1. Proposed title of this feature request
Recovering from losing 2 master nodes in OCP4 DR scenario

2. What is the nature and description of the request?
Can the following procedure be officially supported?
https://gist.github.com/fjcloud/40620f14b3a8ea701296776cc75cad69

There is also this KCS describing a bit different approach to recovering from losing 2 master nodes:
https://access.redhat.com/solutions/5235891

The following official docs describe similar procedure by restoring from etcd backup. It however does not mention how to recover in case of no etcd backup or in case the failed nodes are beyond repair/deleted/lost.
https://docs.openshift.com/container-platform/4.8/backup_and_restore/control_plane_backup_and_restore/disaster_recovery/scenario-2-restoring-cluster-state.html#dr-restoring-cluster-state

3. Why does the customer need this? (List the business requirements here)
To be able to recover the cluster in a case of a full, irreparable lost of 2 master nodes.

4. List any affected packages or components.
etcd, kube-apiserver, kube-scheduler, kube-controller-manager, CNI