Re: Running kickstart-tests in openQA: should we?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, 2016-04-12 at 11:19 -0700, Brian C. Lane wrote:
> 
> > > If you can automate all the things I'd say it's useful to have openQA
> > > run these.
> > > 
> > > I don't want to see the current system go away though, it is pretty
> > > useful to be able to run the tests locally before pushing big changes
> > > out to the world.
> > So to understand the local workflow: you have to build an image with
> > all your changes and run the tests on that, right?
> > 
> > It would be possible to do this with openQA, even without having your
> > own pet openQA instance - though note it's also quite easy to set up a
> > pet openQA instance, you don't have to do a full-fat bare metal
> > deployment like openqa.happyassassin.net , we have a documented
> > workflow for doing it in docker containers:
> > 
> > https://bitbucket.org/rajcze/openqa_fedora_tools/src/a8c83b6929bb4935697b34ae36f56e0d0dae96ce/docker/?at=develop
> > 
> > this is how jskladan and garretraziel work, so it's a pretty safe bet
> > to keep working.

> > Looks interesting, I'll keep that in mind.
> 
> > Regardless, so long as you have an image file and can get it to the
> > openQA server somehow (or put it up on a web server), you can quite
> > trivially tell openQA 'hey, run these tests on this image'. If we

> Right, it's just easier/faster for me to make my boot.iso here and run
> the tests on it than it is to make it and wait to upload it someplace.

That's a point, indeed. I guess I'm just a bit worried about the long-
term ongoing work that would be required to keep two runners working.
Still, we can try it. It would certainly make sense as the short-term
starting point, to keep openQA as the experimental new kid that has to
learn to play nice.

> > wanted to let people run scratch tests on a shared openQA instance we
> > might have to think about ways to do naming and stuff such that there's
> > a clear separation between those and the 'production' tests, or maybe
> > have separate instances for scratch purposes, but all that is solvable.
> > (Depending on the level of activity, it may require someone somewhere
> > to cough up some money, though :>)
> > 
> > I also set up a thing in our openQA scheduling code last week for
> > letting you run tests with an arbitrary updates.img , which you could
> > use for quick and dirty tests too.

> That would be pretty useful, usually that's all it needs, not a full
> rebuild of the iso.

OK, thanks for the feedback. I think what I'm gonna do at this point is
talk to our boss about resources - anything along these lines would
need, I think, at least some increase in our worker capacity - and then
come up with some kind of proposal from there...
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list



[Index of Archives]     [Kickstart]     [Fedora Users]     [Fedora Legacy List]     [Fedora Maintainers]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [Yosemite Photos]     [KDE Users]     [Fedora Tools]
  Powered by Linux