[
https://issues.apache.org/jira/browse/HBASE-4809?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13154285#comment-13154285
]
Phabricator commented on HBASE-4809:
------------------------------------
nspiegelberg has accepted the revision "[jira] [HBASE-4809] Per-CF set RPC
metrics".
could you put the test under TestHeapSize & I'll commit
INLINE COMMENTS
src/test/java/org/apache/hadoop/hbase/regionserver/metrics/TestSchemaMetrics.java:220-222
I think it's better to be consistent than optimal. Right now, heapsize is
easy to refactor because it's done the same way for all classes. I'm +1 on a
heapsize refactor, but I say we put that in another JIRA. It's easier to
review 2 JIRAs, feature + refactor, than it is to combine the two.
REVISION DETAIL
https://reviews.facebook.net/D483
> Per-CF set RPC metrics
> ----------------------
>
> Key: HBASE-4809
> URL: https://issues.apache.org/jira/browse/HBASE-4809
> Project: HBase
> Issue Type: Improvement
> Reporter: Mikhail Bautin
> Assignee: Mikhail Bautin
> Priority: Minor
> Attachments: D483.1.patch, D483.2.patch, D483.3.patch,
> HBASE-4809_Per_CF_set_RPC_metrics.patch
>
>
> Porting per-CF set metrics for RPC times and response sizes from 0.89-fb to
> trunk. For each "mutation signature" (a set of column families involved in an
> RPC request) we increment several metrics, allowing to monitor access
> patterns. We deal with guarding against an explosion of the number of
> metrics in HBASE-4638 (which might even be implemented as part of this JIRA).
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
