Problem guide
Back Up Application Data on a Self-Hosted Server
A self-hosted server can hold databases, uploaded files, and Docker volumes, each with different backup needs. A backup that runs on schedule is only part of the job: operators also need suitable retention, a copy in an appropriate destination, and a practical way to restore data.
Updated November 22, 2026 · Reviewed November 22, 2026 by OSRepos editorial team
How to choose
- Match the backup scope: distinguish database snapshots, Docker volumes, and files or repositories managed across multiple machines.
- Check restore workflows: determine whether the tool restores databases, individual files, archives, or only creates backup copies.
- Choose destinations deliberately: compare local storage, object storage, file shares, and SSH-based destinations against the infrastructure available to you.
- Account for network layout: private databases and endpoints that cannot accept inbound connections may need tunnels, agents, or an outbound-polling design.
- Plan for consistency and protection: look for documented ways to handle live application data, encryption, and retention; verify the configuration required for each.
- Assess operational overhead: consider how many services and components must be deployed and maintained, and whether your team needs a web interface, an agent, or a companion container.
- Check project and license evidence: review measured maintenance signals and the license information, noting where the supplied catalog does not establish a fact.
Compare the options
| Project | Stars | Health | License | Language | Self-hosted | Difficulty |
|---|---|---|---|---|---|---|
| databasement: Manage Self-Hosted Database Backups | 2,633 | Active | MIT | PHP | Yes | Moderate |
| borgbackupserver: Manage Borg Backups Across Endpoints | 405 | Active | NOASSERTION | PHP | Yes | Advanced |
| docker-volume-backup: Back Up Docker Volumes to Local or Remote Storage | 4,078 | Active | MPL-2.0 | Go | Yes | Easy |
databasement: Manage Self-Hosted Database Backups
Best match here for centrally scheduled backups and restores of supported databases.
Strength: Broad listed database-engine support with schedules, multiple storage destinations, job monitoring, and database restore workflows in a self-hosted web app.
Trade-off: Its described scope is database backup management; the supplied evidence does not establish general backup coverage for arbitrary files or all Docker volumes.
Good for: Teams managing multiple supported database engines that want retention, storage configuration, job monitoring, and restore workflows in a self-hosted application.
Consider another option if: Your primary requirement is general file or Docker-volume backup rather than database-aware backup and restore.
Read the repository profile →borgbackupserver: Manage Borg Backups Across Endpoints
A strong workflow match for centrally managing Borg backups across endpoint machines, with the license status to resolve first.
Strength: Central Borg job and repository management across endpoint machines, with an agent-polling workflow and browser-based individual-file restore options.
Trade-off: The supplied license field is NOASSERTION, and deploying it involves a PHP/MySQL server application and Python endpoint agents.
Good for: Administrators coordinating Borg backup plans and file restores across machines where endpoints should poll the server rather than accept inbound connections.
Consider another option if: You need a simple Docker-volume companion or want a clearly established license from the supplied catalog before proceeding.
Read the repository profile →docker-volume-backup: Back Up Docker Volumes to Local or Remote Storage
A focused choice for making local or remote archives of Docker volumes alongside an existing deployment.
Strength: Focused companion-container workflow for recurring or one-off Docker volume archives, with local and remote destinations plus optional retention and GPG encryption.
Trade-off: The supplied evidence does not establish database-specific restore management or a central web interface for coordinating jobs across a machine fleet.
Good for: Docker Compose or Swarm operators who want scheduled or one-off volume backups with optional retention, encryption, and notifications.
Consider another option if: You need a database-focused management interface, documented file-by-file restores, or central coordination across a fleet of machines.
Read the repository profile →Start by identifying what needs to be backed up
“Application data” can mean different things on a self-hosted server. A database may need a database-aware dump and a way to restore it to a compatible server. A Docker deployment may store important data in volumes, even when the application itself is easy to recreate. A multi-machine setup may also need a coordinated way to schedule jobs, monitor them, and retrieve selected files.
