Re: [DISCUSS] stabilizing Hadoop releases wrt. downstream

On Wed, Mar 06, 2013 at 07:24AM, Arun C Murthy wrote:
> Cos,
> 
> On Mar 4, 2013, at 10:15 PM, Konstantin Boudnik wrote:
> 
> > The issue that a downstream
> > integration project is likely to have is - for once - the absence of
> > regularly published development artifacts. In the light of "it didn't
> > happen if there's no picture" here's a couple of examples:
> 
>  Maybe I wasn't clear, apologies, my request is simpler: Can Bigtop validate a Hadoop RC?

No offense taken. I think this is indeed the intention of initial and
follow-up emails. There need to be a bit more coordination between the
two, so BigTop's branches can be set in advance to churn stack builds
and validations with underlying Hadoop RCs or even before; and Hadoop
RM - and potentially downstream components - have more time to react
to the problems.

One of the issues here is a difference in the release schedule between
the two. Perhaps we can setup a validation branch in the bigtop to
help with releases. This is something that needs to be decided within
the project apparently, but the it might work.

Cos

>  Every RC has it's artifacts in staging at the ASF maven repo. In the past we have had Bigtop actually verifying an RC:
>  http://s.apache.org/sjz
> 
>  If Bigtop can do that consistently, it would definitely help. Agree?
> 
> thanks,
> Arun
>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iF4EAREIAAYFAlE3iLUACgkQenyFlstYjhLCgAEAq7e7Pj9Yw5kYhCvXs93gW243
3n+UQePPwvWYbq0ygxIA/1hmvquk1NQyDF1SO/SAbQCiu1+wcViLiTTRB6sGbUZP
=0EMS
-----END PGP SIGNATURE-----