[
https://issues.apache.org/jira/browse/HBASE-12278?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14176398#comment-14176398
]
Dima Spivak commented on HBASE-12278:
-------------------------------------
Can't we all just get along? :) My two cents, since I'm sorry to say that my
day's delay in posting a patch apparently caused this conflict:
I'm totally fine with someone solving a problem that I noticed but didn't fix
(just a week back, Stack fixed HBASE-12212 for me because I wasn't familiar
with what needed to be done). That said, I can understand frustration when an
established member of the community jumps in on an issue that a new contributor
has put effort toward solving; it definitely has the potential to chill
contributions. Personally, I'm less concerned about getting credit for a patch
than about having my test rig stable, but if I were just starting out on this
project, I could see what happened here as evidence that HBase developers
compete to get patches up first instead of cooperating. This all goes out the
door, of course, if someone had filed an unassigned critical blocker that I
assigned myself to before taking my sweet time to fix, but
[[email protected]], you yourself made a point for me to re-prioritize a
nearly-identical fix (HBASE-12229) as minor, so I was frankly a little taken
aback by how important you thought it was to get this patch up before I had the
chance to.
Lost in all of this is that TestSecureLoadIncrementalHFilesSplitRecovery (jeez,
what a mouthful) still has a race condition. Can someone please push the fix so
that we can have a big group hug and go fix other flakey tests? :)
> Race condition in TestSecureLoadIncrementalHFilesSplitRecovery
> --------------------------------------------------------------
>
> Key: HBASE-12278
> URL: https://issues.apache.org/jira/browse/HBASE-12278
> Project: HBase
> Issue Type: Bug
> Components: test
> Affects Versions: 1.0.0
> Reporter: Dima Spivak
> Assignee: Dima Spivak
> Priority: Minor
> Attachments: 12278-v1.txt, 12278-v2.txt
>
>
> I sometimes get #testGroupOrSplitWhenRegionHoleExistsInMeta to error out on
> me because of a TableNotFoundException. Fix is pretty simple and will involve
> a little bit of test refactoring to get rid of some duplicate behavior that's
> done better in HBaseTestingUtility anyway.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
