[openstack-dev] [nova] Next steps for resource providers work

Sylvain Bauza sbauza at redhat.com
Wed Aug 31 11:55:07 UTC 2016



Le 31/08/2016 13:08, Jay Pipes a écrit :
> On 08/29/2016 12:40 PM, Matt Riedemann wrote:
>> I've been out for a week and not very involved in the resource providers
>> work, but after talking about the various changes up in the air at the
>> moment a bunch of us thought it would be helpful to lay out next steps
>> for the work we want to get done this week.
>>
>> Keep in mind feature freeze is more or less Thursday 9/1.
>>
>> Also keep in mind the goal from the midcycle:
>>
>> "Jay's personal goal for Newton is for the resource tracker to be
>> writing inventory and allocation data via the placement API. Get the
>> data pumping into the placement API in Newton so we can start using it
>> in Ocata."
>>
>> 1. The ResourceTracker work starts here:
>>
>> https://review.openstack.org/#/c/358797/
>>
>> That relies on the placement service being in the service catalog and
>> will be optional for Newton. There are details to be sorted about
>> if/when to retry connecting to the placement service with or without
>> requiring a restart of nova-compute, but I don't think those are too 
>> hairy.
>>
>> Jay is working on changes that go on top of that series to push the
>> inventory and allocation data from the resource tracker to the placement
>> service.
>
> OK, just pushed new revision for 358797 that fixes up Sean's patch 
> with unit tests and pep8 fixes and fixes the review comment from Alex 
> Xu regarding a missing log statement.
>
> Also, pushed the patch that adds inventory recording from the resource 
> tracker, rebased onto Sean's patch:
>
> https://review.openstack.org/363583
>
> It needs unit tests added. Will try to get to those this evening but 
> if someone is interested in helping, that would be cool.
>

I'm already working on adding UTs to 
https://review.openstack.org/#/c/363061 which is about to do the same 
things.

-sylvain

> Best,
> -jay
>
>> Chris Dent pointed out that there is remaining work to do with the
>> allocation objects in the placement API, but those can be worked in
>> parallel to the RT work Jay is doing.
>>
>> 2. Chris is going to cleanup the devstack change that adds the placement
>> service:
>>
>> https://review.openstack.org/#/c/342362/
>>
>> The main issue is there isn't a separate placement database, at least
>> not by default, so Chris has to take that into account. In Newton, by
>> default, the Nova API DB will be used for the placement service. You can
>> optionally configure a separate placement database with the API DB
>> schema, but we're not going to test with that as the default in devstack
>> in Newton since that's most likely not what deployers would be doing in
>> Newton as the placement service is still optional.
>>
>> 3. I'm going to work on a job that runs in the experimental queue and
>> enables the placement service. So by default in Newton devstack the
>> placement service will not be configured or running. With the
>> experimental queue job we can test the Nova changes with and without the
>> placement service to make sure we didn't completely screw something up.
>>
>> -- 
>>
>> If I've left something out please add it here.
>>
>
> __________________________________________________________________________ 
>
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: 
> OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev




More information about the OpenStack-dev mailing list