The three projects below address different parts of that problem. Databasement focuses on database backup and restore workflows. offen/docker-volume-backup focuses on Docker volumes. Borg Backup Server coordinates Borg backups across machines. They are not interchangeable: first identify the data and recovery action you need, then choose a tool whose scope fits.
Database backup management: Databasement
Databasement is the most directly targeted option when the important application data lives in databases. Its listed database support includes MySQL, PostgreSQL, MariaDB, Microsoft SQL Server, MongoDB, SQLite, Firebird, Redis, and Valkey. It brings scheduling, storage configuration, job monitoring, and restore workflows into a self-hosted web application rather than asking an operator to manage a separate process for each database.
The documented workflow includes daily or weekly schedules, time-based or GFS retention, and destinations including local storage, S3-compatible storage, Azure Blob Storage, Samba/SMB, SFTP, and FTP. Backups can be compressed with gzip or zstd, or encrypted with AES-256. The project also describes restore operations to compatible servers, including cross-environment restores, and recurring restore scheduling. These capabilities make it relevant to teams that need more than a copy of a database archive: they need to track jobs and manage restores from a central interface.
Network layout is a useful differentiator. The catalog says Databasement can connect to private database networks using SSH tunnels, or use remote agents for isolated networks without inbound access. It also lists Docker, Docker Compose, Kubernetes with Helm, and native Ubuntu installation as deployment options. Detected repository artifacts include Node.js and PHP dependencies, a Dockerfile and Compose setup, and a Helm chart. Those facts identify possible deployment paths, but they do not establish the exact resource requirements or the effort involved in a particular installation.
Choose Databasement when your central requirement is managing database backups and restores across supported database engines. It is not established by the supplied evidence as a general-purpose backup manager for arbitrary files or all Docker volumes. If your application data is stored in volumes rather than a supported database, assess a volume-focused option instead.
Docker volume backups: offen/docker-volume-backup
offen/docker-volume-backup is a companion container for recurring or one-off backups of Docker volumes. It is aimed at people using Docker Compose or Swarm who want backup behavior alongside an existing deployment, without adding backup code to the application container. Archives can be written locally, remotely, or to both places.
Its listed destinations include compatible S3 storage, WebDAV, Azure Blob Storage, Dropbox, Google Drive, and SSH storage. It can rotate old backups when configured, encrypt archives with GPG, and send success or failure notifications. It can also stop and restart selected containers during a backup to help preserve data consistency. That last capability is a useful option for some deployments, but the evidence does not establish that it automatically guarantees application-consistent backups in every setup. Operators still need to configure the tool for their application and consider how that application handles writes during backup.
Security note for SSH destinations: versions v2.49.1 and earlier failed to validate SSH host identity. Use v2.49.2 or later and configure SSH host-key validation; do not disable it. See the upstream release notes.
This project has a narrower scope than Databasement: the supplied description is about Docker volumes, not database-specific scheduling, database restore workflows, or browsing and restoring individual files. Its evidence says it creates archives and supports backup destinations; it does not describe a central web interface for managing a fleet. For an operator who already uses Docker and needs a configurable volume backup process, that narrower scope can mean a closer fit. For a multi-host setup needing a central job and restore interface, the evidence for this project does not establish that workflow.
Choose this project for Docker volumes when you want local and/or remote archives, optional retention, GPG encryption, and notifications in a companion-container workflow. Do not choose it solely because an application runs in Docker if its important data is actually managed in a database and you need database-aware restore operations.
Coordinating Borg backups across machines: Borg Backup Server
Borg Backup Server provides a self-hosted web interface for scheduling and managing BorgBackup across Linux, Mac, and Windows endpoints. Its workflow differs from a local volume companion: agents poll the server over HTTPS for tasks and send backup data over SSH. The project describes this as avoiding the need for endpoints to accept inbound connections from the server, which may suit machines behind restrictive network boundaries.
The listed features include per-client backup plans, configurable retention, manual runs, cataloged-file browsing, individual-file restores, and archive downloads through a browser. The server can give agents append-only SSH access so they cannot delete or prune existing archives. It can also mirror repositories to S3-compatible storage using rclone. For databases, the description specifically mentions MySQL and PostgreSQL pre-dumps with restore options. That is narrower than Databasement’s listed database-engine coverage, and this project’s wider emphasis is on Borg repositories and endpoint backups rather than managing many database engines through a database-focused application.
