[
https://issues.apache.org/jira/browse/HBASE-15631?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16209727#comment-16209727
]
Andrew Purtell commented on HBASE-15631:
----------------------------------------
bq. The {randomAssignment} stuff in HMaster was truly not needed?
[~lhofhansl] It wasn't covered by any RSGroup test. Removing it had no effect
on test outcomes or performance. I'm not sure why it was needed beyond standard
assignment manager functionality and it is better to avoid introducing any
changes to core unless clearly required and/or commented and/or covered by
tests (as you've rightly pointed out). [~toffer] might remember if it was
added (to handle a wrinkle they found in production? but with no test
coverage?). This feature is going to be advertised as _experimental_ at first.
We know of several improvements needed already and will be performing full
cluster integration tests and piloting the feature ahead of any production
dependency so will be able to push the status from experimental to GA as part
of that process.
> Backport Regionserver Groups (HBASE-6721) to branch-1
> ------------------------------------------------------
>
> Key: HBASE-15631
> URL: https://issues.apache.org/jira/browse/HBASE-15631
> Project: HBase
> Issue Type: New Feature
> Affects Versions: 1.4.0
> Reporter: Francis Liu
> Assignee: Andrew Purtell
> Fix For: 1.4.0, 1.5.0
>
> Attachments: HBASE-15631-branch-1.patch, HBASE-15631-branch-1.patch,
> HBASE-15631.branch-1.patch, HBASE-15631.patch
>
>
> Based on dev list discussion backporting region server group should not be an
> issue as it does not: 1. destabilize the code. 2. cause backward
> incompatibility.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
