Which integrations are available for this server?

Supports querying and schema exploration for MariaDB databases, enabling tools to list tables and perform parallel SQL execution. Provides tools to execute SQL queries, list tables, and explore schemas in MySQL databases, supporting simultaneous operations across multiple connections. Allows for querying PostgreSQL databases, exploring schemas, listing tables, and describing table structures, including support for parallel queries across multiple instances. Enables interaction with SQLite database files, providing tools for executing SQL queries, listing tables, and describing database structures.

How do I use Multi-Database MCP Server?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Multi-Database MCP Server Compare the users table schema in postgres_db and mysql_db" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Multi-Database MCP Server

An MCP (Model Context Protocol) server that provides tools to access multiple databases simultaneously. The server runs in Docker and supports querying multiple databases of different types in parallel. This server was inspired by needs to compare databases or support migrations.

Features

Multiple Database Types: Support for PostgreSQL, MySQL/MariaDB, SQL Server, and SQLite
Multiple Database Support: Connect to and query multiple databases simultaneously (even different types)
Parallel Queries: Execute the same query across multiple databases at once
Schema Exploration: List databases, schemas, tables, and describe table structures
Dockerized: Runs in a Docker container for easy deployment
Connection Pooling: Efficient connection management for all database types
Flexible Configuration: Organize databases by type

Available Tools

list_databases - List all configured database connections
query_database - Execute a SQL query on a specific database
list_tables - List all tables in a database schema
describe_table - Get detailed schema information for a table
list_schemas - List all schemas in a database
query_multiple_databases - Execute the same query on multiple databases simultaneously

Setup

1. Create Database Configuration

Copy the example configuration file and edit it with your database credentials:

cp databases.json.example databases.json

Edit databases.json with your database connection details. Organize databases by type at the root level:

{ "postgresql": { "postgres_db": { "host": "localhost", "port": 5432, "user": "postgres", "password": "your_password", "database": "database1" }, "another_postgres": { "host": "remote.example.com", "port": 5432, "user": "admin", "password": "secret", "database": "production" } }, "mysql": { "mysql_db": { "host": "localhost", "port": 3306, "user": "root", "password": "your_password", "database": "mydatabase" } }, "sqlserver": { "sqlserver_db": { "host": "localhost", "port": 1433, "user": "sa", "password": "your_password", "database": "MyDatabase" } }, "sqlite": { "sqlite_db": { "database": "/path/to/database.db" } } }

Supported Database Types

PostgreSQL ("postgresql", "postgres", or "pg")
MySQL/MariaDB ("mysql" or "mariadb")
SQL Server ("sqlserver", "mssql", or "sql server")
SQLite ("sqlite" or "sqlite3")

2. Build and Run with Docker Compose

docker-compose up --build

3. Build and Run with Docker

# Build the image docker build -t multidb-mcp-server . # Run the container docker run -it \ -v $(pwd)/databases.json:/app/databases.json:ro \ multidb-mcp-server

4. Docker Connection Tips

When configuring database connections from within Docker, keep these tips in mind:

Network Connections (PostgreSQL, MySQL, SQL Server)

Accessing host machine databases: Use host.docker.internal as the hostname (Windows/Mac) or 172.17.0.1 (Linux)
Accessing other containers: Use the container name or service name from docker-compose.yml
Accessing remote databases: Use the actual hostname or IP address

Example for accessing PostgreSQL on host machine:

{ "postgresql": { "local_db": { "host": "host.docker.internal", "port": 5432, "user": "postgres", "password": "password", "database": "mydb" } } }

Example for accessing database in another Docker container:

{ "postgresql": { "container_db": { "host": "postgres-container", "port": 5432, "user": "postgres", "password": "password", "database": "mydb" } } }

Example for accessing SQL Server on host machine:

{ "sqlserver": { "sqlserver_db": { "host": "host.docker.internal", "port": 1433, "user": "sa", "password": "password", "database": "MyDatabase" } } }

Note for SQL Server: Ensure that SQL Server is configured to accept TCP/IP connections and that the SQL Server Browser service is running if using named instances. The default port is 1433.

