Restore verification / v0.1

A backup is a promise. Test the restore.

Restore a Postgres dump or volume into a disposable Docker network, probe the database and app, then keep signed proof it worked.

Install the CLI

Internal network required. Production hosts rejected.

Plate 01 / isolated rehearsal Technical cutaway showing a sealed backup entering an isolated test chamber and emerging as a healthy database
Backup artifact in. Measured recovery evidence out.

Recorded drill / local simulation

Watch the proof assemble.

This preview uses no server and touches no data. The real CLI records the same checkpoints from Docker.

weekly-ordersReady
  1. Artifact hashwaiting
  2. Isolated restorewaiting
  3. SQL assertionswaiting
  4. HTTP probewaiting
  5. Signed reportwaiting

RTOSHA-256

The drill boundary

Small enough for cron. Strict enough to trust.

  1. 01

    Seal the room

    Pull images first, then create a new Docker --internal network. Existing networks and undeclared HTTP hosts are refused.

  2. 02

    Restore the artifact

    Load a plain/custom Postgres dump or a validated volume tar. The original stays read-only; secrets stay in local env files.

  3. 03

    Probe usability

    Run exact SQL expectations and app-level HTTP probes from inside the isolated network—not against production.

  4. 04

    Sign and clear

    Write hashes, image IDs, outcomes, and recovery time to Ed25519-signed JSON; remove containers, volume, and network.

Install / Linux x86-64

One binary. Your Docker engine.

Build from source today. Release binaries are attached to tagged GitHub releases.

cargo install --git https://github.com/B-Divyesh/sf-restore-drill

Evidence, not optimism

Know the recovery time before the incident.

A signed local report ties the exact artifact to the images, probes, and elapsed time used in the rehearsal. Verify it later without a service account.

Status
Passed
Recovery time
02m 41.382s
Artifact
latest.dump · 8.4 GB
SHA-256
8d71…c204
Assertions
7 / 7 passed
Signature
Ed25519 · valid

One-time operator upgrade

Turn a command into a recovery habit.

$39 once · yours for every environment your team operates

  • Ready-to-edit GitHub Actions and cron schedules
  • Four-week recovery scorecard template
  • Failure injection and retention checklist
  • All future Team Kit revisions

The core CLI, signed reports, safety checks, and JSON export stay free.

Buy the Team Kit · $39

No license stored on this device.