hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Yu <yuzhih...@gmail.com>
Subject Re: [NOTICE] PreCommit now running against Hadoop3 artifacts successfully
Date Sun, 01 Oct 2017 23:35:22 GMT
It would be nice if the link is included in QA report when compilation
against hadoop 3 fails:

https://issues.apache.org/jira/browse/HBASE-17441?focusedCommentId=16186839&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-16186839

Thanks

On Sun, Oct 1, 2017 at 4:28 PM, Sean Busbey <busbey@apache.org> wrote:

> The output of each hadoop-specific build are available in the build's
> artifacts. They show up in the 'patchprocess' directory with the name
> 'patch-javac-%hadoop_version%.txt'.
>
> So the hadoop-3 build is here:
>
> https://builds.apache.org/job/PreCommit-HBASE-Build/8875/
> artifact/patchprocess/patch-javac-3.0.0-alpha4.txt
>
> these aren't super intuitive names, maybe you could update them to
> include 'hadoopcheck' or something like that?
>
> On Sun, Oct 1, 2017 at 4:01 PM, Ted Yu <yuzhihong@gmail.com> wrote:
> > I was looking under https://builds.apache.org/job/
> PreCommit-HBASE-Build/8875
> > but didn't seem to find the output which shows the compilation error
> > against hadoop-3.
> >
> > Do you know if the output can be retrieved from Jenkins or, does
> developer
> > have to perform build against hadoop-3 locally ?
> >
> > Thanks, Josh.
> >
> > On Sun, Oct 1, 2017 at 1:37 PM, Josh Elser <elserj@apache.org> wrote:
> >
> >> tl;dr You may start seeing some new "failures" WRT Hadoop3 compatibility
> >> with PreCommit builds. They are (very likely) true negatives.
> >>
> >> Sean, Appy, and myself think we got to the bottom of why our PreCommit
> had
> >> let (at least two) patches that broke Hadoop3 compatibility on
> HBASE-17441.
> >> From what I can tell so far, there haven't been too many changes
> upstream,
> >> but thought it would be good to give a heads up to folks (lest someone
> >> wastes time thinking it's just PreCommit being flaky).
> >>
> >> - Josh
> >>
>

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