt, 1553-06.txt, 1553.txt, ringer > > > I'd like to enhance nodetool ring to show the % of the token range that it is > responsible for according to the partitioner (that is, before considering > replication). For RandomParitioner this is easy. For OPP we can estimate > using describe_splits. I'd also like to issue a warning if any node N has > more than 25% more of the ring than another node M. > (The reason we need to use the range for this instead of load, is load may > include an arbitrary amount of data that no longer belongs to the node, > pending removal by cleanup.) -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.