[
https://issues.apache.org/jira/browse/HBASE-3102?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12925194#action_12925194
]
Gary Helmling commented on HBASE-3102:
--------------------------------------
@Nicolas: I agree that the naming issue isn't critical. It's a bit wonky but
not a big deal. And understood that this is pushed down on us from the Hadoop
metrics limitation.
I'm fine with closing this out and opening a new JIRA (low priority) to address
the naming issue as time allows. Hadoop metrics are already slated for an
overhaul, hopefully in time for 0.22:
https://issues.apache.org/jira/browse/HADOOP-6728
I hope that those changes will make some of these issues a bit easier. Will
dig into the discussion there when I can. Encourage you to do the same if you
see anything.
> Enhance HBase rMetrics for Long-running Stats
> ---------------------------------------------
>
> Key: HBASE-3102
> URL: https://issues.apache.org/jira/browse/HBASE-3102
> Project: HBase
> Issue Type: Improvement
> Components: regionserver
> Reporter: Nicolas Spiegelberg
> Assignee: Nicolas Spiegelberg
> Fix For: 0.90.0
>
> Attachments: HBASE-3102_90.patch
>
>
> Example: a useful metric to observe track is compaction count + duration.
> Since compactions are long running and only happen infrequently, the
> avg/opcount stats are should be reset longer than the default polling period
> (5 sec). In addition to 'hbase.period', we should allow a different duration
> after which long-running metrics should expire. This would also fix our
> existing metrics problem where min/max stats are never reset until the
> process is restarted/upgraded.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
