hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Purtell <apurt...@apache.org>
Subject Re: On filing JIRAs for failed ASF Jenkins tests
Date Mon, 25 Mar 2013 19:04:43 GMT
Simply pointing out a test failure does not add anything beyond what we can
find out from Jenkins. Pasting a log snippet does not add anything beyond
what we can find out from looking at the log. Posting a source code snippet
does not add anything beyond what we Google for ourselves. This comment
adds a marginally useful hypothesis about the generic cause but nothing
about what specifically in HBase is the problem. So apparently that is left
up to someone else no matter who the issue is assigned to.


On Mon, Mar 25, 2013 at 7:54 PM, Ted Yu <yuzhihong@gmail.com> wrote:

> HBASE-7871 was assigned to me.
>
> Were you referring to another JIRA ?
>
> On Mon, Mar 25, 2013 at 11:47 AM, Andrew Purtell <apurtell@apache.org
> >wrote:
>
> > Yes, that's good, but the pasted source is not necessary, the useful bit
> is
> > "This means that concurrent deregister() resulted in the NPE." It would
> be
> > nice, since the hypothesis is unexpected/unprotected concurrency, if the
> > HBase code that could be racing is also identified. Otherwise you are
> just
> > pushing this work onto someone else, why not take it on yourself?
> >
> > On Mon, Mar 25, 2013 at 7:40 PM, Ted Yu <yuzhihong@gmail.com> wrote:
> >
> > > How about this one
> > >
> > >
> >
> https://issues.apache.org/jira/browse/HBASE-7871?focusedCommentId=13612966&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13612966
> > > ?
> > >
> > > On Mon, Mar 25, 2013 at 11:26 AM, Andrew Purtell <apurtell@apache.org
> > > >wrote:
> > >
> > > > No, that is not an analysis. Why did we NPE there? The answer to that
> > > > question would qualify as analysis.
> > > >
> > > >
> > > > On Mon, Mar 25, 2013 at 7:20 PM, Ted Yu <yuzhihong@gmail.com> wrote:
> > > >
> > > > > I agree that analysis of test output for failed test should be
> > > conducted
> > > > > before logging JIRA.
> > > > >
> > > > > Below is an example of analysis for a test failure from yesterday:
> > > > >
> > > > >
> > > > >
> > > >
> > >
> >
> https://issues.apache.org/jira/browse/HBASE-7871?focusedCommentId=13612855&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13612855
> > > > >
> > > > > Cheers
> > > > >
> > > > > On Fri, Mar 22, 2013 at 1:21 PM, Andrew Purtell <
> apurtell@apache.org
> > >
> > > > > wrote:
> > > > >
> > > > > > Filing a JIRA for a flaky test just checked in is not really
what
> > > > > motivated
> > > > > > my original mail. Let me just say if we think opening JIRAs
> > > > > > simply for Jenkins test failures is good practice, then we should
> > > > simply
> > > > > > automate it and have Jenkins do it. My opinion is without detail,
> > > > > analysis,
> > > > > > or patches such JIRAs are noise, they don't add anything to
> reports
> > > we
> > > > > can
> > > > > > pull from build result mails.
> > > > > >
> > > > > > On Friday, March 22, 2013, Sergey Shelukhin wrote:
> > > > > >
> > > > > > > I think filing JIRAs for recently added/recently broken
tests
> > that
> > > > > became
> > > > > > > flaky (outside of jenkins) should be ok, I filed one some
time
> > ago
> > > :)
> > > > > > > Otherwise it's ok to commit a flaky test and then not fix
it
> > > yourself
> > > > > > > because whoever raises a JIRA has to provide a patch himself..
> > > > > > >
> > > > > > > On Thu, Mar 21, 2013 at 6:57 AM, Andrew Purtell <
> > > apurtell@apache.org
> > > > > > <javascript:;>>
> > > > > > > wrote:
> > > > > > >
> > > > > > > > Can we refrain from filing JIRAs for failed ASF Jenkins
tests
> > > until
> > > > > the
> > > > > > > > builds are running faster? At least, let's confirm
the test
> > also
> > > > > fails
> > > > > > > > locally. Preferably the filer also has a patch that
fixes the
> > > > > problem.
> > > > > > > >
> > > > > > > > I may go into JIRA and reap a bunch of issues which
do not
> meet
> > > the
> > > > > > above
> > > > > > > > criteria. Please let me know if you have concerns
about that.
> > > > > > > >
> > > > > > > >
> > > > > > > > --
> > > > > > > > Best regards,
> > > > > > > >
> > > > > > > >    - Andy
> > > > > > > >
> > > > > > > > Problems worthy of attack prove their worth by hitting
back.
> -
> > > Piet
> > > > > > Hein
> > > > > > > > (via Tom White)
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > Best regards,
> > > > > >
> > > > > >    - Andy
> > > > > >
> > > > > > Problems worthy of attack prove their worth by hitting back.
-
> Piet
> > > > Hein
> > > > > > (via Tom White)
> > > > > >
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Best regards,
> > > >
> > > >    - Andy
> > > >
> > > > Problems worthy of attack prove their worth by hitting back. - Piet
> > Hein
> > > > (via Tom White)
> > > >
> > >
> >
> >
> >
> > --
> > Best regards,
> >
> >    - Andy
> >
> > Problems worthy of attack prove their worth by hitting back. - Piet Hein
> > (via Tom White)
> >
>



-- 
Best regards,

   - Andy

Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message