[openstack-dev] [zaqar][all] PTL No-Candidacy

Flavio Percoco flavio at redhat.com
Fri Sep 11 11:51:03 UTC 2015


I'm sending this email to announce that I wont be running for Zaqar's
PTL position this cycle.

I've been Zaqar's PTL for two cycles and I believe it is time for me
to move on. More importantly, I believe it's time for this great,
still small, project to be led by someone else. The reasons behind
this belief have nothing to do with neither the previous state of the
project or even its current success story. If anything, my current
decision of not running has everything to do with the project's
current growth.

As many of you know, Zaqar (formerly known as Marconi) went through
many ups and downs. From great discussions and growth attempts to
almost being shutdown[0]. This has taugh me a lot but more
importantly, it's made the team stronger and it's cleared the team's
goals and path. And to prove that, let me share some of the success
stories the team has had since Vancouver:

3 great milestones

Let me start by sharing the progress the project has made code-wise.
While it may not be the most important for many people, I believe it's
extremly valuable for the project. The reason for this being that
every single member of this team is not a full-time Zaqar developer.
That means, every single member of this team has a different full-time
responsibility and every contribution made to the project has been
made in their spare working (or free) time. From amazing Outreachy
mentees (we've mentored participants of the Outrechy program since
cycle 1) to great contributors from other projects in OpenStack.

In milestone #1[1], we closed several bugs while we discussed the
features that we wanted to work on during Liberty. In milestone #2[1],
some of the features we wanted to have in Liberty started to land and
several bugs were fixed as well. In milestone #3, many bugs were fixed
due to a heavy testing session. But it doesn't end there. In RC1[4], 3
FFE were granted - not carelessly, FWIW - to complete all the work
we've planned for Liberty and, of course, more bug fixes.

We now even have a websocket example in the code base... ZOMG!

In addition to the above, the client library has kept moving forward
and it's being aligned with the current, maintained, API. This
progress just makes me happy and happier. Keep reading and you'll know

Adoption by other projects

If you read the call for adoption thread[0], you probably know how
important that was for the project to move forward. After many
discussions in Vancouver, on IRC, conferences, mailing lists, pigeons,
telegrams, etc. projects started to see[5] the different use-cases for
Zaqar and we started talking about implementations and steps forward.
One good example of this is Heat's use of Zaqar for
software-config[6], which was worked on and implemented.

Things didn't stop there on this front. Other projects, like Sahara,
are also considering using Zaqar to communicate with guests agents.
While this is under discussion on Sahara's side, the required features
for it to happen and be more secure have been implemented in Zaqar[7].
Other interesting discussions are also on-going that might help with
Zaqar's adoption[8].

That said, I believe one of the works I'm most excited about right now
is the puppet-zaqar project, which will make it simpler for
deployments based on puppet to, well, deploy zaqar[9].

Community Growth

None of the above would have been possible without a great community
and especially without growing it. I'm not talking about the core
reviewers team growth - although we did have an addition[10] - but the
growth of the community accross OpenStack. Folks from other teams -
OpenStack Puppet, Sahara, Heat, Trove, cross-project efforts - have
joined the efforts of pushing Zaqar forward in different ways (like
the ones I've mentioned before).

Therefore, I owe a huge THANK YOU to each and one of these people that
helped making this progress possible.

Oh God, please, stop talking

Sure, fine! But before I do that, let me share why I've said all the

The above is not to show off what the team has accomplished. It's
definitely not to take any credits whatsoever. It's to show exactly
why the team needs a new PTL.

I believe PTLs should rotate every 2 cycles (if not every cycle). I've
been the PTL for 2 cycles (or probably even more) and it's time for
the vision and efforts of other folks to jump in. It's time for folks
with more OPs knowledge than me to help making Zaqar more
"maintainable". It's time for new technical issues to come up and for
us as a community to work together on achieving those. More
cross-project collaboration, more APIs improvement, more user stories
is what Zaqar needs right now and I believe there are very capable
folks in Zaqar's team that would be perfect for this task.

One thing I'd like the whole team to put some efforts on, regardless
what technical decisions will be taken, is on increasing the diversity
of the project. Zaqar is not as diverse[11] (company wise) as I'd like
that worries me A LOT. Growth will, hopefully, bring more people and
reaching out to other communities remain something important.

It's been an honor to serve as Zaqar's PTL and it'll be an honor for
me to contribute to the next PTL's future plans and leads.


P.S: #openstack-zaqar remains the funiest channel ever, just sayin'.

[0] http://lists.openstack.org/pipermail/openstack-dev/2015-April/061967.html
[1] https://launchpad.net/zaqar/+milestone/liberty-1
[2] https://launchpad.net/zaqar/+milestone/liberty-2
[3] https://launchpad.net/zaqar/+milestone/liberty-3
[4] https://launchpad.net/zaqar/+milestone/liberty-rc1
[5] http://lists.openstack.org/pipermail/openstack-dev/2015-May/064739.html
[6] https://github.com/openstack/heat-specs/blob/master/specs/kilo/software-config-zaqar.rst
[7] http://specs.openstack.org/openstack/zaqar-specs/specs/liberty/pre-signed-url.html
[8] https://review.openstack.org/#/c/185822/
[9] https://github.com/openstack/puppet-zaqar
[10] http://lists.openstack.org/pipermail/openstack-dev/2015-August/072191.html
[11] http://stackalytics.com/?module=zaqar-group

Flavio Percoco
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150911/527acc74/attachment.pgp>

More information about the OpenStack-dev mailing list