[openstack-dev] [all] [tc] Community Goals for Rocky

Tim Bell Tim.Bell at cern.ch
Fri Jan 12 17:09:16 UTC 2018

I was reading a tweet from Jean-Daniel and wondering if there would be an appropriate community goal regarding support of some of the later API versions or whether this would be more of a per-project goal.


Interesting numbers about customers tools used to talk to our @OpenStack APIs and the Keystone v3 compatibility:
- 10% are not KeystoneV3 compatible
- 16% are compatible
- for the rest, the tools documentation has no info

I think Keystone V3 and Glance V2 are the ones with APIs which have moved on significantly from the initial implementations and not all projects have been keeping up.


-----Original Message-----
From: Emilien Macchi <emilien at redhat.com>
Reply-To: "OpenStack Development Mailing List (not for usage questions)" <openstack-dev at lists.openstack.org>
Date: Friday, 12 January 2018 at 16:51
To: OpenStack Development Mailing List <openstack-dev at lists.openstack.org>
Subject: Re: [openstack-dev] [all] [tc] Community Goals for Rocky

    Here's a quick update before the weekend:
    2 goals were proposed to governance:
    Remove mox
    Champion: Sean McGinnis (unless someone else steps up)
    Ensure pagination links
    Champion: Monty Taylor
    2 more goals are about to be proposed:
    Enable mutable configuration
    Champion: ChangBo Guo
    Cold upgrades capabilities
    Champion: Masayuki Igawa
    Thanks everyone for your participation,
    We hope to make a vote within the next 2 weeks so we can prepare the
    PTG accordingly.
    On Tue, Jan 9, 2018 at 10:37 AM, Emilien Macchi <emilien at redhat.com> wrote:
    > As promised, let's continue the discussion and move things forward.
    > This morning Thierry brought the discussion during the TC office hour
    > (that I couldn't attend due to timezone):
    > http://eavesdrop.openstack.org/irclogs/%23openstack-tc/latest.log.html#t2018-01-09T09:18:33
    > Some outputs:
    > - One goal has been proposed so far.
    > Right now, we only have one goal proposal: Storyboard Migration. There
    > are some concerns about the ability to achieve this goal in 6 months.
    > At that point, we think it would be great to postpone the goal to S
    > cycle, continue the progress (kudos to Kendall) and fine other goals
    > for Rocky.
    > - We still have a good backlog of goals, we're just missing champions.
    > https://etherpad.openstack.org/p/community-goals
    > Chris brought up "pagination links in collection resources" in api-wg
    > guidelines theme. He said in the past this goal was more a "should"
    > than a "must".
    > Thierry mentioned privsep migration (done in Nova and Zun). (action,
    > ping mikal about it).
    > Thierry also brought up the version discovery (proposed by Monty).
    > Flavio proposed mutable configuration, which might be very useful for operators.
    > He also mentioned that IPv6 support goal shouldn't be that far from
    > done, but we're currently lacking in CI jobs that test IPv6
    > deployments (question for infra/QA, can we maybe document the gap so
    > we can run some gate jobs on ipv6 ?)
    > (personal note on that one, since TripleO & Puppet OpenStack CI
    > already have IPv6 jobs, we can indeed be confident that it shouldn't
    > be that hard to complete this goal in 6 months, I guess the work needs
    > to happen in the projects layouts).
    > Another interesting goal proposed by Thierry, also useful for
    > operators, is to move more projects to assert:supports-upgrade tag.
    > Thierry said we are probably not that far from this goal, but the
    > major lack is in testing.
    > Finally, another "simple" goal is to remove mox/mox3 (Flavio said most
    > of projects don't use it anymore already).
    > With that said, let's continue the discussion on these goals, see
    > which ones can be actionable and find champions.
    > - Flavio asked how would it be perceived if one cycle wouldn't have at
    > least one community goal.
    > Thierry said we could introduce multi-cycle goals (Storyboard might be
    > a good candidate).
    > Chris and Thierry thought that it would be a bad sign for our
    > community to not have community goals during a cycle, "loss of
    > momentum" eventually.
    > Thanks for reading so far,
    > On Fri, Dec 15, 2017 at 9:07 AM, Emilien Macchi <emilien at redhat.com> wrote:
    >> On Tue, Nov 28, 2017 at 2:22 PM, Emilien Macchi <emilien at redhat.com> wrote:
    >> [...]
    >>> Suggestions are welcome:
    >>> - on the mailing-list, in a new thread per goal [all] [tc] Proposing
    >>> goal XYZ for Rocky
    >>> - on Gerrit in openstack/governance like Kendall did.
    >> Just a fresh reminder about Rocky goals.
    >> A few questions that we can ask ourselves:
    >> 1) What common challenges do we have?
    >> e.g. Some projects don't have mutable configuration or some projects
    >> aren't tested against IPv6 clouds, etc.
    >> 2) Who is willing to drive a community goal (a.k.a. Champion)?
    >> note: a Champion is someone who volunteer to drive the goal, but
    >> doesn't commit to write the code necessarily. The Champion will
    >> communicate with projects PTLs about the goal, and make the liaison if
    >> needed.
    >> The list of ideas for Community Goals is documented here:
    >> https://etherpad.openstack.org/p/community-goals
    >> Please be involved and propose some ideas, I'm sure our community has
    >> some common goals, right ? :-)
    >> Thanks, and happy holidays. I'll follow-up in January of next year.
    >> --
    >> Emilien Macchi
    > --
    > Emilien Macchi
    Emilien Macchi
    OpenStack Development Mailing List (not for usage questions)
    Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe

More information about the OpenStack-dev mailing list