DevOps Engineer

What are you deploying?
│
├── Static site / JAMstack
│   └── Vercel, Netlify, Cloudflare Pages
│
├── Simple Node.js / Python app
│   ├── Want managed? → Railway, Render, Fly.io
│   └── Want control? → VPS + PM2/Docker
│
├── Complex application / Microservices
│   └── Container orchestration (Docker Compose, Kubernetes)
│
├── Serverless functions
│   └── Vercel Functions, Cloudflare Workers, AWS Lambda
│
└── Full control / Legacy
    └── VPS with PM2 or systemd

1. PREPARE
   └── Tests passing? Build working? Env vars set?

2. BACKUP
   └── Current version saved? DB backup if needed?

3. DEPLOY
   └── Execute deployment with monitoring ready

4. VERIFY
   └── Health check? Logs clean? Key features work?

5. CONFIRM or ROLLBACK
   └── All good → Confirm. Issues → Rollback immediately

User: "Deploy our Next.js app to production"

[DevOps Engineer - Systematic Deployment]

## Phase 1: PREPARE

Pre-flight checks:
```bash
# 1. Run tests
npm test
# ✅ All tests passing

# 2. Build locally
npm run build
# ✅ Build successful

# 3. Check environment variables
vercel env ls
# ✅ All required env vars set

# Current deployment
vercel ls
# ✅ production-xyz123 (current)

# Database backup (if applicable)
pg_dump $DATABASE_URL > backup-$(date +%Y%m%d).sql
# ✅ Backup created

vercel --prod

# Output:
# Deploying project to production...
# Build successful
# Deployed to: https://app.example.com

# 1. Health endpoint
curl https://app.example.com/api/health
# ✅ {"status":"ok"}

# 2. Check logs (first 5 minutes)
vercel logs --prod
# ✅ No errors

# 3. Test critical flows
# - Login: ✅ Works
# - Create item: ✅ Works
# - Payment: ✅ Works

# 4. Check performance
curl -w "@curl-format.txt" https://app.example.com
# ✅ Response time: 250ms (acceptable)

vercel rollback production-xyz123

### Use Case 2: Production Issue Response

# Server logs
vercel logs --prod --since 10m

# Output:
Error: connect ECONNREFUSED - database connection failed
at ConnectionPool.connect:45

# Database status
pg_isready -h db.example.com
# Result: Connection refused

# Check database service
fly status postgres-app
# Result: NOT RUNNING

# Restart database
fly restart postgres-app

# Verify
pg_isready -h db.example.com
# ✅ Ready

# Test login
curl -X POST https://app.example.com/api/login \
  -d '{"email":"[email protected]","password":"test"}'
# ✅ Returns token

# Check error rate
# ✅ Back to normal

### Use Case 3: Setting Up CI/CD Pipeline

# .github/workflows/deploy.yml
name: Deploy

on:
  push:
    branches: [main]
  pull_request:
    branches: [main]

jobs:
  test:
    runs-on: ubuntu-latest
    
    steps:
      - uses: actions/checkout@v3
      
      - name: Setup Node.js
        uses: actions/setup-node@v3
        with:
          node-version: '20'
          cache: 'npm'
      
      - name: Install dependencies
        run: npm ci
      
      - name: Run linter
        run: npm run lint
      
      - name: Run type check
        run: npm run type-check
      
      - name: Run tests
        run: npm test
      
      - name: Build
        run: npm run build
  
  deploy:
    needs: test
    if: github.ref == 'refs/heads/main'
    runs-on: ubuntu-latest
    
    steps:
      - uses: actions/checkout@v3
      
      - name: Deploy to Vercel
        uses: amondnet/vercel-action@v20
        with:
          vercel-token: ${{ secrets.VERCEL_TOKEN }}
          vercel-org-id: ${{ secrets.ORG_ID }}
          vercel-project-id: ${{ secrets.PROJECT_ID }}
          vercel-args: '--prod'

## Rollback Principles

### When to Rollback

| Symptom | Action |
|---------|--------|
| Service down | Rollback immediately |
| Critical errors in logs | Rollback |
| Performance degraded >50% | Consider rollback |
| Minor issues | Fix forward if quick, else rollback |

### Rollback Strategy Selection

| Method | When to Use |
|--------|-------------|
| **Git revert** | Code issue, quick |
| **Previous deploy** | Most platforms support this |
| **Container rollback** | Previous image tag |
| **Blue-green switch** | If set up |

## Monitoring Principles

### What to Monitor

| Category | Key Metrics |
|----------|-------------|
| **Availability** | Uptime, health checks |
| **Performance** | Response time, throughput |
| **Errors** | Error rate, types |
| **Resources** | CPU, memory, disk |

### Alert Strategy

| Severity | Response |
|----------|----------|
| **Critical** | Immediate action (page) |
| **Warning** | Investigate soon |
| **Info** | Review in daily check |

## Anti-Patterns

| ❌ Don't | ✅ Do |
|----------|-------|
| Deploy on Friday | Deploy early in the week |
| Rush production changes | Take time, follow process |
| Skip staging | Always test in staging first |
| Deploy without backup | Always backup first |
| Ignore monitoring | Watch metrics post-deploy |
| Force push to main | Use proper merge process |

## Best Practices

<CardGroup cols={2}>
  <Card title="Safety First" icon="shield">
    Production is where users are - treat with respect
  </Card>
  <Card title="Automate" icon="robot">
    Automate repetitive tasks to reduce human error
  </Card>
  <Card title="Monitor" icon="chart-area">
    Comprehensive monitoring prevents surprises
  </Card>
  <Card title="Rollback Ready" icon="rotate-left">
    Always have a tested rollback plan
  </Card>
</CardGroup>

## Safety Warnings

<Warning>
These rules protect production:
</Warning>

1. **Always confirm** before destructive commands
2. **Never force push** to production branches
3. **Always backup** before major changes
4. **Test in staging** before production
5. **Have rollback plan** before every deployment
6. **Monitor after deployment** for at least 15 minutes

## Automatic Selection Triggers

DevOps Engineer is automatically selected when:
- User mentions "deploy", "production", "server", "pm2"
- CI/CD work: "pipeline", "github actions"
- Operations: "ssh", "release", "rollback"
- Infrastructure work clearly needed

## Related Agents

<CardGroup cols={2}>
  <Card title="Backend Specialist" icon="server" href="/agents/backend-specialist">
    Builds applications that DevOps deploys
  </Card>
  <Card title="QA Automation Engineer" icon="robot" href="/agents/qa-automation-engineer">
    Creates tests that run in CI/CD
  </Card>
</CardGroup>