Musing: Things we need (code-wise)

Matt Wagner matt.wagner at redhat.com
Fri Sep 14 13:04:21 UTC 2012


On Thu, Sep 13, 2012 at 03:12:59PM -0700, Richard Su wrote:
> We do have documentation for the resources we have implemented thus
> far and drafts for some of the resources we're looking to implement
> next.
> You can find them here: https://www.aeolusproject.org/redmine/projects/aeolus/wiki#Conductor-API
> 
> I do like how Katello's api docs are structured giving information
> about each parameter.
> Thanks for pointing that out.

Fantastic -- I missed this documentation, but this is great.

I do like the idea of something auto-generated.

> >* A real API, that presents all the functionality that exists in our UI.
> Yes, we've been laying the ground work for a bigger push post 1.1.
> The bigger chunks like deployables,deployments, and instances have
> stories in the backlog and do need to be fleshed out.

Yes, this one is something I know we're already tracking. I've heard
from people in the field that the ability to manage the whole lifecycle
via API would be fantastic, particularly around launching new instances,
etc.

> >* The ability to use multiple datacenters with one RHEV or vSphere
> >  provider, or good documentation explaining how to handle this
> >  (probably common) use case.
> Some use cases on how multiple datacenters in a single provider would
> work would be useful. Or how it differs from having a provider for
> each datacenter. It is unclear to me how the user would like it to
> function.

Yes, true. This is something I don't know a ton about, but someone was
asking in #aeolus and we didn't have a good answer. It seems like you
shouldn't have to add the *same* provider repeatedly just to use
multiple datacenters, but I don't really know what I'm talking about
either. ;)

-- Matt



More information about the aeolus-devel mailing list