Teradata MCP Server

🚀 Complete Guide to Implementing MCP Server with Teradata

This guide provides a comprehensive walk - through on setting up and running an MCP server using Teradata, covering installation, configuration, and advanced features.

📚 Table of Contents

1. 📦 Install Dependencies
2. 🗃️ Clone the Repository
3. ⚙️ Configure the Environment
4. 🚀 Start the Server
5. 🔌 Integration and Expansion

📖 Main Content

📦 1. Install Dependencies

Before you start, make sure the following tools are properly installed:

• Python: It is recommended to use version 3.8 or higher.
• pip: Used to manage Python packages.

Run the following command in the terminal to install the required packages:

pip install uv

This will install the uv tool for subsequent code running and management.

🗃️ 2. Clone the Repository

Execute the corresponding commands in the terminal according to your operating system to clone the Teradata MCP server repository.

On Windows:

Use PowerShell:

mkdir MCP
cd MCP
git clone https://github.com/Teradata/teradata-mcp-server.git
cd teradata-mcp-server
uv sync
.\.venv\Scripts\activate

On Mac:

Run in the terminal:

mkdir MCP
cd MCP
git clone https://github.com/Teradata/teradata-mcp-server.git
cd teradata-mcp-server
uv sync
source .venv/bin/activate

⚙️ 3. Configure the Environment

1. Create and edit the .env file: In the root directory of the repository, find or create a file named .env and add the following content. Fill in the database credentials and other settings according to your actual situation.

   DATABASE_URI=teradata://username:password@host:1025/databaseName
   SSE_HOST=127.0.0.1
   SSE_PORT=8001
   

2. Optional: Configure LLM credentials If you need to integrate a large language model (LLM), add the following content to the .env file:

   LLM_API_KEY=your_api_key_here
   

🚀 4. Start the Server

Run the following command in the terminal to start the MCP server:

uv run src/teradata_mcp_server/server.py

Make sure you have activated the virtual environment; otherwise, you need to install the dependencies required by server.py globally.

🔌 5. Integration and Expansion

Expose REST endpoints using the mcpo tool

Run the following command to expose the MCP service as a REST API:

uvx mcpo --port 8001 --api-key "your_api_key" -- uv run src/teradata_mcp_server/server.py

In this way, your Teradata tool will become an OpenAPI - compatible HTTP endpoint. Visit http://localhost:8001/docs to view and test the API documentation.

Integrate into Open WebUI

1. Install and run Open WebUI:

   python -m venv ./env
   source ./env/bin/activate  # or .env\Scripts\activate on Windows
   pip install open-webui
   open-webui start
   

2. Add an MCP service connection: Open the settings interface of Open WebUI, add a new data source or tool connection, and enter your MCP server address and API key.

⚠️ 6. Troubleshooting

• Problem: Unable to connect to the database Check if the DATABASE_URI in the .env file is correct. Confirm that the Teradata database service is running and the network configuration allows connections.

• Problem: Module not found error when starting the server Make sure you have activated the virtual environment or installed all necessary dependencies globally.

💪 7. Advanced Features

Deploy to a production environment

Use a reverse proxy (such as Nginx) to manage traffic and SSL certificates. Configure as follows:

server {
    listen 80;
    server_name your_domain.com;

    location / {
        proxy_pass http://localhost:8001;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto $scheme;
    }
}

Monitoring and logging

Integrate Prometheus and Grafana to monitor server performance and health status. Modify server.py to add the corresponding monitoring module and configure data collection.

📝 8. Summary

Through the above steps, you can successfully set up and run a Teradata - based MCP server and expand its functionality to REST endpoints or integrate it into Open WebUI. When encountering problems, carefully check the configuration and log information and troubleshoot step by step to solve the problems.