File Paths (SQLite)

SQLite databases on host: Mount the directory containing the database file as a volume and use the container path
SQLite databases in container: Use absolute paths within the container

Example Docker run command with SQLite volume:

docker run -it \ -v $(pwd)/databases.json:/app/databases.json:ro \ -v $(pwd)/data:/app/data:ro \ multidb-mcp-server

Example SQLite configuration:

{ "sqlite": { "local_db": { "database": "/app/data/mydatabase.db" } } }

Note: The path /app/data/mydatabase.db is the path inside the container, which maps to ./data/mydatabase.db on your host machine via the volume mount.

Usage

The server communicates via stdio using the MCP protocol. Connect your MCP client to the Docker container's stdio streams.

Example MCP Client Configuration

If using with an MCP client, configure it to connect to the Docker container:

{ "mcpServers": { "multidb": { "command": "docker", "args": [ "run", "-i", "-v", "[YOUR/PATH]/databases.json:/app/databases.json:ro", "multidb-mcp-server" ] } } }

Tool Examples

List All Databases

{ "tool": "list_databases", "arguments": {} }

Query a Single Database

{ "tool": "query_database", "arguments": { "database_name": "db1", "query": "SELECT * FROM users LIMIT 10" } }

List Tables

{ "tool": "list_tables", "arguments": { "database_name": "db1", "schema": "public" } }

Describe a Table

{ "tool": "describe_table", "arguments": { "database_name": "db1", "table_name": "users", "schema": "public" } }

Query Multiple Databases Simultaneously

You can query multiple databases of different types simultaneously:

{ "tool": "query_multiple_databases", "arguments": { "database_names": ["db1", "db2", "sqlserver_db"], "query": "SELECT COUNT(*) as total FROM users" } }

Note: When querying multiple databases of different types, ensure the SQL syntax is compatible across all database types, or use database-specific queries for each database separately.

Development

Local Development (without Docker)

Install dependencies:

pip install -r requirements.txt

Set environment variable:

export DB_CONFIG_PATH=./databases.json

Run the server:

python server.py

Database-Specific Notes

PostgreSQL

Default port: 5432
Default schema: public
Uses asyncpg for async operations

MySQL/MariaDB

Default port: 3306
Schema parameter is optional (uses current database)
Uses aiomysql for async operations

SQL Server

Default port: 1433
Default schema: dbo
Requires ODBC Driver 17 for SQL Server (installed in Docker image)
Uses pyodbc with asyncio wrapper
Important: SQL Server must be configured to accept TCP/IP connections
For named instances, ensure SQL Server Browser service is running
Use host.docker.internal to connect to SQL Server on the host machine from Docker

SQLite

No network connection required
Use database field to specify file path (or :memory: for in-memory)
Default schema: main
Uses aiosqlite for async operations

Security Notes

Never commit - It contains sensitive credentials
Use environment variables or secrets management in production
Consider using SSL/TLS connections for remote databases
Limit network access to the Docker container
For SQLite, ensure file paths are secure and accessible

Troubleshooting

Connection Issues

Verify database credentials in databases.json
Ensure databases are accessible from the Docker container
Check network connectivity (use docker network for container-to-container communication)

Permission Issues

Ensure the databases.json file has proper read permissions
Check that the Docker volume mount is correct

License

MIT

Multi-Database MCP Server

Multi-Database MCP Server

Features

Available Tools

Setup

1. Create Database Configuration

Supported Database Types

2. Build and Run with Docker Compose

3. Build and Run with Docker

4. Docker Connection Tips

Network Connections (PostgreSQL, MySQL, SQL Server)

File Paths (SQLite)

Usage

Example MCP Client Configuration

Tool Examples

List All Databases

Query a Single Database

List Tables

Describe a Table

Query Multiple Databases Simultaneously

Development

Local Development (without Docker)

Database-Specific Notes

PostgreSQL

MySQL/MariaDB

SQL Server

SQLite

Security Notes

Troubleshooting

Connection Issues

Permission Issues

License

Resources

New MCP Servers

Latest Blog Posts

MCP directory API