Follow on to
CONSOLE-2748. The cluster list should no longer be hard-coded to one additional cluster. Get the list of clusters from ACM. The backend will need to be updated so that it can proxy to any cluster.
- Determine if there is a way to get the CA file for the cluster if it's a self-signed certificate.
- Determine if the backend (or console operator) should request the cluster list or the frontend. Regardless there will be backend changes.
- This story does not deal with version skew across clusters in the fleet. For the POC, we can assume all clusters will be at the same version.
- Assume the user is kubeadmin. We don't necessarily need to use the console service account to fetch the cluster list for this POC.
Once you have ACM installed, the kube api server might be available at:
ManagedClusterInfo > spec.masterEndpoint
ClusterDeployment > status.apiURL (if cluster was provisioned from ACM/Hive)
For a quick list of managed clusters, you could query the projects by this label:
ACM has a namespace per managed cluster.
- Working code that builds on the existing prototype and uses the ACM cluster list to populate the dropdown.
- Document any limitations or problems that arise during the prototype.