mcp-mermiad

# Railway Deployment Guide ## Environment Variables to Set in Railway Dashboard When deploying to Railway, configure these environment variables in the Railway dashboard: | Variable | Value | Description | |----------|-------|-------------| | `NODE_ENV` | `production` | Node.js environment | | `PORT` | `3000` | HTTP server port (Railway auto-assigns) | | `PUPPETEER_EXECUTABLE_PATH` | `/usr/bin/chromium` | Path to Chromium in container | | `PUPPETEER_SKIP_CHROMIUM_DOWNLOAD` | `true` | Skip Puppeteer's Chromium download | ## Deployment Steps ### 1. Install Railway CLI ```bash npm install -g @railway/cli ``` ### 2. Login to Railway ```bash railway login ``` ### 3. Initialize Project ```bash cd /Users/narasimharao.ponnada/Documents/Mermaid railway init ``` ### 4. Deploy ```bash railway up ``` ### 5. Set Environment Variables In Railway Dashboard: - Go to your project - Click on Variables tab - Add the environment variables listed above ### 6. Get Public URL ```bash railway domain ``` Or in Railway Dashboard: - Go to Settings - Generate domain under "Domains" ## Local Testing with Docker ### Build and Run ```bash # Build the image docker build -t mermaid-mcp-connector . # Run with docker-compose docker-compose up -d # Or run directly docker run -p 3000:3000 -p 3001:3001 \ -e NODE_ENV=production \ -e PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium \ -e PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true \ mermaid-mcp-connector ``` ### Test Endpoints ```bash # Health check curl http://localhost:3000/health # Generate diagram curl -X POST http://localhost:3000/api/diagram/generate \ -H "Content-Type: application/json" \ -d '{ "code": "graph TD\n A[Start] --> B[End]", "format": "svg" }' ``` ### View Logs ```bash # docker-compose docker-compose logs -f # Docker docker logs -f mermaid-mcp-connector ``` ### Stop Container ```bash # docker-compose docker-compose down # Docker docker stop mermaid-mcp-connector docker rm mermaid-mcp-connector ``` ## ChatGPT Integration Once deployed to Railway: 1. **Get Railway URL**: e.g., `https://your-app.railway.app` 2. **Create Custom GPT**: - Go to ChatGPT → Create a GPT - Add action with this OpenAPI spec: ```yaml openapi: 3.0.0 info: title: Mermaid Diagram Generator version: 1.0.0 servers: - url: https://your-app.railway.app paths: /api/diagram/generate: post: operationId: generateDiagram summary: Generate Mermaid diagram requestBody: required: true content: application/json: schema: type: object properties: code: type: string description: Mermaid diagram code format: type: string enum: [svg, png] default: svg responses: '200': description: Diagram generated successfully ``` 3. **Test**: Ask ChatGPT to generate diagrams! ## Troubleshooting ### Puppeteer/Chromium Issues If Puppeteer fails to launch: - Check `PUPPETEER_EXECUTABLE_PATH` is set to `/usr/bin/chromium` - Verify Chromium dependencies in Dockerfile - Check container logs for missing libraries ### Memory Issues Railway free tier has memory limits: - Reduce `maxBrowserInstances` in puppeteer-renderer.ts - Monitor memory usage in Railway dashboard - Consider upgrading Railway plan if needed ### Slow Cold Starts First request may take 3-5 seconds: - This is normal (Chromium initialization) - Subsequent requests are faster (300-500ms) - Consider keeping service warm with health checks ## Cost Estimation Railway Pricing: - **Free Tier**: $5 credit/month, good for testing - **Hobby**: $5/month base + usage - **Pro**: $20/month base + usage Typical usage: - ~100 diagram generations/day ≈ $0.50/month - Keep within free tier for testing - Upgrade if needed for production ## Monitoring Railway provides: - CPU/Memory metrics - Request logs - Deployment history - Custom metrics via health endpoint Monitor these metrics: - Response time (target: <500ms) - Error rate (target: <1%) - Memory usage (target: <512MB) - CPU usage (target: <50%)