Merge lp:~fginther/cupstream2distro-config/update-autopilot-qt-ppa into lp:cupstream2distro-config

[Francis Ginther]

Updated autopilot-qt to dput into ppa:autopilot/unstable.

Updated autopilot-qt to dput into ppa:autopilot/unstable.

[PS Jenkins bot (ps-jenkins)]

PASSED: Continuous integration, rev:454
http://jenkins.qa.ubuntu.com/job/cupstream2distro-config-ci/321/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/cupstream2distro-config-saucy-amd64-ci/119

Click here to trigger a rebuild:
http://s-jenkins:8080/job/cupstream2distro-config-ci/321/rebuild

review: Approve (continuous-integration)

[Francis Ginther (fginther)]

Self approving.

review: Approve

[Didier Roche-Tolomelli (didrocks)]

Francis: is it really needed? Autopilot is already going daily into distro.

As the autopilot runs are not even done before going into that ppa, I'm afraid that delivering that in addition to distro is:
- missing build power for distro
- not focusing of having the autopilot tests failing fixed (which is happening for the past 2 days).

Shouldn't that be removed rather?

[Martin Mrazik (mrazik)]

On 21.6.2013 08:24, Didier Roche wrote:
> Francis: is it really needed? Autopilot is already going daily into distro.

It is going daily into saucy. We want to support people developing on
raring as well (to make it possible for them to use the latest autopilot
rather then the version which is in raring). I don't think there is any
way how to achieve that other than PPA.

--
Martin Mrazik | Product Strategy Quality Lead
Key fingerprint:
0373 C5E5 6A04 39FF 9D06 31B8 B296 2F5D 35FF D83B

[Didier Roche-Tolomelli (didrocks)]

What is the use case for developers to use raring? All our internal upstream should be developping on raring as per company requirements as it's our goal.

For other contributors, we have a version in distro for raring, this one can't add/remove features, don't we support that one already?

[Martin Mrazik (mrazik)]

On 21.6.2013 08:47, Didier Roche wrote:
> What is the use case for developers to use raring? All our internal upstream should be developping on raring as per company requirements as it's our goal.

I assume s/raring/saucy/g...

> For other contributors, we have a version in distro for raring, this one can't add/remove features, don't we support that one already?

One use case are for example the ubuntu core apps developers. I don't
think all of those developers are on saucy and we certainly want them to
use the latest autopilot with touch features, etc.

I'm not sure you are aware but last UDS there were even people asking
for latest autopilot on precise (thin client).

But, is this worth the discussion? Are we in such a position that we can
not afford few builds in launchpad?

[Didier Roche-Tolomelli (didrocks)]

Le 21/06/2013 13:29, Martin Mrazik a écrit :
> On 21.6.2013 08:47, Didier Roche wrote:
>> What is the use case for developers to use raring? All our internal upstream should be developping on raring as per company requirements as it's our goal.
> I assume s/raring/saucy/g...
>
>> For other contributors, we have a version in distro for raring, this one can't add/remove features, don't we support that one already?
> One use case are for example the ubuntu core apps developers. I don't
> think all of those developers are on saucy and we certainly want them to
> use the latest autopilot with touch features, etc.
>
> I'm not sure you are aware but last UDS there were even people asking
> for latest autopilot on precise (thin client).
>
> But, is this worth the discussion? Are we in such a position that we can
> not afford few builds in launchpad?
>
>
>
>

I'm more afraid of having autopilot tests failing right now for 2 days
and not being fixed, making latest revision unavailable to distro. The
fallback is "oh, but it's in the ppa, so fine". Making people not "I
want to deliver my latest software to saucy".

[Martin Mrazik (mrazik)]

On 21.6.2013 14:39, Didier Roche wrote:
> Le 21/06/2013 13:29, Martin Mrazik a écrit :
>> On 21.6.2013 08:47, Didier Roche wrote:

> I'm more afraid of having autopilot tests failing right now for 2 days
> and not being fixed, making latest revision unavailable to distro. The
> fallback is "oh, but it's in the ppa, so fine". Making people not "I
> want to deliver my latest software to saucy".

Which is fairly off-topic for this merge proposal.

Sorry for the failures. I fear we have no clear responsibilities here
(i.e. nobody was appointed to look after this job) which is my fault. I
started a different thread on that.

[Francis Ginther (fginther)]

So I hope the question over the PPA is resolved. The PPA is there primarily to support the community and I don't believe dropping it is the right answer.

[Didier Roche-Tolomelli (didrocks)]

Le 21/06/2013 15:17, Francis Ginther a écrit :
> So I hope the question over the PPA is resolved. The PPA is there primarily to support the community and I don't believe dropping it is the right answer.
As long as it's pushing to raring but not saucy, I'm fine :)

=== modified file 'stacks/head/qa.cfg'
--- stacks/head/qa.cfg	2013-06-20 11:31:12 +0000
+++ stacks/head/qa.cfg	2013-06-20 21:41:33 +0000
@@ -91,7 +91,7 @@
           saucy-armhf:
             archive_artifacts: '**/output/*deb'
       autolanding:
-        ppa_target: ppa:autopilot/ppa
+        ppa_target: ppa:autopilot/unstable # Needed for armhf hardware builds
         distributions: raring,saucy
     dbus-test-runner:
       contact_email: [e-mail]