Open Stack

Hi !

I thought my perspective might be valuable as the author of a project that
would likely end up being slashed: ARA [1].

In a nutshell, ARA provides easy and seamless Ansible reporting on playbook
runs.

It has nothing to do with OpenStack, to be honest: it doesn't require
OpenStack to run, it doesn't run on OpenStack.
We highlight that ARA is not an "official" OpenStack project in the
contributors' documentation [2].

But, it was born out of necessity inside the RDO community (call that the
extended OpenStack community, if you will) due to the vast amount of our
workloads that are driven by Ansible.

I like to draw a parallel between ARA and JJB [3] (Jenkins Job Builder) as
two tools that the OpenStack community develops and use without slapping
the OpenStack brand on them.
If we take a step back, would projects like Nodepool or Zuul also be
affected by this kind of cut ? Where do we draw the line ? Are they
OpenStack projects or not ?

For me, applying for ARA to be hosted as a "community" project [4], or what
would've been a "Stackforge" project long ago, was a completely logical
choice to me.

I could go and praise how the Gerrit workflow is vastly superior than
GitHub pull requests or that the Zuul-driven CI is completely awesome.
But... The truth is that I sincerely hoped it would help the project gain
exposure and credibility in order to drive adoption *inside* the OpenStack
community.

We say that hindsight is 20/20, but, in retrospect [5], I guess we could
say it worked.
Maybe out of luck, maybe out of sheer dedication and long weekends spent
working on the project.

The reality is that today, ARA is used by many different projects inside
the OpenStack ecosystem -- to name a few:
- OpenStack-Ansible
- TripleO
- Browbeat
- Kolla/Kolla-Ansible
- Devstack-Gate

But also, like Jenkins Job Builder, it has gained adoption outside of the
OpenStack community -- such as inside the OpenShift Ansible community or
BonnyCI.

Being part of the OpenStack "ecosystem" allows ARA to do things like easily
implement a gate job almost verbatim from OpenStack-Ansible [6] to make
sure that each new commit doesn't break them.
I feel that this is useful and powerful not just for ARA but for
OpenStack-Ansible and the other projects that use it as well.

Ironically, some projects feel hindered by this ecosystem (see: Gnocchi)
and have made an exit.
I do not share these feelings and I would in fact be very sad to be shown
the door.

My 0.02$CAD (not worth much right now)

​[1]​: https://github.com/openstack/ara
​[2]: ​http://ara.readthedocs.io/en/latest/contributing.html#contributing
[3]: https://github.com/openstack-infra/jenkins-job-builder
[4]: https://review.openstack.org/#/c/321226/
​[5]:
https://dmsimard.com/2017/05/08/ara-is-one-year-old-a-look-back-at-the-past-year/
​
​[6]:
https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/ara.yaml#L21-L89
​

David Moreau Simard
Senior Software Engineer | Openstack RDO

dmsimard = [irc, github, twitter]

On Jun 28, 2017 6:46 PM, "James E. Blair" <corvus at inaugust.com> wrote:

Thierry Carrez <thierry at openstack.org> writes:

> Removing the root cause would be a more radical move: stop offering
> hosting to non-OpenStack projects on OpenStack infrastructure
> altogether. We originally did that for a reason, though. The benefits of
> offering that service are:
>
> 1- it lets us set up code repositories and testing infrastructure before
> a project applies to be an official OpenStack project.
>
> 2- it lets us host things that are not openstack but which we work on
> (like abandoned Python libraries or GPL-licensed things) in a familiar
> environment
>
> 3- it spreads "the openstack way" (Gerrit, Zuul) beyond openstack itself

I think this omits what I consider the underlying reason for why we did
it:

It helps us build a community around OpenStack.

Early on we had so many people telling us that we needed to support
"ecosystem" projects better.  That was the word they used at the time.
Many of us said "hey, you're free to use github" and they told us that
wasn't enough.

We eventually got the message and invited them in, and it surpassed our
expectations and I think surprised even the most optimistic of us.  We
ended up in a place where anyone with an OpenStack related idea can try
it out and collaborate frictionlessly with everyone else in the
OpenStack community on it, and in doing so, become recognized in the
community for that.  The ability for someone to build something on top
of OpenStack as part of the OpenStack community has been empowering.

I confess to being a skeptic and a convert.  I wasn't thrilled about the
unbounded additional responsibility when we started this, but now that
we're here, I think it's one of the best things about the project and I
would hate to cleave our community by ending it.

-Jim

__________________________________________________________________________
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20170628/58736cd1/attachment.html>