# Disaster Recovery

## Objectives (initial targets)

| Metric | Target |
| --- | --- |
| RPO | ≤ 24 hours (improve with more frequent backups) |
| RTO | ≤ 8 hours for full stack restore |

## Failure Scenarios

| Scenario | Response |
| --- | --- |
| App host loss | Provision host, restore compose + env, restore DB dumps, repoint Nginx upstreams |
| Edge host loss | Rebuild Nginx from `infrastructure/nginx/*`, restore certs/maps, re-run provision as needed |
| DB corruption | Restore from last known good dump; forward-fix if possible |
| Keycloak loss | Restore Keycloak DB + realm export; validate OIDC clients |

## Communication

Record incident time, impact, RPO/RTO achieved, and follow-ups in ops notes; update [progress.md](../progress.md) only for lasting platform changes.

## Related Documents

- [Backup](backup.md)
- [Restore](restore.md)
- [Monitoring](monitoring.md)
- [Production](production.md)