[aerogear-dev] How to propagate results for non Travis CI?

[Karel Piwko]

I have an early bird here:
http://hudson.jboss.org/hudson/job/aerogear-unifiedpush-server-integration-tests/

I've asked for Aerogear view that would list Aerogear related jobs.

More comments inline.

Karel

On Tue, 24 Sep 2013 14:06:32 -0300
Douglas Campos <qmx at qmx.me> wrote:

> On Tue, Sep 24, 2013 at 06:30:34PM +0200, Karel Piwko wrote:
> > > > * When to notify? [failure only|unstable|always]
> > > 
> > > Failure only
> > Sorry, I was not explicit here:
> > * Failure: job didn't finish (aborted to to some error)
> > * Unstable: some tests failed
> > 
> > What about specific states, like still-failing or still-unstable?
> 
> The less noise, the better - I think jenkins has a "state changed only"
> notification setting

+1. Too much noise and people will ignore it.

> 
> > > 
> > > > * How often to run tests? [nightly|per commit]
> > > 
> > > This really boils down to what kind of tests are being run, and how busy
> > > is the build-slave queue.
> > 
> > Well, per commit strategy is unrealistic setup when trying to rerun
> > everyting.
> 
> So nightly is the obvious choice, right?

Very likely. Especially if build publisher does not sync to public instance
immediately but with a delay.

> > 
> > > 
> > > > * How to enable developers to easily rerun something? [email|irc]
> > > 
> > > irc would be easier, email safer?
> > 
> > I prefer IRC, however QE does not cover all timezones. In such cases email
> > makes more sense.
> 
> Aaah, will it be manual intervention? email then.
> 

I think that aerogear-issues or aerogear-ci might be a good list for both
sending results and requesting a rerun.