[openstack-dev] VMware Workstation / Fusion / Player Nova driver

Alessandro Pilotti apilotti at cloudbasesolutions.com
Tue Dec 3 21:59:46 UTC 2013



> On 03.12.2013, at 21:20, "Russell Bryant" <rbryant at redhat.com> wrote:
> 
>> On 12/03/2013 04:27 AM, Daniel P. Berrange wrote:
>>> On Mon, Dec 02, 2013 at 07:23:19PM +0000, Alessandro Pilotti wrote:
>>> 
>>> On 02 Dec 2013, at 04:52 , Kyle Mestery (kmestery) <kmestery at cisco.com> wrote:
>>> 
>>>> This is very cool Alessandro, thanks for sharing! Any plans to try and get this
>>>> nova driver upstreamed?
>>> 
>>> My personal opinion is that drivers should stay outside of Nova in a separate project.
> 
> Oh, so when are we dropping Hyper-V (again) ?  :-)

Now we have two drivers to drop at the price of one!

BTW didn't you want to come to the Hyper-V session in HK to talk about this (off)topic? :-)

> 
>> If drivers were to live in separate projects we would be forced to maintain
>> Nova internal code as stable APIs to avoid breaking drivers during the dev
>> cycle. This would place a significant burden on Nova development and have a
>> negative impact on the overal ease of development. It would also discourage
>> collaboration and sharing of code between virt drivers, which is already a
>> significant problem today whereby drivers come up with different ways todo
>> the same thing. If you want to be isolated from the community in a separate
>> project then expect your code to be broken periodically during development.
>> If you don't want that, then put the code in tree and be an active part of
>> the community effort working together, instead of in isolation.
> 
> I generally agree here.  If a driver wants to be separate, that's fine,
> there's nothing stopping you.  It just has to be done with the
> understanding that it's build on unstable APIs, and you will need
> something to ensure you keep up.  That is probably CI.  And if you're
> doing CI on the driver, you've met one of the most difficult requirments
> to being in Nova.
> 
> Someone may also want to have a driver separate for control reasons.
> I'd be surprised if that was really a factor with this particular
> driver, though.
> 
> -- 
> Russell Bryant
> 
> _______________________________________________
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev



More information about the OpenStack-dev mailing list