Open Stack

> I believe the only difference between your thoughts on this and my 
> own are the implementation details of how those placement HTTP API 
> calls would be made. I believe you want to see those calls done in 
> the nova.objects.Inventory[List] object whereas I was hoping to have 
> the resource tracker instead call a 
> placement_client.update_inventory() call which would be responsible 
> for talking to the placement REST API endpoint and the placement
> REST API endpoint would save inventory state to the Nova API database
> via calls to a nova.objects.ResourceProvider.update_inventory()
> method.

No, I'm not adamant about where they go. I suggested we put them in the
Inventory object purely to hide the where-does-it-go-when-I-save details
from the upper layers in compute. If you want compute to use different
models internally that it passes to the placement client, or have it
never really store them internally and just make calls into the
placement client when it has something to say, then that's fine with me.

--Dan