When using Cluster pools, if you load this controller, it will create ManagedCluster, so that all you have to do is apply a simple clusterClaim resource.

---
apiVersion: hive.openshift.io/v1
kind: ClusterClaim
metadata:
  name: my-cluster
  namespace: aws-east
spec:
  clusterPoolName: aws-eas

Note: See ./examples/clusterclaim.yaml

To install this controller, you must be kubeadmin. The controller will be deployed to the open-cluster-management namespace.

oc apply -k ./deploy

It will take 1-2min for the image to download the first time. The controller runs two pods, and chooses a leader to reduce the possibility of an outage.

The following steps assume you will use the ./examples/clusterclaim.yaml

1. Navigate to the ACM console and click Infrastructure > Clusters > Cluster pools
2. Click the Create cluster pool button
3. Choose aws, then pick an Infrastructure provider credential, and press Next to continue
4. Name the pool aws-east and use namespace aws-east, choose a size and release image, then click Next to continue
5. Customize any additional settings in steps 3-6
6. Then on step 7 Review and click Create (To use Single node clusters, choose a 4.8.0 image and make sure the install-config has controlPlane.replicas: 1 and compute[0].replicas: 0)
7. At this point you can apply the clusterclaim.yaml and a cluster will be claimed.

You can claim as many clusters as you want and they will be queued up and provisioned if the pool is empty. It is recommended to create a subscription that points to the exmples folder, so you can commit more clusterclaim.yaml files to Git. ACM will automatically claim those clusters, giving you a very simple Cluter Create GitOps flow. You can create multiple pools, to support different cluster configurations and providers(AWS, GCP & Azure).

• OpenShift GitOps can also be used to deliver the clusterclaim.yaml from the examples directory to the ACM Hub.
• When creating a namespace to hold your cluster pools, if you add the label:

  metadata:
    labels:
      open-cluster-management.io/managed-by: clusterpools
      ...

  Then as the last cluster pool is removed, the namespace will be deleted. If the label is not present, the namespace will not be removed.