Database MCP Servers Setup Guide

What is Database MCP Servers Setup Guide?

This guide explains how to set up Model Context Protocol (MCP) servers for various AWS database services—Amazon RDS MySQL, DocumentDB, DynamoDB, Redshift, and ElastiCache Redis—to use with Amazon Bedrock models.

How to use Database MCP Servers Setup Guide?

After meeting prerequisites, configure credentials using a template, validate database connections, run the secure setup script (setup_db_mcp_servers_secure.sh), and test the servers with test_mcp_servers.sh. Then issue natural language prompts through Amazon Bedrock to interact with databases.

Key features of Database MCP Servers Setup Guide

• Supports five AWS database services: MySQL, DocumentDB, DynamoDB, Redshift, and Redis.
• Provides a validation script to check database connectivity before setup.
• Automates installation, configuration, and restart of MCP services.
• Includes a test script to verify all servers are working correctly.
• Offers example prompts for each database type.
• Emphasizes security best practices like credential management and least privilege.

Use cases of Database MCP Servers Setup Guide

• Query MySQL tables and describe schemas using natural language.
• Analyze sales data by region in Redshift with SQL prompts.
• Retrieve specific items from DynamoDB tables based on key conditions.
• Manage Redis keys and set members through Bedrock models.
• Find documents and explore collection schemas in MongoDB/DocumentDB.

FAQ from Database MCP Servers Setup Guide

What prerequisites are needed?

AWS access with appropriate permissions, existing database instances, Node.js, npm, Python 3, Git, and optionally Docker.

How do I provide database credentials?

Copy db_credentials_template.sh to set_db_credentials.sh, then edit the file with your actual database connection info.

How can I test if the servers are set up correctly?

Run test_mcp_servers.sh after the setup script completes; it will verify all MCP servers are working.

What security measures are recommended?

Never commit credentials to version control. Use AWS Secrets Manager or Parameter Store for credential management, IAM roles instead of access keys, and apply least privilege access.

Where does the MCP service run?

The MCP service runs at http://127.0.0.1:7002, and the Chatbot UI is accessible at http://localhost:8502.