CATA Bus MCP Server

# 🚀 Deployment Guide for CATA Bus MCP Server ## FastMCP Cloud Deployment ### Prerequisites 1. FastMCP Cloud account at https://fastmcp.cloud 2. GitHub repository with this code 3. Python 3.11+ ### Quick Deploy 1. **Push to GitHub:** ```bash git remote add origin https://github.com/yourusername/catabus-mcp.git git push -u origin main ``` 2. **Deploy to FastMCP Cloud:** ```bash # Install FastMCP CLI pip install fastmcp-cli # Login to FastMCP Cloud fastmcp login # Deploy from current directory fastmcp deploy # Or deploy from GitHub fastmcp deploy --github yourusername/catabus-mcp ``` 3. **Monitor deployment:** ```bash fastmcp logs catabus-mcp fastmcp status catabus-mcp ``` ## Configuration The `fastmcp.toml` file contains all deployment configuration: - **Memory**: 512MB (sufficient for GTFS data) - **Timeout**: 30 seconds per request - **Caching**: 24-hour TTL for static GTFS data - **Health check**: Every 60 seconds ## Environment Variables Set these in FastMCP Cloud dashboard if needed: ```env TZ=America/New_York # CATA operates in Eastern Time LOG_LEVEL=INFO # Logging verbosity ``` ## Testing Deployment Once deployed, test with: ```bash # List available tools fastmcp test catabus-mcp tools/list # Test a specific tool fastmcp test catabus-mcp list_routes_tool # Health check curl https://your-deployment.fastmcp.cloud/health ``` ## Monitoring ### Logs ```bash fastmcp logs catabus-mcp --follow ``` ### Metrics - Check FastMCP Cloud dashboard for: - Request rate - Response times - Error rate - Memory usage ### Alerts Set up alerts in FastMCP Cloud for: - Error rate > 5% - Response time > 2s - Memory usage > 80% ## Scaling The server automatically scales based on load: - Min instances: 1 - Max instances: 10 - Scale up at 70% CPU - Scale down at 30% CPU ## Troubleshooting ### Common Issues 1. **"No routes loaded"** - Check GTFS feed URL is accessible - Verify cache directory permissions - Check logs for download errors 2. **"No real-time data"** - CATA real-time feeds may be temporarily down - Check network connectivity - Verify 15-second polling interval 3. **"High memory usage"** - Normal with full GTFS data (~50MB) - Consider increasing memory to 1GB if needed ### Debug Mode Enable debug logging: ```bash fastmcp env set LOG_LEVEL=DEBUG fastmcp restart catabus-mcp ``` ## Local Development Run locally before deploying: ```bash # Install dependencies pip install -e . # Run server python -m catabus_mcp.server # Or with FastMCP CLI fastmcp run src/catabus_mcp/server.py:mcp ``` ## CI/CD GitHub Actions automatically: 1. Run tests on push 2. Lint and type check 3. Deploy to FastMCP Cloud on main branch (if configured) To enable auto-deploy: 1. Add `FASTMCP_API_KEY` to GitHub secrets 2. Uncomment deploy step in `.github/workflows/ci.yml` ## Support - FastMCP Cloud docs: https://docs.fastmcp.cloud - CATA Developer Tools: https://catabus.com/developer-tools/ - Issues: https://github.com/yourusername/catabus-mcp/issues