An Unraid plugin that turns BorgBackup into a guided backup and restore workflow: configure jobs and repositories, browse archives, verify restores, review reports, automate optional Pre/Post tasks, and monitor results from one web interface. The UI and user manuals are available in English and German.
Borg Backup UI is built for Unraid users who want the reliability of BorgBackup without maintaining custom shell scripts for everyday operation. BorgBackup remains the backup engine; this project adds the Unraid-focused control layer around jobs, storage targets, schedules, restore workflows, checks, and support diagnostics.
Status: public beta, available through Unraid Community Apps. Start with non-critical data and verify a restore before relying on any backup setup. This project is not affiliated with BorgBackup or other Borg UI projects.
Community Apps listing | English manual | Deutsches Handbuch | Unraid support thread | Issues
Follow the German/English tutorial, tasks, reference and concepts. For an outage, use Notfallwiederherstellung or Disaster Recovery. The same manuals are available offline under Help > User Manual.
A walkthrough of reusable Pre/Post scripts, the job wizard, Docker/VM handling, exclusions, retention, multi-file restores, scheduled restore tests, and the Prometheus/Grafana integration.
7:11 minutes | English narration | Recorded with release 2026.10.09.1435.
Watch on YouTube · Pre/Post scripts · Restore files and folders · Grafana
The demo uses generated data in an isolated environment. Grafana shows synthetic metrics; Docker/VM behavior is explained through the job configuration.
- Open Apps in the Unraid web interface.
- Install Python 3 for Unraid if it is not already installed.
- Search for Borg Backup UI and open its Community Apps listing.
- Review the public beta notice and install the plugin.
- Open Settings > Borg Backup UI, start the service, and complete the initial setup.
If the Apps tab is unavailable, follow the official Community Applications instructions.
- Guided job wizard for common Unraid backup workflows.
- Reusable Pre/Post scripts managed in Settings and assigned per job.
- Storage profiles for local paths, USB devices, SMB shares, SSH targets, and Hetzner Storage Box style repositories.
- Docker/VM handling with restart after archive creation, before maintenance.
- Dashboard overview for backup runs, restore proof, storage data, and checks.
- Multi-file and folder restores with configurable safe restore target roots.
- Scheduled restore tests with separate evidence validity and structured reports.
- Optional Prometheus metrics and a supplied Grafana dashboard.
- Notifications through Apprise profiles, Unraid notifications, and email.
- Import/export, support bundle, system health, and auditable migrations.
- Create and edit jobs through a guided multi-step wizard.
- Choose backup type, source paths, repository target, compression, retention, schedule, optional scripts, and runtime behavior.
- Exclude paths, skip folders containing marker files, or import Borg pattern files. Keep archives by retention tiers, age, latest count, or keep all.
- Stop all Docker containers, selected containers, or all except selected containers before a backup.
- Stop selected VMs when a VM backup requires consistent runtime state.
- Restart workloads stopped by the job after successful or warning-only archive creation, before pruning, compaction, and final repository statistics.
- Configure Docker restart priority and delays for service dependencies.
- Schedule jobs in the wizard and optionally show added, modified, and errored files in the live log.
- Review a flow preview before saving the job.
- Manage reusable Bash scripts under Settings > Scripts, with descriptions,
timeouts, syntax validation, duplication, and
.shimport/export. - Assign scripts in the wizard's Pre/Post scripts step. Pre runs before backup preparation; Post runs after maintenance and cleanup according to the selected condition, optionally including failure, skip, or cancellation.
- Use job context and results for custom automation, such as publishing a status for external tools. Normal backup checks work without custom scripts.
- Review script outcomes in the job log and history. A failed Pre script stops the backup from starting; a failed Post script marks the job failed while retaining an archive already created.
- Manage storage targets and profiles for local, USB, SMB, SSH, and Storagebox style repositories.
- Run connection and repository checks from the UI.
- View repository size, compression, deduplication, and check state.
- Keep location grouping consistent across Dashboard, Jobs, History, Reports, Restore Tests, and Browse & Restore.
- Browse Borg archives and select multiple files and folders in one restore.
- Review destination paths, conflicts, and available space before starting.
- Restrict restore targets to administrator-approved safe roots.
- Prevent unsafe backup and restore write operations from using the same repository at the same time.
