Keeping the platform available and your data recoverable.

• Recovery Point Objective (RPO): 24 hours for app-level snapshot exports; near-continuous for the managed database's point-in-time recovery window.
• Recovery Time Objective (RTO): 4 hours for a full-region failover; 8 hours for restoration of a specific customer dataset from snapshot.

The managed database provides automated daily backups and point-in-time recovery. In addition, LDPS Africa runs an app-level snapshot export covering 12 critical tables (profiles, farms, records, subscriptions, audit and financial ledgers). Snapshots carry SHA-256 checksums for integrity verification and are visible to super-admins under /founder/backups.

The application layer is stateless and served through a global edge network, which absorbs single-node and single-region failures automatically. The database provider maintains standby replicas and manages promotion during a regional incident.

1. Declare an incident and open an sla_incidents record with the affected scope.
2. Freeze customer writes for the affected tables to prevent divergent state during recovery.
3. Identify the correct recovery point (PITR timestamp or snapshot ID).
4. Restore into a staging schema and verify checksums against the snapshot registry.
5. Swap the verified schema in and re-enable writes.
6. Notify affected customers and post the outcome to the Status page.

Snapshot integrity is exercised each time an operator downloads and verifies a backup through the file-integrity checker. A full restore drill into a staging environment is performed at least once per calendar year.