---
title: blossom backup:restore
url: "/docs/reference/cli/backup/restore/"
canonical_url: https://docs.blossom88.com/docs/reference/cli/backup/restore/
markdown_url: https://docs.blossom88.com/docs/reference/cli/backup/restore.md
summary: Restore a managed data recovery backup
categories:
- cli
- reference
---

# blossom backup:restore

Restore a managed data recovery backup

## Usage

    blossom backup:restore --id ID [--data DATA] [--follow] [--timestamps]

## Flags

| Flag | Description |
| --- | --- |
| `--id (required)` | Backup id |
| `--data, -d` | YAML or JSON restore data, a path to a data file, or an HTTP(S) URL |
| `--follow, -f` | Follow backup restore logs |
| `--timestamps` | Show log timestamps. Default: true. |

Restore a completed managed data recovery backup into a new Managed Database instance
or Managed Cache instance. The backup format determines the restored resource type.

    blossom backup:restore --id backup-100 --data restore.yml --follow

restore.yml:

```yaml
restore:
  name: restored-postgres
  location_id: location-100
  availability: shared
  size: medium
  port: 5433
  target_time: 2026-07-17T14:30:00.123456Z
```

`target_time` is optional and supported for PostgreSQL only. Use an ISO 8601
timestamp with a time zone to recover through archived WAL to that exact time.
Omit it to restore the selected backup. The source resource is not changed.

Choose `availability: shared` to use your Data Capacity, or `dedicated` for an
instance on its own Server. Shared is the default where the Location supports
it; otherwise Dedicated is the default. Each restored instance has its own
storage and endpoint.

For Shared, `size` accepts `small`, `medium`, `large`, or a custom
`{cpu_vcpus: 0.5, memory_gb: 1}`. For Dedicated Databases, use
`dedicated_size: 2vcpu-4gb`, `4vcpu-8gb`, or `8vcpu-16gb` instead. Dedicated
Caches use Standard compute. Restores use the same capacity limits as creation.