Railway Deployment

Railway Deployment Skill

When to Use This Skill

Use this skill when:

1. Deploying applications to Railway platform
2. Configuring nixpacks.toml for custom builds
3. Managing environment variables across services
4. Debugging deployment failures
5. Setting up multi-service Railway projects
6. Troubleshooting production issues
7. Configuring database migrations on Railway
8. Setting up health checks and monitoring

Core Principles

1. Build Configuration First

Always start with proper nixpacks configuration:

• Specify all system dependencies in nixPkgs
• Define build phases explicitly
• Test locally with nixpacks CLI when possible

2. Environment Variable Management

• Use Railway's service references: ${{ServiceName.VARIABLE}}
• Never hardcode secrets or URLs
• Use Railway secrets for sensitive data
• Configure all environment variables before deployment

3. Port Binding

Always bind to Railway's PORT environment variable:

port = int(os.environ.get('PORT', 8000))

4. Health Checks

Implement health check endpoints for Railway to monitor:

@app.get("/health")
async def health_check():
    return {"status": "healthy"}

Common Deployment Workflows

Initial Deployment

1. Create nixpacks.toml in project root
2. Configure environment variables in Railway dashboard
3. Add Procfile or start command (optional if using nixpacks)
4. Deploy via GitHub integration or CLI
5. Monitor build logs for errors
6. Verify deployment with health check

Multi-Service Deployment

project/
├── backend/
│   ├── nixpacks.toml
│   └── requirements.txt
├── frontend/
│   ├── nixpacks.toml
│   └── package.json
└── railway.toml  # Optional: multi-service config

Configure service references:

# In frontend service
VITE_API_URL=https://${{backend.RAILWAY_PUBLIC_DOMAIN}}

# In backend service
FRONTEND_URL=https://${{frontend.RAILWAY_PUBLIC_DOMAIN}}

Database Migration Strategy

# nixpacks.toml
[phases.build]
cmds = [
  'pip install -r requirements.txt',
  'alembic upgrade head'  # Run migrations during build
]

Alternative: Separate migration service

# Create one-off migration job
railway run alembic upgrade head

nixpacks Configuration Patterns

Python + PostgreSQL

[phases.setup]
nixPkgs = ['python310', 'postgresql']
nixLibs = ['libpq']

[phases.install]
cmds = ['pip install -r requirements.txt']

[start]
cmd = 'uvicorn main:app --host 0.0.0.0 --port $PORT'

Node.js + TypeScript

[phases.setup]
nixPkgs = ['nodejs-18_x']

[phases.install]
cmds = ['npm ci']

[phases.build]
cmds = ['npm run build']

[start]
cmd = 'node dist/index.js'

Full-Stack (Python Backend + React Frontend)

[phases.setup]
nixPkgs = ['python310', 'nodejs-18_x', 'postgresql']
nixLibs = ['libpq']

[phases.install]
cmds = [
  'pip install -r requirements.txt',
  'cd frontend && npm ci'
]

[phases.build]
cmds = [
  'cd frontend && npm run build',
  'alembic upgrade head'
]

[start]
cmd = 'gunicorn main:app --workers 4 --worker-class uvicorn.workers.UvicornWorker --bind 0.0.0.0:$PORT'

Environment Variable Patterns

Database Connection

DATABASE_URL=${{Postgres.DATABASE_URL}}
DATABASE_PRIVATE_URL=${{Postgres.DATABASE_PRIVATE_URL}}

Redis Connection

REDIS_URL=${{Redis.REDIS_URL}}
REDIS_PRIVATE_URL=${{Redis.REDIS_PRIVATE_URL}}

Service-to-Service Communication

# Use private networking for internal communication
BACKEND_PRIVATE_URL=http://${{backend.RAILWAY_PRIVATE_DOMAIN}}
BACKEND_PUBLIC_URL=https://${{backend.RAILWAY_PUBLIC_DOMAIN}}

Application Configuration

# Environment
ENVIRONMENT=production
DEBUG=false

# Security
JWT_SECRET_KEY=${{secrets.JWT_SECRET}}
ALLOWED_HOSTS=${{RAILWAY_PUBLIC_DOMAIN}}

# CORS
CORS_ORIGINS=https://${{frontend.RAILWAY_PUBLIC_DOMAIN}}

Troubleshooting Guide

Build Failures

Error: "Package not found"

• Add missing package to nixPkgs in nixpacks.toml
• Check nixpkgs search: https://search.nixos.org/packages

Error: "Command failed"

• Check build logs for specific error
• Verify commands work locally
• Ensure correct working directory

Runtime Failures

Error: "Application failed to respond"

• Verify binding to 0.0.0.0 not localhost
• Check PORT environment variable usage
• Ensure health check endpoint exists

Error: "Database connection failed"

• Verify DATABASE_URL is set
• Check database service is running
• Use private URL for better performance

Deployment Best Practices

1. Always test locally first

   nixpacks build . --name myapp
   docker run -p 8000:8000 myapp
   

2. Use railway.toml for monorepos

   [build]
   builder = "nixpacks"
   buildCommand = "cd backend && pip install -r requirements.txt"
   
   [deploy]
   startCommand = "cd backend && uvicorn main:app --host 0.0.0.0 --port $PORT"
   restartPolicyType = "on-failure"
   

