Re: Should we maybe move Cirrus-CI jobs away from Gitlab again?

[Alex Bennée]

Daniel P. Berrangé <berrange@redhat.com> writes:

> On Tue, Sep 27, 2022 at 08:40:44PM +0200, Thomas Huth wrote:
>> On 27/09/2022 19.57, Daniel P. Berrangé wrote:
>> > On Tue, Sep 27, 2022 at 01:36:20PM -0400, Stefan Hajnoczi wrote:
>> > > On Tue, 27 Sept 2022 at 11:54, Daniel P. Berrangé <berrange@redhat.com> 
>> > > wrote:
>> > > > 
>> > > > On Tue, Sep 27, 2022 at 11:44:45AM -0400, Stefan Hajnoczi wrote:
>> > > > > On Tue, 27 Sept 2022 at 05:02, Thomas Huth <thuth@redhat.com> wrote:
>> > > > > > now that Gitlab is giving us pressure on the amount of free CI 
>> > > > > > minutes, I
>> > > > > > wonder whether we should maybe move the Cirrus-CI jobs out of the 
>> > > > > > gitlab-CI
>> > > > > > dashboard again? We could add the jobs to our .cirrus-ci.yml file 
>> > > > > > instead,
>> > > > > > like we did it in former times...
>> > > > > > 
>> > > > > > Big advantage would be of course that the time for those jobs 
>> > > > > > would not
>> > > > > > count in the Gitlab-CI minutes anymore. Disadvantage is of course 
>> > > > > > that they
>> > > > > > do not show up in the gitlab-CI dashboard anymore, so there is no 
>> > > > > > more
>> > > > > > e-mail notification about failed jobs, and you have to push to 
>> > > > > > github, too,
>> > > > > > and finally check the results manually on cirrus-ci.com ...
>> > > > > 
>> > > > > My understanding is that .gitlab-ci.d/cirrus.yml uses a GitLab CI job
>> > > > > to run the cirrus-run container image that forwards jobs to 
>> > > > > Cirrus-CI.
>> > > > > So GitLab CI resources are consumed waiting for Cirrus-CI to finish.
>> > > > > 
>> > > > > This shouldn't affect gitlab.com/qemu-project where there are private
>> > > > > runners that do not consume GitLab CI minutes.
>> > > > > 
>> > > > > Individual developers are affected though because they most likely
>> > > > > rely on the GitLab shared runner minutes quota.
>> > > > 
>> > > > NB, none of the jobs should ever be run automatically anymore in
>> > > > QEMU CI pipelines. It always requires the maintainer to set the
>> > > > env var when pushing to git, to explicitly create a pipeline.
>> > > > You can then selectively start each individual job as desired.
>> > > 
>> > > Cirrus CI is not automatically started when pushing to a personal
>> > > GitLab repo? If starting it requires manual action anyway then I think
>> > > nothing needs to be changed here.
>> > 
>> > No pipeline at all is created unless you do
>> > 
>> >    git push -o ci.variable=QEMU_CI=1 <your-fork-remote>
>> > 
>> > that creates the pipeliune but doesn't run any jobs - they're manual
>> > start.
>> 
>> Yes, sure, the jobs are not started automatically. But I *do* want to run
>> the jobs before sending pull requests - but since the gitlab-CI minutes are
>> now very limited, I'd like to avoid burning these minutes via gitlab and
>> start those jobs directly on cirrus-ci.com again. For that the jobs would
>> need to be moved to our .cirrus-ci.yml file again.
>
> We do need a better story for maintainers sending pull requests to have
> ability to run CI. We have 50+ jobs in the bujild stage of which the
> cirrus jobs are just 3 - removing the cirrus jobs won't make a difference
> to how quickly we run out of minutes if people try to run all of them.
>
> We need to define a much tighter minimalist set of recommended jobs to
> run.
>
> I believe that if QEMU joins the OSS program, then the forks of QEMU
> also benefit from a reduced cost factor for jobs they run, effectively
> giving you much higher CI quota

We shall find out soon enough. The code came through today so I've
applied it to the project which shows we are now on the "Ultimate" tier.

>
>> Well, maybe we could also have both, jobs via cirrus-run for those who want
>> to see them in their gitlab-CI dashboard, and via .cirrus-ci.yml for those
>> who want to avoid burning CI minutes on Gitlab. It's a little bit of
>> double-maintenance, but maybe acceptable?
>
> Key info about the jobs is in .gitlab-ci.d/cirrus/freebsd-12.vars which
> could be referenced from the cirrus-ci.yml to reduce duplication
>
> With regards,
> Daniel


-- 
Alex Bennée