[openstack-dev] Horizon PTL candidacy
Gabriel.Hurley at nebula.com
Mon Nov 11 19:14:45 UTC 2013
We all agree on the benefits of an event-driven "push" system, but the basis for this already exists in OpenStack. The RPC notification code exists in Oslo and most of the IaaS projects are using it now. There were (should have been?) conversations about how to leverage this data stream in the Horizon sessions in Hong Kong.
As I understood it, Marconi has a completely different goal (queuing service a la SQS) which operates as a "building block" service for user applications, not as a publication channel for the core OpenStack infrastructure.
Anyhow, we agree on the goal and have plans for how to do it already; Marconi ought to be an orthogonal concern when it's ready for graduation.
My two cents,
> -----Original Message-----
> From: Clint Byrum [mailto:clint at fewbar.com]
> Sent: Monday, November 11, 2013 11:02 AM
> To: openstack-dev
> Subject: Re: [openstack-dev] Horizon PTL candidacy
> Excerpts from Matthias Runge's message of 2013-11-11 00:04:52 -0800:
> > On 11/10/2013 11:53 PM, John Dickinson wrote:
> > > A random off-the-top-of-my-head use case would be to subscribe to
> > > events from creating or changing objects in a particular Swift
> > > account or container. This would allow much more efficient listings
> > > in Horizon for active containers (and may also be consumed by other
> > > listeners too).
> > >
> > > --John
> > >
> > yupp.
> > There are many many usecases for this, and we'd get rid of pulling
> > services for status.
> Note that Heat can also make use of this, and a need for such things came up
> during several session discussions. I forget who said this, but it sums up the
> "Polling sucks"
> I'd hope that we keep it a public API of some sorts, so that users can also take
> advantage of it.
> OpenStack-dev mailing list
> OpenStack-dev at lists.openstack.org
More information about the OpenStack-dev