hadoop-mapreduce-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Konstantin Boudnik <...@apache.org>
Subject Re: [Vote] Merge branch-trunk-win to trunk
Date Sat, 02 Mar 2013 03:33:49 GMT
Suresh, I appreciate all the troubles you're going through wrt syncing up the
huge patch for a long time - I really do.

I am not asking to have full test-patch process in place. But I think it is a
real good idea to have a way to run the full test suite once in a while - or
as Konstantin proposing - for a given patch, to make sure that codebase
doesn't bitrot at the edges.

"Official" support has nothing to do with the issue - you're trying to build a
straw man argument around this. What is relevant, on the other hand, is that
Windows is so different from _any_ Unix or pseudo-Unix flavors, including
Windows with Cygwin - that even multi-platform Java has hard hard time
dealing with it. This is enough, IMO, to warrant a separate checkpoint.

I hope I have explained myself better.

On Fri, Mar 01, 2013 at 05:55PM, Suresh Srinivas wrote:
> > It seems that with the HW in place, the matter of setting at least nightly
> > build is trivial for anyone with up to date Windows knowledge. I wish I
> > could
> > help. Going without a validation is a recipe for a disaster IMO.
> >
> > -1 until some reasonable solution is implemented.
> >   Cos
> Cos, I have hard time understanding your veto?
> Here is my rationale for merge:
> Currently with all the cross platform support, the merge patch has +1 from
> Jenkins on Linux. So no regression has been introduced in Hadoop on Linux.
> As regards to Windows support I want to make two points:
> 1. Until Jenkins is setup and we are passing all the tests, I am okay, as
> Konstntin proposed, if we do not officially declare Windows as supported. I
> do not want to tie the patch merge with setting up Windows Jenkins. I have
> been maintaining this branch for a long time and keeping it in sync with
> trunk is non-trivial.
> 2. After Jenkins is setup, there is a concern in the community about -1
> from Windows hindering patch commits. As others have already suggested in
> the thread, I am okay committing new patches even if -1 is posted by
> Jenkins on Windows. The team that worked on Windows will fix the issues. I
> do not see many such issues cropping up.

View raw message