Limits of Scalingo Backup Restoration: In‑Place vs New Database

Goal

Confirm that a recovered PostgreSQL database contains the same data as the original production instance while respecting Scalingo’s documented backup behavior.

Constraints & Uncertainty

• Backups are logical dumps with retention governed by the add‑on tier, not user‑configurable.
• Documentation states restoration is performed by downloading a dump and loading it into a new or local instance; an in‑place one‑click rollback is not described.
• The impact of restoring to the live database—potential downtime, data loss, or overwrite—is not explicitly covered.

Open Questions

• Can a backup be restored directly into the existing Scalingo database without triggering downtime?
• If not, what is the recommended workflow for restoring to a new database and performing a cutover?
• What programmatic checks can best verify that the restored data matches the production state?