Open Stack

On Mon, Jan 9, 2017, at 01:58 PM, James E. Blair wrote:
> corvus at inaugust.com (James E. Blair) writes:
> 
> >> Yup, I think this makes sense and avoids duplicate image data. One other
> >> similarish use case that I don't think this addresses that we should
> >> consider is the one we had in hpcloud and what we do in osic-cloud1
> >> currently. Basically chunk up a provider in several different ways to
> >> affect distribution of nodes based on attributes within that provider. I
> >> don't have any great ideas for how that might look right now, but wonder
> >> if that might also solve the flavor problem. Probably something to think
> >> about before we commit to this.
> >
> > Yeah, I don't think this addresses that problem.  I suspect a real
> > solution to it would look a lot different than what we have now.  I'm
> > open to suggestions.
> 
> Perhaps something like this?  It creates a new "pools" section which is
> groups of instances+labels within a provider.  The actual image uploads
> are still at the provider level.  It's a bit more complicated, in that
> we have to explicitly cross-link the label to the cloud image (the
> previous suggestion implicitly did that by being underneath it in the
> yaml hierarchy).  It does provide a nice retcon for 'nodepool' though.
> :)

I like this. Would definitely make the current osic setup simpler (and
reduce the amount of work that the builder has to do uploading to those
"providers").

Clark