Open Stack

On Thu, Apr 2, 2020 at 10:31 AM Radosław Piliszek <
radoslaw.piliszek at gmail.com> wrote:

> It's April already so I think a progress report might be beneficial to
> not lose the topic.
>
> The change switching DevStack to CirrOS 0.5.1 [1] is still paused.
> This is mostly because I would really prefer to hit CI early in the
> Victoria cycle rather than late in Ussuri.
> Please let me know if you beg to differ. ;-)
> That said, thank you Dmitry, Iury and Maciej for taking the time to
> test it on Ironic [2] and Neutron [3] sides.
> Re Neutron - glad to see Xenial go away with this. Finally! :-)
>
> Neutron looks overally happy with 0.5.1.
> Ironic in general too, but metalsmith's job seems to have issues with
> available storage space.
>

I hope that https://review.opendev.org/716894 will solve this issue.


>
> As for merged progress, the CI images now have CirrOS 0.5.1 cached for
> x86_64 [4] and aarch64 [5].
> Also, we already switched Kolla Ansible to test against CirrOS 0.5.1
> (for better aarch64 support).
>
> Even if we do the switch in Victoria, it's best to test early, so
> please let me know if you have questions and/or need help with
> switching your CI.
>
> [1] https://review.opendev.org/711492
> [2] https://review.opendev.org/712728
> [3] https://review.opendev.org/711425
> [4] https://review.opendev.org/714475
> [5] https://review.opendev.org/714481
>
> PS: I just noticed I wrote 'cinder' in one place in the original
> message. Please forgive me the confusion.
>
> -yoctozepto
>
> czw., 12 mar 2020 o 12:55 Radosław Piliszek
> <radoslaw.piliszek at gmail.com> napisał(a):
> >
> > Hiya Folks,
> >
> > as you might have noticed, cinder 0.5.1 has been released.
> > This build seems to be passing the current devstack gate. [1]
> > Big thanks to hrw and smoser for letting cirros 0.5.1 happen (and
> > cirros having bright future yet again).
> > Also thanks to mordred for cleaning up SDK testing to pass. :-)
> >
> > I think it would be nice to merge this in Ussuri still, preferably
> before April.
> > On the other hand, we all know that devstack gate is not super
> > comprehensive and hence I would like to ask teams whose tests depend
> > on interaction with guest OS to test their gates on this patch (or
> > help me help you do that).
> > I deliberately marked it W-1 to avoid merging too early.
> >
> > Let the discussion begin. :-)
> >
> > [1] https://review.opendev.org/711492
> >
> > -yoctozepto
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20200402/47fe2f42/attachment-0001.html>