Based on the Repair/Backup/Restore tasks for multi datacenter ScyllaDBClusters internal discussion document, the scope of this issue is to put the shape of the solution on paper based on the design decisions made so far.

Includes:

• goals and non-goals of the effort
  • what operations we intend to support and what we don't
  • what this means to existing users (breaking changes, migration paths, deprecations of APIs)
• new APIs/CRDs
• new data flows necessary (including dependencies like storage buckets, PKI, secret sharing, data synchronization between DCs)
• support plan (what is getting deprecated, maturity of the newly introduced APIs)
• testing plan

Acceptance criteria

• a document (enhancement proposal) exists that covers the above
• shared with the internal stakeholders