Backup schedules and policies
Schedule backups for every game server from the FeatherPanel admin area, pick targets and timezones, control retention, and find node-level Wings backup policies.
On this page
FeatherPanel has two kinds of admin backups:
- Backup Schedules make normal server backups for many servers at once. They are the same backups users see on their Backups page.
- Wings Backups archive whole nodes to SFTP or S3. They are covered in Node backups.
Creating a backup schedule
Admin area, Backup Schedules, Create schedule.
| Section | Settings |
|---|---|
| Targets | Name, and the scope: All servers, All servers on a node, or Specific servers. The target count updates as you choose. |
| Schedule | Cron fields (minute, hour, day of month, month, day of week) and the timezone they are read in. 0 3 * * * with Europe/Paris is 03:00 Paris time. |
| Backup type | Server files, Databases or Full server. Ignored files use gitignore-style patterns; empty means the server’s .featherpanelignore. A full backup can include panel metadata, encrypted secrets and the activity log. |
| Options | Concurrency (1 to 5 servers per batch), Active, Only when online (skip stopped servers), Notify on failure (post an admin announcement). |
For multi-server scopes, database backups always include every database on each server.
The panel cron checks due schedules every minute and starts a backup on each target. Wings creates the archives in the background. Each schedule row shows the last and next run; History lists past runs, and Run now starts one at once, which is a good way to test a new schedule.
Retention
Schedules do not have their own retention. Each server’s backup limit still applies, with the retention mode from its Limits tab or the panel default:
| Mode | When the server is full |
|---|---|
Hard limit (hard_limit, default) | New backups are refused. |
| FIFO rolling | The oldest unlocked backup is deleted to make room. |
The default is the server_backup_retention_mode setting (Settings, Servers). With
server_allow_user_backup_policy_edit on, owners can change the mode for their own servers, but only admins change the
backup limit. Locked and in-progress backups are never rotated away.
Storage uses each server’s node backup adapter (local, S3 and the others in Server backup drivers). Plan off-peak times and low concurrency on busy nodes.
API
| Method | Path | Permission |
|---|---|---|
| GET | /api/admin/backup-schedules | admin.backup_schedules.view |
| PUT | /api/admin/backup-schedules | admin.backup_schedules.create |
| PATCH, DELETE | /api/admin/backup-schedules/{id} | admin.backup_schedules.edit, admin.backup_schedules.delete |
| POST | /api/admin/backup-schedules/{id}/run | admin.backup_schedules.run |
| GET | /api/admin/backup-schedules/{id}/runs | admin.backup_schedules.view |
| POST | /api/admin/backup-schedules/preview-targets | admin.backup_schedules.view |
Frequently asked questions
Each backup goes to the backup adapter configured on the server's node, the same place as manual backups. There is no separate destination for schedules.
Only through the server's own retention. With FIFO rolling the oldest unlocked backup is removed to make room; with a hard limit new backups fail once the server is full.