Open Stack

Thanks for the additional info!

Cheers
Alex.

On Thu, Mar 28, 2019 at 7:31 AM Rikimaru Honjo <
honjo.rikimaru at po.ntt-tx.co.jp> wrote:

> Hello Alex,
>
> On 2019/03/27 0:33, Alex Kavanagh wrote:
> > Hi
> >
> > On Fri, Mar 22, 2019 at 9:25 AM Rikimaru Honjo <
> > honjo.rikimaru at po.ntt-tx.co.jp> wrote:
> >
> >> Hello,
> >>
> >> I'm using the nova-lxd with the latest nova.
> >>
> >> In my understanding, nova-lxd already supports live-migration. But,it
> >> didn't work
> >> when I tried it. Is there anyone who can do live-migration with
> nova-lxd?
> >>
> >
> > I'm currently one of the maintainers of nova-lxd and I don't think that
> it
> > supports live migration.  I'm fairly certain migration is broken (
> > https://bugs.launchpad.net/nova-lxd/+bug/1818078) as it didn't work for
> me
> > when I was testing.  Live migration appears to be disabled in the tempest
> > testing (legacy-tempest-dsvm-lxd-ovs test):
> >
> > setUpClass
> > (tempest.api.compute.admin.test_live_migration.LiveMigrationTest) ...
> > SKIPPED: LiveMigrationTest skipped as live-migration is not available
> OK, I see.
>
> >> I think that the root cause is that the LXDLiveMigrateData object is not
> >> registered with nova-condcutor & compute.
> >>
> >
> > I say "don't think it supports live migration" because I became a
> > maintainer after that functionality appeared to be implemented; but I'm
> not
> > entirely sure the work is complete (i.e. it returns an object that
> doesn't
> > function with Nova as your mail indicates).  Whether it's because nova
> > changed and nova-lxd didn't notice (quite likely!) or it never worked
> > properly, I don't know, sadly.
> >
> >
> >>
> >> I wrote more information as a comment#2 in the following bug report.
> >> https://bugs.launchpad.net/nova-lxd/+bug/1711568
> >
> >
> > Thanks, I'll reference it in the bug I raised, too.  In terms of getting
> it
> > sorted out, it's not going to happen before the stein release, but I do
> > want to get to it at the beginning of the next cycle.
> Thanks a lot.
>
> By the way, I confirmed the detail of the solution provided by Matt after
> that.
> As a result, I found that the solution requires a special nova-conductor
> process "nova-conductor-powervm".
>
>
> https://github.com/openstack/nova-powervm/blob/newton-eol/README.rst#live-migration
>
> I think that it is troublesome for users...
>
> Current PowerVMLiveMigrateData was moved to nova tree in the following
> patch.
> https://review.openstack.org/#/c/391284/
>
> Best regards,
>
> > Thanks
> > Alex.
> >
> >
> >>
> >>
> >> Best regards,
> >> --
> >> _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> >> Rikimaru Honjo
> >> E-mail:honjo.rikimaru at po.ntt-tx.co.jp
> >>
> >>
> >>
> >
>
> --
> _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
> ★部署名が変わりました。
>
> NTTテクノクロス株式会社
> IoTイノベーション事業部 第二ビジネスユニット（IV2BU）
> 本上力丸
> TEL.  :045-212-7539
> E-mail:honjo.rikimaru at po.ntt-tx.co.jp
> 〒220-0012
>    横浜市西区みなとみらい4丁目4番5号
>    横浜アイマークプレイス 13階
>
>
>

-- 
Alex Kavanagh - Software Engineer
OpenStack Engineering - Data Centre Development - Canonical Ltd
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20190328/2c32ee35/attachment-0001.html>