Manage every database from one control plane.
Provision, monitor, back up, restore, and migrate self-hosted database instances across your servers from one secure interface.
Open source. Self-hosted. Built for database fleets.
Servers online
3 / 3
DB instances
24
Backups (24h)
18
Database nodes
3 nodes- db-node-eu-013 inst.
- db-node-eu-022 inst.
- db-node-eu-031 inst.
Instances
showing 3 of 24- customer-001MariaDB 11.4
- customer-002PostgreSQL 16
- customer-003MySQL 8.4
Recent operations
- Backup completedcustomer-001 → r2-eu-backups
- Migration in progresscustomer-004 · eu-02 → eu-01
- Instance provisionedcustomer-006 on db-node-eu-01
Fleet capacity
- CPU39%
- Memory57%
- Disk52%
Database operations should not depend on SSH and shell scripts.
When databases are spread across many servers, the operations that keep them healthy stay manual, inconsistent, and invisible. Fleetdock standardizes them.
Manual SSH administration
Every provision, backup, and fix is a shell session on a specific host. Knowledge lives in someone's terminal history.
Scattered backup cron jobs
Backups run from ad-hoc crontabs with no shared status. You find out a job stopped when a restore fails.
Undocumented restore steps
Restores depend on tribal knowledge and one-off scripts, so they are slow and risky exactly when they matter most.
Unsafe credential handling
Database passwords end up in env files, chat messages, and wikis with no rotation or central control.
No central inventory
No single answer to which database runs on which node, at what version, with how much headroom left.
Risky server migrations
Moving a database means hand-assembling dump, transfer, and import steps — and hoping nothing drifts.
Limited capacity visibility
CPU, memory, and disk are checked per host, so you notice a node is full only after it starts failing.
Expensive managed databases
Managed cloud databases solve the operations problem but get costly at scale and take away infrastructure ownership.
The database workflows your team runs, standardized.
Every day-to-day operation becomes a consistent, tracked action instead of a bespoke shell script — the same way on every server.
Provision
Launch isolated MariaDB, MySQL, or PostgreSQL instances with a generated root password, a named data volume, and a published port — as a Docker container on a connected server.
Learn moreMonitor
Track node availability, CPU, memory, disk, and connection metrics from 30-second heartbeats, plus per-instance health and a fleet-wide overview.
Learn moreBack up
Run manual or scheduled backups with each engine's native dump tool — gzipped and checksummed — streamed to S3, Cloudflare R2, or any S3-compatible bucket with retention.
Learn moreRestore
Restore from any backup through a tracked operation that verifies the artifact checksum before touching the target and reports the restored table count.
Learn moreMigrate
Move a database to another instance or server with a backup → restore → verify saga — no hand-assembling dump, transfer, and import scripts.
Learn moreManage credentials
Create application credentials with scoped profiles for external apps. Instance and storage secrets are encrypted at rest; public access routes through the Fleetdock Gateway.
Learn moreExternal access
Expose databases to applications outside your private network through the Fleetdock Gateway — one stable port per database, CIDR allowlists, and disabled by default.
Learn moreOne manager. Multiple database nodes.
Fleetdock separates the control plane from the servers that run your databases, so you operate the whole fleet from one place without opening shells.
Fleetdock Manager
The central control plane. It serves the dashboard and API, tracks every operation, and stores fleet metadata in PostgreSQL.
Fleetdock Agent
A lightweight service on each database node. It enrolls with a single-use token, heartbeats health, and executes approved operations locally.
Database nodes
Your own servers. Provisioned MariaDB, MySQL, and PostgreSQL instances run in isolated Docker containers; the Manager never touches a node except through its agent.
Fleetdock Gateway
An optional HAProxy service colocated with the Manager. It exposes stable, CIDR-gated TCP endpoints for external applications — one port per database, with config owned by the control plane.
