Open Stack

On 5/28/2018 7:31 AM, Sylvain Bauza wrote:
> That said, given I'm now working on using Nested Resource Providers for 
> VGPU inventories, I wonder about a possible upgrade problem with VGPU 
> allocations. Given that :
>   - in Queens, VGPU inventories are for the root RP (ie. the compute 
> node RP), but,
>   - in Rocky, VGPU inventories will be for children RPs (ie. against a 
> specific VGPU type), then
> 
> if we have VGPU allocations in Queens, when upgrading to Rocky, we 
> should maybe recreate the allocations to a specific other inventory ?

For how the heal_allocations CLI works today, if the instance has any 
allocations in placement, it skips that instance. So this scenario 
wouldn't be a problem.

> 
> Hope you see the problem with upgrading by creating nested RPs ?

Yes, the CLI doesn't attempt to have any knowledge about nested resource 
providers, it just takes the flavor embedded in the instance and creates 
allocations against the compute node provider using the flavor. It has 
no explicit knowledge about granular request groups or more advanced 
features like that.

-- 

Thanks,

Matt