Looks like underlying issue for all the failing tests is same. Looks like
one change/commit caused failures in all these tests.
java.io.IOException: Shutting down
at
org.apache.hadoop.hbase.regionserver.TestHRegionServerBulkLoad.testAtomicBulkLoad(TestHRegionServerBulkLoad.java:347)
Caused by: java.lang.RuntimeException: Master not initialized after
200000ms seconds
at
org.apache.hadoop.hbase.regionserver.TestHRegionServerBulkLoad.testAtomicBulkLoad(TestHRegionServerBulkLoad.java:347)
Created https://issues.apache.org/jira/browse/HBASE-23000 for tracking
purpose and assigned to myself.
Rushabh Shah
- Software Engineering SMTS | Salesforce
-
- Mobile: 213 422 9052
On Sun, Sep 8, 2019 at 6:47 PM Andrew Purtell <andrew.purt...@gmail.com>
wrote:
> File a JIRA for each constant failure? Then we can divide up the work. At
> the least, the task of bisecting to find the breaking change.
>
> > On Sep 8, 2019, at 6:00 PM, Sean Busbey <bus...@apache.org> wrote:
> >
> > I've filed HBASE-22991 to track a 1.4 release and have started the
> process.
> >
> > For others looking to help out, branch-1.3 has a few tests that show up
> as
> > always failing according to the flakey report:
> >
> >
> https://builds.apache.org/view/H-L/view/HBase/job/HBase-Find-Flaky-Tests/job/branch-1.3/Flaky_20Test_20Report/
> >
> >
> >> On Fri, Sep 6, 2019, 20:53 Andrew Purtell <apurt...@apache.org> wrote:
> >>
> >> I have been pretty busy with work and life so haven't had the bandwidth
> to
> >> make the next set of 1.3 and 1.4 releases, nor another 1.5.0 candidate
> >> either. I have started a few times. I will get as far as observing some
> >> issues still in progress targeting the respective release(s) on JIRA, or
> >> finding that looped unit suite execution reveals a flaky unit test or
> two
> >> deserving of a bit of investigation.
> >>
> >> Next week I am heading out for vacation for two weeks. Therefore It is
> >> unlikely I will have any time in the entire month of September for
> >> branch-1.3, branch-1.4, or branch-1 release candidates.
> >>
> >> If someone else would like to contribute time and attention to release
> >> management, especially for branch-1.3 (1.3.6) and branch-1.4 (1.4.11),
> that
> >> would be great. Both of these branches need releasing to get the bug fix
> >> out for those code lines where recovered.edits files may be misplaced
> when
> >> root FS != wal FS.
> >>
> >> If not I'll spin a set when I return, end of September maybe, or early
> >> October.
> >>
> >> --
> >> Best regards,
> >> Andrew
> >>
> >> Words like orphans lost among the crosstalk, meaning torn from truth's
> >> decrepit hands
> >> - A23, Crosstalk
> >>
>
