You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hbase.apache.org by Jan Kunigk <ja...@gmail.com> on 2017/08/28 13:02:34 UTC

Question on Hadoop QA

Hi I currently have a patch pending for HBASE-16415.

I am having trouble to successfully executing all tests in the HBase
upstream master branch, while executing my own module tests work fine.

Since I cannot successfully run the tests on master I am having a hard time
to see if my own changes as part of the patch are breaking any other tests.

What is the recommended way of dealing with this?
Is there an RC branch I should be using to qualify my changes?

Best, J

Re: Question on Hadoop QA

Posted by Ted Yu <yu...@gmail.com>.
Jan:
If you look at:
https://builds.apache.org/job/PreCommit-HBASE-Build/8149/artifact/patchprocess/patch-unit-hbase-server.txt

you can see that QA bot excluded known flaky tests:

-Dtest.exclude.pattern=...

However, there might be a few tests which failed that were not related
to your patch (TestSplitTableRegionProcedure in your JIRA).

Hopefully you get a clean run for your next patch.

Cheers


On Mon, Aug 28, 2017 at 6:10 AM, Sean Busbey <bu...@apache.org> wrote:

> Hi Jan!
>
> I'll go take a look. Our flaky list should be pretty robust at this point,
> but sometimes regressions make it into master and our nightly tests aren't
> great at raising the alarm.
>
> On Aug 28, 2017 08:02, "Jan Kunigk" <ja...@gmail.com> wrote:
>
> > Hi I currently have a patch pending for HBASE-16415.
> >
> > I am having trouble to successfully executing all tests in the HBase
> > upstream master branch, while executing my own module tests work fine.
> >
> > Since I cannot successfully run the tests on master I am having a hard
> time
> > to see if my own changes as part of the patch are breaking any other
> tests.
> >
> > What is the recommended way of dealing with this?
> > Is there an RC branch I should be using to qualify my changes?
> >
> > Best, J
> >
>

Re: Question on Hadoop QA

Posted by Sean Busbey <bu...@apache.org>.
Hi Jan!

I'll go take a look. Our flaky list should be pretty robust at this point,
but sometimes regressions make it into master and our nightly tests aren't
great at raising the alarm.

On Aug 28, 2017 08:02, "Jan Kunigk" <ja...@gmail.com> wrote:

> Hi I currently have a patch pending for HBASE-16415.
>
> I am having trouble to successfully executing all tests in the HBase
> upstream master branch, while executing my own module tests work fine.
>
> Since I cannot successfully run the tests on master I am having a hard time
> to see if my own changes as part of the patch are breaking any other tests.
>
> What is the recommended way of dealing with this?
> Is there an RC branch I should be using to qualify my changes?
>
> Best, J
>