Databases
Manage Backups
View backups and restore your managed database to a point in time
Manage Backups
Managed Databases with automated backups enabled provide continuous WAL archiving and periodic full backups, enabling point-in-time recovery to any second within the retention window.
Backup Overview
When automated backups are enabled on a database:
- Full backups are taken periodically
- WAL (Write-Ahead Log) archiving captures every transaction continuously
- Default retention is 7 days
- Backups are stored in a separate region for durability
Viewing Backups
- Navigate to Databases in the sidebar
- Click the Backups tab
- Expand a database card to see its backup history
Each backup entry shows:
| Column | Description |
|---|---|
| Label | Backup identifier |
| Type | FULL (complete backup) or WAL (incremental log segment) |
| Taken | When the backup was created (relative time) |
| Size | Backup file size |
| Duration | How long the backup took to complete |
Recovery Window
Each database with backups displays a recovery window:
- Oldest recovery point — earliest timestamp you can restore to
- Latest recovery point — most recent recoverable timestamp (near real-time)
You can restore to any point in time between these two bounds.
Restoring a Database
Point-in-time restore creates a new database from an existing database's backup. The source database is not modified.
Restore Steps
- Navigate to Databases in the sidebar
- Click Create Database
- Select Restore from Backup mode
- Choose the Source Database from the dropdown (only databases with backups are listed)
- Select the Restore Point:
- Default: latest recoverable time
- Custom: pick any date/time within the recovery window
- Use Reset to Latest to snap back to the most recent point
- Configure the new database name and settings
- Click Create Database
The restored database provisions as a new cluster with data recovered up to your chosen point in time.
Backup Status Indicators
| Status | Meaning |
|---|---|
| ok | Backups running normally |
| WAL archiving active | Continuous log shipping is working |
| Alert | An issue with the backup pipeline — check the status message |
Best Practices
- Enable backups on all production databases — the 7-day retention window provides a safety net against accidental data loss
- Test restores periodically — verify that your recovery process works before you need it in an emergency
- Note the recovery window — if the oldest recovery point is less than 7 days ago, the database was recently created or backups were recently enabled
Next Steps
- Manage Database — monitor status and cluster nodes
- Create a Database — launch a new database