Epic Goal

Can a two node scenario sustain the restore of an ACM to one of it's Managed Clusters.

Since by default local-cluster is enabled, how does that affect a restore on a cluster that used to be Managed Cluster.

Why is this important?

Two rack scenarios

Scenarios

1. ACM on hub 1 on Rack 1 fails.  Can you restore ACM to hub 2 on Rack 2 that was managed by ACM? 

2. Needs to support Ramen DR recovery flows

Acceptance Criteria

Either this can or can not be supported.

Dependencies (internal and external)

1. Agent support when restoring.  
   1. Use local-cluster vs remote
   2. Could use hosted mode
   3. restore with local-cluster agent on a cluster where the managed cluster agent is already installed

Previous Work (Optional):

1. local-cluster rename
2. Hosted Mode
3. Agent namespace change
4. Backup and restore

Open questions:

1. Can we set local managed FALSE, then import the cluster with its regular name
2. Should we use rename local cluster
3. Can you restore on a Managed Cluster

Done Checklist

• Release Enablement <link to Feature Enablement Presentation>
• DEV -Process to move ACM between two nodes that it manages is complete.
• QE - Validated
• DOC - Doc for product or just for ODF