Fleetdock Manager
Control plane · UI · API · worker
- MariaDB
- PostgreSQL
- MySQL
- MariaDB
- PostgreSQL
Agents enroll over a single outbound HTTPS connection and execute only approved operations on their own node.
From zero to a managed database in five steps.
Each step maps to a documented procedure with the real commands from the repository.
- 1
Deploy Fleetdock Manager
Bring up the control plane with Docker Compose (a local Postgres is included) or point it at managed PostgreSQL.
- 2
Add a node and generate a token
Register a database server from the dashboard to get a single-use registration token and a ready-to-paste install command.
- 3
Install the Fleetdock Agent
Run one curl command on the server. The agent enrolls, installs the database client tools, and appears within about 30 seconds.
- 4
Provision a database instance
Launch a new MariaDB, MySQL, or PostgreSQL instance in an isolated container, or register a database the server already runs.
- 5
Configure backups and monitoring
Add a backup destination, schedule recurring backups with retention, and watch node and instance health from the overview.
Built for teams running database fleets.
Fleetdock is designed for the specific problem of operating many isolated database instances across many servers.
Database-per-tenant SaaS
Manage isolated databases for every customer from one control plane.
ERP and CRM platforms
Operate many customer database instances across multiple servers.
Hosting providers
Provision and maintain database instances for clients.
Agencies
Manage customer databases without sharing manual server access.
Self-hosted software vendors
Offer managed database operations on customer-controlled infrastructure.
Internal infrastructure teams
Replace scripts, spreadsheets, and undocumented procedures.
Ownership and central management, without the trade-off.
Fleetdock sits between running everything by hand and giving your databases to a managed cloud provider — you keep the infrastructure and gain the operations layer.
| Capability | Manual operations | Managed cloud database | Fleetdock |
|---|---|---|---|
| Infrastructure ownership | Yes | No | Yes |
| Central management | Limited | Yes | Yes |
| Multi-server visibility | Manual | Provider-specific | Yes |
| Automated backups | Custom scripts | Included | Centralized |
| Database migration | Manual | Provider-specific | Controlled workflow |
| Vendor lock-in | Low | High | Low |
| Operational effort | High | Low | Reduced |
Comparison reflects implemented Fleetdock capabilities. Fleetdock does not provide automatic failover or high availability.
Database infrastructure requires more than a login screen.
Fleetdock holds credentials for your entire fleet, so its security model is explicit — including the boundaries it does not yet cover.
Single-use enrollment
Agents register with short-lived, single-use tokens and then authenticate with their own bearer token.
Secrets encrypted at rest
Instance passwords and storage keys are envelope-encrypted with a key you control, and re-wrapped on rotation.
Presigned backup access
Backups stream to object storage through time-limited presigned URLs — agents never receive your storage credentials.
Role-based access control
Owner, admin, operator, and viewer roles plus custom roles from a permission catalog, with scoped API tokens.
CIDR-gated public access
External database access routes through the Fleetdock Gateway with CIDR allowlists — disabled by default, with application credentials instead of root passwords.
Tracked operations & audit
Every mutating action is a tracked operation with status and history, alongside an audit log of changes.
Rotation & revocation
Rotate the encryption and JWT secrets, revoke API tokens, and suspend accounts to cut off sessions immediately.
Known limitations in the current release
The dashboard stores its session token in localStorage, so deploy it behind TLS on a trusted network. SSO/OIDC and 2FA are on the roadmap, not yet available. The agent currently runs as root. Gateway TLS is passthrough only in the current MVP. See the security documentation for the full trust model and mitigations.
Open source. Self-hosted. No database infrastructure lock-in.
Fleetdock is Apache-2.0 licensed and runs entirely on infrastructure you control. Read the code, file issues, and follow the roadmap in the open.
A Go control plane and agent with a Next.js dashboard, released in the open.
Stop managing databases one server at a time.
Connect your infrastructure, standardize database operations, and manage every instance from one control plane.