proposed-migration now runs/blocks on failed ppc64el tests

[Martin Pitt]

Hello again,

Martin Pitt [2015-08-25  6:55 +0200]:
> I just enabled autopkgtests for armhf, i. e. from now on
> 
>   http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html
> 
> will trigger, show, and block on regressions of armhf tests.
> 
> Since the new cloud infrastructure we do the regression detection per
> architecture, so a failing armhf test which never succeeded will *not*
> block packages. But one which succeeded in the past will now. This
> will ensure that we avoid regressions on armhf.
> 
> These are still run on the same manually maintained hardware
> (cyclops-nodeXX) as we did under Jenkins, in LXC. Proper ScalingStack
> support for armhf is in the works, until then I'll keep a daily eye on
> the cyclopses.
> 
> If this causes too many problems, we can disable it again, but let's
> give this a try. Please let me know if you run into inordinate
> blockers!
> 
> Thanks,
> 
> Martin
> 
> P.S. I'd also like to do that on ppc64el, but the containers
> frequently die through the OOM killer, so they aren't suitable for
> putting back into production right now.

I fixed/robustified this, so all of the above now applies to ppc64el
too. gcc-5 in -proposed helpfully triggered some 350 tests which made
for a nice smoke test. The queue is still at ~ 60, but stability is
looking okay for now.

Thanks,

Martin
-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <https://lists.ubuntu.com/archives/ubuntu-devel-announce/attachments/20150826/2dbb3b1a/attachment.pgp>