Snapshots & backups

A snapshot is a point-in-time capture of an environment’s database and files. You take one on demand before a risky change, or let scheduled snapshots run on their own, and you restore from one with a single action. Snapshots are your undo button and the data half of a rollback.

Snapshots vs backups

The two words describe the same underlying capability from different angles.

• A snapshot is an explicit, point-in-time capture — usually one you take yourself right before doing something risky.
• A backup is the scheduled, retained series of snapshots that runs automatically so you’re never without a recent restore point.

Both are restorable the same way. Throughout the dashboard and API you’ll see snapshots for the point-in-time objects and backups for the scheduled capture; restoring either is identical.

Take a snapshot

Capture any environment — production, staging, or a preview — on demand.

Dashboard

Open the environment, go to Snapshots, and choose Take snapshot. The capture runs in the background and appears in the list when it’s ready.

The Snapshots panel for the production environment, listing recent snapshots with capture time, size, and DB/files badges, plus a “Take snapshot” button and per-row “Restore” and “Download” actions.

CLI

Take a snapshot with mf

mf backup create --site acme-store --env production

API

Take a snapshot

curl -X POST https://api.managed.dev/v1/sites/site_01J7.../backups \
  -H "Authorization: Bearer mfk_live_..." \
  -H "Idempotency-Key: 4f1a...b9"

Capture is an async job and returns 202 Accepted. See the backups API reference for the full surface.

Snapshot before anything risky

Take a snapshot immediately before a plugin update, a bulk content change, or a production promotion. If it goes wrong, restore is one click instead of a support ticket.

Restore

A restore brings back the database and files as they were at capture time. You can restore into the same environment to undo a change, and partial restores let you bring back just the database or just the files.

1. Open the environment’s Snapshots list.
2. Pick the snapshot you want.
3. Choose Restore, select database, files, or both, and confirm.

Restore a snapshot

curl -X POST https://api.managed.dev/v1/sites/site_01J7.../snapshots/snap_01J8.../restore \
  -H "Authorization: Bearer mfk_live_..." \
  -H "Idempotency-Key: 7c20...e1" \
  -H "Content-Type: application/json" \
  -d '{ "selectors": ["db", "files"] }'

Restore overwrites the target

Restoring replaces the target environment’s current database and/or files with the snapshot’s. It’s the right tool to recover, but it’s destructive to whatever is there now — take a fresh snapshot first if you might want to come back to the current state.

Download & export

You can download a snapshot to keep an off-platform copy or to inspect it locally — useful for compliance archives or for moving data into a local dev environment.

Get a download URL for a snapshot

curl https://api.managed.dev/v1/sites/site_01J7.../snapshots/snap_01J8.../download \
  -H "Authorization: Bearer mfk_live_..."

Scheduled backups & retention

Scheduled backups run automatically so there’s always a recent restore point without you remembering to capture one. Snapshots are retained on a rolling window and pruned as they age out.

Exact backup cadence and retention windows vary by plan; confirm the figures for your plan in the dashboard.

The backup schedule settings for a site, showing the automatic capture cadence, the retention window, and the list of scheduled snapshots with their expiry dates.

Next steps

Promote & roll backUse a snapshot to bring back code and data together after a bad deploy.

Backups API referenceCapture, restore, and download snapshots programmatically.