Recovery guide for Teleport HA for disaster recovery scenarios

[benarent]

Summary

We should provide guidance around break glass procedure guide for various recovery scenarios.

These will include

• [ ] If Teleport Auth cluster is offline
• [ ] Backing up core Teleport config and restoring
• [ ] Rolling out certs if an account has been compromised.

Audience

This guide should be aimed at users evaluating the product, and power users who want to cover all edge cases.

What steps are you concerned with, and what do you do?

Location

TBD

[russjones]

If the configuration management stack uses a non-SSH protocol for communication, that can be used to rebootstrap a node into a known safe state.

[benarent]

also TIL

Trust will automatically reestablish if you backup your CAs with tctl get --with-secrets cas > cas.yaml and, after full loss of backend, start the fresh auth server with --bootstrap cas.yaml.