The server-side web application uses MySQL and PHP, while backup endpoints use a Python agent. That means adopting it involves more than running a single backup container: plan for the server application, its database, and endpoint agents. The evidence provided does not specify an installation method or detailed system requirements, so the deployment effort for a particular environment is uncertain. The project also lists role-based user management, TOTP two-factor authentication, Prometheus metrics, and Apprise notifications, useful operational features for administrators coordinating multiple clients.
Choose Borg Backup Server when you need central coordination of Borg backups across machines, especially where endpoints should initiate contact rather than accept inbound connections. It is not the strongest match if you want a lightweight Docker-volume-only process or broad database-engine support in a database-specific manager.
Recommendations by scenario
- Several databases and a managed restore workflow: start with Databasement. Its documented engine list, scheduling, storage choices, monitoring, and restore workflows map directly to that need. Confirm that your database type and network arrangement fit the supported setup.
- A Docker Compose or Swarm application whose data is in volumes: consider offen/docker-volume-backup. It is specifically designed as a companion container and offers local or remote archives, optional retention and encryption, and notifications. Check how your application behaves while its volumes are being backed up.
- Borg-based backups across multiple endpoints: consider Borg Backup Server. Its agent-polling workflow, repository operations, and file-level restore options are a better match for a fleet than a volume-only companion. Review its license status before adoption.
- A mixed server: the projects may cover different data layers, but the supplied evidence does not establish an integrated workflow among them. Avoid assuming that one tool’s schedules, retention, alerts, or restore process automatically manage another tool’s backups.
Evidence limits and practical trade-offs
The catalog describes features and deployment approaches, not a comparative benchmark. It does not establish backup speed, storage cost, supported data volumes, recovery-time objectives, or compatibility with a particular application configuration. It also does not provide test notes, so no claim is made here that any project was installed or tested. Before relying on any backup process, verify the project’s current documentation and determine whether you can restore the data you intend to protect. In particular, the evidence does not establish whether a feature works for every deployment simply because it is listed.
Repository health labels are dated snapshots, not forecasts or support commitments. Review current upstream releases before adoption. Borg Backup Server’s license status still needs to be resolved with the project before use.
Finally, choose based on the recovery task rather than the word “backup.” If recovery means restoring a database to another compatible server, Databasement’s scope is the closest fit. If it means keeping Docker volume archives, the Offen companion container is more focused. If it means administering Borg repositories and restoring selected files from several machines, Borg Backup Server addresses that workflow. Each choice leaves different questions to verify in your own environment, particularly around configuration, consistency, and restore procedures.
Sources and further reading
Frequently asked questions
Which project is the best fit for backing up a self-hosted database?
Databasement is the most database-focused option in this comparison. Its listed supported engines, schedules, storage destinations, monitoring, and restore workflows make it a close fit, subject to confirming support for your specific setup.
Can offen/docker-volume-backup back up Docker application data?
It is specifically described as a companion container for recurring or one-off Docker volume backups. It supports local and several remote destinations, but the supplied evidence does not establish database-specific restore workflows or a central fleet-management interface.
Which option works with endpoints that should not accept inbound connections?
Borg Backup Server describes agents polling the server over HTTPS and sending backup data over SSH, so endpoints do not need to accept inbound connections from the server. Databasement also lists SSH tunnels and remote agents for private or isolated database networks.
Do these projects guarantee that a backup can be restored?
No such guarantee is established by the supplied evidence. Databasement and Borg Backup Server describe restore features, while the Docker volume project describes archive creation and related controls. Check the current documentation and verify the restore workflow for your data and configuration.
Are the projects equally clear about licensing?
No. The supplied catalog lists Databasement as MIT and offen/docker-volume-backup as MPL-2.0. Borg Backup Server’s license field is NOASSERTION, which does not establish a license; review its repository terms before adopting it.