[jira] [Commented] (CASSANDRA-10464) "nodetool compactionhistory" output should be sorted on compacted_at column and the timestamp shown in human readable format

["Michael Edge (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-10464?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15008177#comment-15008177 ] 

Michael Edge commented on CASSANDRA-10464:
------------------------------------------

Patch attached. I made the changes in CompactionHistory.java to ensure only NodeTool is impacted. It would have been easier to change CompactionHistoryTabularData.java, since this is where the data retrieved from the system table is formatted, but I was unsure whether this would impact any other consumers or JMX users/apps.

My first Cassandra patch - please be gentle...

> "nodetool compactionhistory" output should be sorted on compacted_at column and the timestamp shown in human readable format
> ----------------------------------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-10464
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-10464
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Wei Deng
>            Priority: Minor
>             Fix For: 3.x
>
>         Attachments: CASSANDRA-10464-CompactionHistory.patch
>
>
> "nodetool compactionhistory" (introduced in CASSANDRA-5078) is a useful tool for Cassandra DBAs. However, the current output limits its usefulness without some additional parsing.
> We should improve it in the following two areas:
> 1. The output should be sorted on the compacted_at column, so that the most recently finished compaction will show up last (which is what the DBAs would expect);
> 2. The compacted_at column should be printed in human-readable timestamp.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)