3. Set up health checks

   @app.get("/health")
   async def health():
       # Check database connection
       try:
           await db.execute("SELECT 1")
           return {"status": "healthy", "database": "connected"}
       except:
           return {"status": "unhealthy", "database": "disconnected"}
   

4. Configure logging

   import logging
   logging.basicConfig(
       level=logging.INFO,
       format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
   )
   

5. Use Railway CLI for debugging

   railway login
   railway link  # Link to your project
   railway logs  # View logs
   railway run python manage.py shell  # Run commands
   

Production Deployment Checklist

Pre-Deployment

• All environment variables configured in Railway
• Database migrations tested locally
• nixpacks.toml configured and tested
• Health check endpoint implemented
• CORS configured for production domains
• Secrets stored in Railway secrets (not env vars)
• Logging configured
• Error tracking configured (Sentry, etc.)

During Deployment

• Monitor build logs in Railway dashboard
• Watch for build phase completion
• Verify deployment status shows "Active"
• Check health endpoint responds

Post-Deployment

• Test critical user flows
• Verify database connectivity
• Check external API integrations
• Monitor error rates in logs
• Test authentication flows
• Verify static assets loading
• Check performance metrics

Rollback Procedure

1. Go to Railway dashboard
2. Navigate to deployments tab
3. Click "Redeploy" on previous working deployment
4. Monitor rollback completion
5. Verify application health

Advanced Patterns

Zero-Downtime Deployments

Railway handles this automatically with:

• Health check monitoring
• Gradual traffic shifting
• Automatic rollback on health check failures

Custom Domains

# Add custom domain in Railway dashboard
# Configure DNS:
CNAME record: your-domain.com -> your-app.up.railway.app

Private Networking

Use Railway's private networking for service-to-service communication:

# Faster and more secure than public URLs
INTERNAL_API_URL=http://${{backend.RAILWAY_PRIVATE_DOMAIN}}

Environment-Specific Configuration

# Use Railway environments (production, staging)
# Configure different variables per environment
ENVIRONMENT=${{RAILWAY_ENVIRONMENT}}

Common Integration Patterns

PostgreSQL

from sqlalchemy import create_engine
import os

DATABASE_URL = os.environ.get('DATABASE_URL')
if DATABASE_URL and DATABASE_URL.startswith('postgres://'):
    DATABASE_URL = DATABASE_URL.replace('postgres://', 'postgresql://')

engine = create_engine(DATABASE_URL)

Redis

import os
import redis

REDIS_URL = os.environ.get('REDIS_URL')
redis_client = redis.from_url(REDIS_URL)

File Storage (Railway Volumes)

# railway.toml
[deploy]
volumes = [
  { name = "data", mountPath = "/app/data" }
]

Monitoring and Observability

Logging Best Practices

import logging
import sys

logging.basicConfig(
    stream=sys.stdout,
    level=logging.INFO,
    format='{"time": "%(asctime)s", "level": "%(levelname)s", "message": "%(message)s"}'
)

Metrics Collection

from prometheus_client import Counter, Histogram, generate_latest

request_count = Counter('http_requests_total', 'Total HTTP requests')
request_duration = Histogram('http_request_duration_seconds', 'HTTP request duration')

@app.get("/metrics")
async def metrics():
    return Response(generate_latest(), media_type="text/plain")

Railway CLI Commands

# Login and setup
railway login
railway link

# Deployment
railway up  # Deploy current directory
railway up --detach  # Deploy without streaming logs

# Environment management
railway variables set KEY=value
railway variables delete KEY

# Logs and debugging
railway logs
railway logs --deployment <id>
railway shell  # Open shell in deployment

# Service management
railway service  # List services
railway domain  # Manage domains

Security Considerations

1. Never commit secrets - Use Railway secrets
2. Use HTTPS only - Railway provides automatic SSL
3. Configure CORS properly - Restrict to known domains
4. Validate environment variables - Check all required vars on startup
5. Use private networking - For service-to-service communication
6. Enable Railway's Web Application Firewall - If available
7. Rotate secrets regularly - Update JWT keys, API keys, etc.

Performance Optimization

Build Time Optimization

[phases.install]
# Use caching for faster rebuilds
cmds = [
  'pip install --cache-dir /root/.cache/pip -r requirements.txt'
]

Runtime Optimization

# Use production-grade servers
# Gunicorn with Uvicorn workers for async Python
import multiprocessing

workers = multiprocessing.cpu_count() * 2 + 1
worker_class = 'uvicorn.workers.UvicornWorker'

Database Connection Pooling

from sqlalchemy import create_engine

engine = create_engine(
    DATABASE_URL,
    pool_size=10,
    max_overflow=20,
    pool_pre_ping=True  # Verify connections before use
)

Skill Output Format

When using this skill, provide:

1. nixpacks.toml configuration - Customized for the project
2. Environment variable list - All required variables with Railway references
3. Deployment command - Exact commands to run
4. Health check implementation - Code for monitoring
5. Troubleshooting steps - For any potential issues
6. Rollback procedure - How to revert if needed

References

See the references/ directory for detailed documentation on:

• nixpacks configuration patterns
• Environment variable management
• Troubleshooting guides
• Production deployment checklists

Converted and distributed by TomeVault — claim your Tome and manage your conversions.