- Track active restore runs and completed restore history.
- Schedule recurring restore tests or run them manually; configure how long a successful test remains valid independently of the next scheduled run.
- Choose L1-L3 checks and inspect archive metadata, dry-run extraction, and sample extraction results according to the selected level. L3 records SHA-256 hashes of extracted sample data; it does not compare against independent expected hashes or prove full application recovery.
- Dashboard overview for backup runs, restore verification, and growth/check state.
- Backup history and reports for operational review.
- Enable the optional read-only Prometheus exporter under Settings > Integrations > Prometheus & Grafana, then import the supplied Grafana dashboard into your separate monitoring installation.
- Monitor job results, restore evidence, repository storage efficiency, and collector health. See the Prometheus/Grafana setup guide.
- Notification events for successful, warning, failed, skipped, and overdue backup runs.
- Restore test notifications for success, failure, and overdue tests.
- Notification channels are configured through Apprise profiles and can include providers such as ntfy, email-capable services, and other Apprise-supported targets.
- System health and migration status inside Settings.
- Structured migration state and JSONL audit logs.
- Support bundle creation for troubleshooting.
- Import/export flows for jobs, profiles, and secrets.
- Plugin packaging for Unraid with bundled BorgBackup runtime.
- Admin access recovery from the Unraid control page through a short-lived, one-time recovery link.
- English/German UI, light/dark themes, and bundled offline help.
- Unraid with a running array.
- Python 3.10 or newer.
- Recommended on Unraid:
Python 3 for Unraidfrom Community Applications. - Reachable storage targets for the selected backup locations.
BorgBackup is bundled with the plugin package. No pip-based runtime dependency installation is required for normal operation.
- Create the first administrator account.
- Confirm the main data directory in the setup wizard.
- Create or select a storage target and repository.
- Create the first backup job with the guided wizard.
- Run the first backup manually and review Dashboard, History, and Reports.
- Restore a small sample into a separate test folder and verify the result.
- Configure restore tests and notifications for ongoing verification.
For custom automation, create scripts under Settings > Scripts before assigning them to a job. For external monitoring, follow the Prometheus/Grafana setup guide.
For first tests, start with a small test folder or non-critical share. Do not begin with irreplaceable data until backup and restore behavior has been verified on your system.
Screenshots use demo data. The Grafana example uses synthetic metrics.
Backup status, restore verification, repository health, and recent activity in one overview.
Transparent overview of what Borg Backup UI touches, when background activity can happen, and which operations may wake storage or access repositories.
Grouped backup jobs with operational status, next run information, restore verification state, and manual start controls.
A step-by-step workflow for sources, repositories, exclusions, Docker/VM handling, retention, scheduling, Pre/Post scripts, and flow preview. The image shows the script assignment step.
Create and validate reusable Bash scripts before assigning them to jobs. Validation checks syntax without executing the script.
Browse archives, select multiple files and folders, precheck the restore target, and restore into administrator-approved paths.
Set recurring test times and evidence validity, then review results and reports.
Monitor backup status, restore evidence, and storage efficiency with the supplied dashboard. Prometheus and Grafana run separately from the plugin.
The plugin stores configuration below the Unraid flash configuration area and keeps runtime data below the configured data directory. Repository passphrases and profile secrets are stored in dedicated secret files and are excluded from normal logs, public artifacts, and support bundle output.
Backups are only useful when restores are proven. Borg Backup UI therefore treats restore tests and restore history as first-class operating data.
Use the Unraid support thread for setup questions and community discussion. Use GitHub Issues for bugs, feature requests, and design discussions. When reporting runtime problems, include the generated support bundle where possible and remove private infrastructure details before sharing logs publicly.
Borg Backup UI is free and open source.
If Borg Backup UI makes your backups easier and you'd like to support its continued development and maintenance, you can support BorgForge on Ko-fi.
Donations are completely optional. Bug reports, feedback, feature suggestions, and contributions are just as appreciated!
- Project: MIT, see LICENSE
- Bundled BorgBackup: BSD-3-Clause, see runtime/licenses/borg/LICENSE
- Bundled Apprise notification runtime: BSD-2-Clause and permissive Python dependency licenses, see runtime/licenses/THIRD-PARTY-NOTICES.md