# Multi-Environment Setup for Bigeye MCP Server
This guide explains how to configure and run the Bigeye MCP Server for multiple Bigeye environments (demo.bigeye.com and app.bigeye.com) using Docker.
## Overview
The multi-environment setup allows you to:
- Connect to different Bigeye instances (demo and production)
- Use different workspace IDs for each environment
- Manage separate API keys for each environment
- Run multiple MCP server instances simultaneously
## Configuration Files
### Environment-Specific Docker Compose Files
- `docker-compose.yml` - Base configuration
- `docker-compose.demo.yml` - Demo environment overrides
- `docker-compose.app.yml` - App (production) environment overrides
### Environment Variable Files
- `.env.demo` - Demo environment variables
- `.env.app` - App environment variables
Copy and configure these files:
```bash
cp .env.example .env.demo
cp .env.example .env.app
```
Edit each file with your environment-specific values:
- `BIGEYE_DEMO_API_KEY` / `BIGEYE_APP_API_KEY`
- `BIGEYE_DEMO_WORKSPACE_ID` / `BIGEYE_APP_WORKSPACE_ID`
## Running the Servers
### Using Helper Scripts
Run the demo environment:
```bash
./scripts/run-demo.sh
```
Run the app environment:
```bash
./scripts/run-app.sh
```
### Using Docker Compose Directly
For demo environment:
```bash
docker-compose -f docker-compose.yml -f docker-compose.demo.yml --env-file .env.demo up bigeye-mcp-demo
```
For app environment:
```bash
docker-compose -f docker-compose.yml -f docker-compose.app.yml --env-file .env.app up bigeye-mcp-app
```
## Claude Desktop Configuration
Add the following to your Claude Desktop configuration file (`~/Library/Application Support/Claude/claude_desktop_config.json` on macOS):
### Option 1: Direct Docker Configuration
```json
{
"mcpServers": {
"bigeye-demo": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--env", "BIGEYE_API_KEY=${BIGEYE_API_KEY}",
"--env", "BIGEYE_BASE_URL=${BIGEYE_BASE_URL}",
"--env", "BIGEYE_WORKSPACE_ID=${BIGEYE_WORKSPACE_ID}",
"--env", "BIGEYE_DEBUG=${BIGEYE_DEBUG:-false}",
"bigeye-mcp-server:latest",
"python",
"server.py"
],
"env": {
"BIGEYE_API_KEY": "your_demo_api_key_here",
"BIGEYE_BASE_URL": "https://demo.bigeye.com",
"BIGEYE_WORKSPACE_ID": "your_demo_workspace_id_here",
"BIGEYE_DEBUG": "false"
}
},
"bigeye-app": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--env", "BIGEYE_API_KEY=${BIGEYE_API_KEY}",
"--env", "BIGEYE_BASE_URL=${BIGEYE_BASE_URL}",
"--env", "BIGEYE_WORKSPACE_ID=${BIGEYE_WORKSPACE_ID}",
"--env", "BIGEYE_DEBUG=${BIGEYE_DEBUG:-false}",
"bigeye-mcp-server:latest",
"python",
"server.py"
],
"env": {
"BIGEYE_API_KEY": "your_app_api_key_here",
"BIGEYE_BASE_URL": "https://app.bigeye.com",
"BIGEYE_WORKSPACE_ID": "your_app_workspace_id_here",
"BIGEYE_DEBUG": "false"
}
}
}
}
```
### Option 2: Docker Compose Configuration
```json
{
"mcpServers": {
"bigeye-demo-compose": {
"command": "docker-compose",
"args": [
"-f", "/path/to/bigeye_mcp_server/docker-compose.yml",
"-f", "/path/to/bigeye_mcp_server/docker-compose.demo.yml",
"--env-file", "/path/to/bigeye_mcp_server/.env.demo",
"run",
"--rm",
"-T",
"bigeye-mcp-demo",
"python",
"server.py"
]
},
"bigeye-app-compose": {
"command": "docker-compose",
"args": [
"-f", "/path/to/bigeye_mcp_server/docker-compose.yml",
"-f", "/path/to/bigeye_mcp_server/docker-compose.app.yml",
"--env-file", "/path/to/bigeye_mcp_server/.env.app",
"run",
"--rm",
"-T",
"bigeye-mcp-app",
"python",
"server.py"
]
}
}
}
```
Remember to replace `/path/to/bigeye_mcp_server` with the actual path to your project directory.
## Environment Variables
The following environment variables are used:
- `BIGEYE_API_KEY` - Your Bigeye API key
- `BIGEYE_BASE_URL` - The Bigeye instance URL (https://demo.bigeye.com or https://app.bigeye.com)
- `BIGEYE_WORKSPACE_ID` - Your Bigeye workspace ID
- `BIGEYE_DEBUG` - Enable debug logging (true/false)
## Building the Docker Image
Before running any environment, build the Docker image:
```bash
docker build -t bigeye-mcp-server:latest .
```
Or use the build script:
```bash
./scripts/build.sh
```
## Testing
To test a specific environment:
```bash
# Test demo environment
docker run --rm -it \
-e BIGEYE_API_KEY="your_demo_key" \
-e BIGEYE_BASE_URL="https://demo.bigeye.com" \
-e BIGEYE_WORKSPACE_ID="your_demo_workspace_id" \
bigeye-mcp-server:latest \
python server.py
# Test app environment
docker run --rm -it \
-e BIGEYE_API_KEY="your_app_key" \
-e BIGEYE_BASE_URL="https://app.bigeye.com" \
-e BIGEYE_WORKSPACE_ID="your_app_workspace_id" \
bigeye-mcp-server:latest \
python server.py
```
## Troubleshooting
1. **Container name conflicts**: If you get an error about container names already in use, stop the existing container:
```bash
docker stop bigeye-mcp-server-demo
docker stop bigeye-mcp-server-app
```
2. **Environment variables not loading**: Ensure your `.env.demo` and `.env.app` files are in the correct location and have the proper format.
3. **Connection issues**: Verify that your API keys and workspace IDs are correct for each environment.
4. **Claude Desktop not recognizing the server**: Restart Claude Desktop after updating the configuration file.
