Goal

Automate the recovery process from an etcd quorum loss scenario in OpenShift, automating the manual steps required and improving the user experience.

Why is this important

• Reduced Complexity Eliminates the need to follow a manual recovery process
• Improved Efficiency Saves time and minimizes human error during disaster recovery situations
• Enhanced Support Makes it easier for the etcd team to support and test disaster recovery workflows
• OpenShift Virtualization Topologies All of these points are important for OpenShift Virtualization topologies with 2+2 control plane nodes across two sites, providing high availability and fast recovery.*

Initiative: Improve etcd disaster recovery experience (part3)

With OCPBU-252 and OCPBU-254 we create the foundations for an enhanced experience of a recovery procedure in the case of full control plane loss. This requires researching total control-plane failure scenarios of clusters deployed using the various deployment methodologies.

Scope of this feature:

• Spike to research if restoring full control plane with identical properties as the original control plane allow re-importing workers and document workload behavior
• Document procedure to restore from full control plane failure using compact cluster to restore control plane and the re-attachment of workers
• Enhanced e2e testing for validation of the updated manual procedure under this feature