Cloud & DevOps

const config = {
  port: process.env.PORT || 3000,
  databaseUrl: process.env.DATABASE_URL,
};

console.log(JSON.stringify({ level: 'info', message: 'Server started' }));

Common follow-ups: Why does storing state on a specific instance's local disk violate the stateless principle, and what's the cloud-native alternative?;How does this methodology specifically enable horizontal auto-scaling?

app.get('/healthz/live', (req, res) => res.status(200).send('OK'));

app.get('/healthz/ready', async (req, res) => {
  const dbOk = await checkDatabaseConnection();
  res.status(dbOk ? 200 : 503).send(dbOk ? 'