Open Stack

On Wed, Nov 28, 2012 at 03:16:12PM +0100, Julien Danjou wrote:
> On Wed, Nov 28 2012, Steven Hardy wrote:
> 
> > So are you saying you won't support adding a CW compatible API, or that you
> > won't support the concept (or at least requirements) of CW at all inside
> > ceilometer?
> 
> I'm saying we are not directing our next efforts on implementing a CW
> compatible API, that's for sure. But, we would really like to support
> the concept or at least the requirements to provide what's needed for a
> project like Heat to use Ceilometer metrics for its purpose. And we're
> discussing this right now with the dimension blueprint¹, for example.

Ok, thanks, your comments combined with the revised project goal on the wiki
made me concerned that you were shifting focus permanently away from
becoming "the infrastructure for all measurements within OpenStack" and back
to a metering specific project making little provision (other than the hooks
mentioned) for more general purpose/monitoring use-cases of the metric data.

The dimension blueprint looks good, that's exactly the sort of thing we need
to start using ceilometer as a metric service for heat.

> 
> > It seems like you're saying you'll only support a hook to push metric data
> > to an external CW API (which has it's own datastore), so heat will need to
> > either maintain it's own CW API implementation and datastore or move to
> > using something like Synaps (*and* ceilometer) to provide CW style metrics?
> 
> That's another thing we are going to support, among other things for
> projects like Synaps for example, like Eoghan said.

Ok, so I totally understand why the hook to push metrics to an external
service would be useful, but it does not solve the problem of a converged
metric API and datastore, which is, I think, what most people actually want.

I got pretty excited when ceilometer announced it was broadening its scope
to help move towards some sort of converged openstack-specific metric
solution - it's just dissapointing to see that shift in scope reversed for
now :(

Thanks,

Steve