[openstack-dev] Removing Nova V2 API xml support

Matthew Treinish mtreinish at kortar.org
Mon Nov 17 18:57:59 UTC 2014


On Mon, Nov 17, 2014 at 12:40:12PM -0500, Davanum Srinivas wrote:
> Matt,
> 
> Thanks for the help/reviews!. Looks like the sequence seems to be
> 
> DevStack / Icehouse - https://review.openstack.org/#/c/134972/
> DevStack / Juno - https://review.openstack.org/#/c/134975/
> Tempest - https://review.openstack.org/#/c/134924/
> Tempest - https://review.openstack.org/#/c/134985/
> Nova - https://review.openstack.org/#/c/134332/
> 
> Right?

Yeah, that looks right. 

The only caveat is that before 134985 can land the procedure is that a Nova core
has to +2 the 134332, and just leave a comment on the Tempest review saying it's
got Nova core sign off with the link. (it's just procedural, I know that
everyone wants to kill the XML API, myself included)

-Matt Treinish

> On Mon, Nov 17, 2014 at 8:48 AM, Matthew Treinish <mtreinish at kortar.org> wrote:
> > On Mon, Nov 17, 2014 at 08:24:47AM -0500, Davanum Srinivas wrote:
> >> Sean, Joe, Dean, all,
> >>
> >> Here's the Nova change to disable the V2 XML support:
> >> https://review.openstack.org/#/c/134332/
> >>
> >> To keep all the jobs happy, we'll need changes in tempest, devstack,
> >> devstack-gate as well:
> >>
> >> Tempest : https://review.openstack.org/#/c/134924/
> >> Devstack : https://review.openstack.org/#/c/134685/
> >> Devstack-gate : https://review.openstack.org/#/c/134714/
> >>
> >> Please see if i am on the right track.
> >>
> >
> > So this approach will work, but the direction that neutron took and that
> > keystone is in the process of undertaking for doing the same was basically the
> > opposite. Instead of just overriding the default tempest value on master
> > devstack to disable xml testing the devstack stable branches are update to
> > ensure xml_api is True when running tempest from the stable branches, and then
> > the default in tempest is switched to False. The advantage of this approach is
> > that the default value for tempest running against any cloud will always work.
> >
> > The patches which landed for neutron doing this:
> >
> > Devstack:
> > https://review.openstack.org/#/c/130368/
> > https://review.openstack.org/#/c/130367/
> >
> > Tempest:
> > https://review.openstack.org/#/c/127667/
> >
> > Neutron:
> > https://review.openstack.org/#/c/128095/
> >
> >
> > Ping me on IRC and we can work through the process, because things need to land
> > in a particular order to make this approach work. But, basically the approach
> > is first the stable devstack changes need to land which enable the testing on
> > stable, followed by a +2 on the failing Nova patch saying the approach is
> > approved, and then we can land the tempest patch which switches the default,
> > which will let the Nova change get through the gate.
> >
> > -Matt Treinish
> >
> > _______________________________________________
> > OpenStack-dev mailing list
> > OpenStack-dev at lists.openstack.org
> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
> >
> 
> 
> 
> -- 
> Davanum Srinivas :: https://twitter.com/dims
> 
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
-------------- 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/20141117/ad9c5996/attachment.pgp>


More information about the OpenStack-dev mailing list