infra.aap_configuration is asking for an objectID instead of a name

[jgoupil26]

Is your feature request related to a problem? Please describe.

As described in :
https://github.com/redhat-cop/infra.aap_configuration/tree/devel/roles/gateway_role_user_assignments

you must input object_id for the target of the role, instead of a human readable name.

To my knowledge, there is no way get that object ID with the collection.

Describe the solution you'd like

An option to pass a human readable name for the target should be provided for the target object.

Describe alternatives you've considered

What i do personally is I do an API call and construct an input yml in that pipeline.

Additional context

The object_id Works, but it doesn't follow the principal of self contained human readable inputs of the collection.

[Tompage1994]

Hi @jgoupil26

I've also come across this issue, and I think the solution will be to add an extra piece to the role to do the lookup as it is currently a limitation of the module. That said, we are currently awaiting a couple of modules to be created in some of the upstream collections, and for the subsequent roles to also be created. That will probably be the point to make this implementation. I'll leave this open for now and we will revisit once ready

[Tompage1994]

The tracking issue is #1046

[hobby65]

I created a workaround for this, the object_id is asked for teams and organizations.
Check my site for a workaround:

http://ansible.isyour.guru/rhaap25/base_gateway_hub/

In the pages a solution for the problem is implemented in main.yml, by splitting the dispatch_variable in 2 parts, thus stopping the pipeline, query gateway for the organizations and teams, then using jinja to merge it to the role_user_assignments variable that the collection wants.

It is a workaround, but it works, even if there is a gap in the object_id's, this is where it breaks, recovering after a organization or team is deleted.