[
https://issues.apache.org/jira/browse/HBASE-20111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16386427#comment-16386427
]
Josh Elser commented on HBASE-20111:
------------------------------------
{quote}Currently when we are not able split or any failures we are retrying the
split so when we can't split the region better not to retry that's why changed
to DNRIOE.
{quote}
Ok, so this is implicitly getting retried client-side when, for Phoenix, we'd
just want it to fail-fast, right?
I think that this change is at odds with a situation like the following:
Balancer moves Region1, user issues a split on a key which falls in the range
of Region1, split fails because Region1 is closed (being moved), split is
retried and succeeds after that Region1 is re-opened on the new server. With
your change, it would require users to re-issue the split since the retry
wouldn't happen.
I think that would mean that we have to differentiate splitPolicy decisions
that we know will never allow splits to happen from our normal {{isSplittable}}
checks.
> Able to split region explicitly even on shouldSplit return false from split
> policy
> ----------------------------------------------------------------------------------
>
> Key: HBASE-20111
> URL: https://issues.apache.org/jira/browse/HBASE-20111
> Project: HBase
> Issue Type: Bug
> Reporter: Rajeshbabu Chintaguntla
> Assignee: Rajeshbabu Chintaguntla
> Priority: Critical
> Fix For: 2.0.0
>
> Attachments: HBASE-20111.patch, HBASE-20111_test.patch
>
>
> Currently able to split the region explicitly even when the split policy
> returns from shouldSplit.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
