As a follow on from my post in the Foreman users group:
https://groups.google.com/forum/#!topic/foreman-users/wzRlD4ZB83k
I want to ability from the Foreman API to access compute resource
information such as cluster, network interface & volume information.
Since some of these concepts are handled differently across the various
resource types, what's the best way of implementing this?
Similar to the available_images call, I was thinking along the lines of
adding something like:
GET /api/compute_resources/:id/networks
GET /api/compute_resources/:id/clusters
GET /api/compute_resources/:id/volumes
Adding top level methods to the compute_resource model class and then
having each individual compute resource class override it (Ovirt in my
case).
Thoughts?
That makes sense - you'll probably find some of it is already access
from Fog and is used in our UI forms (e.g.
app/views/compute_resources_vms/form.
You can also use RABL in the API to customise the response for different
compute resources types, see app/views/api/v2/compute_resources/ for
examples.
···
On 27/01/14 16:06, James Netherton wrote:
> As a follow on from my post in the Foreman users group:
>
> https://groups.google.com/forum/#!topic/foreman-users/wzRlD4ZB83k
>
> I want to ability from the Foreman API to access compute resource
> information such as cluster, network interface & volume information.
>
> Since some of these concepts are handled differently across the various
> resource types, what's the best way of implementing this?
>
> Similar to the available_images call, I was thinking along the lines of
> adding something like:
>
> GET /api/compute_resources/:id/networks
> GET /api/compute_resources/:id/clusters
> GET /api/compute_resources/:id/volumes
>
> Adding top level methods to the compute_resource model class and then
> having each individual compute resource class override it (Ovirt in my
> case).
–
Dominic Cleal
Red Hat Engineering
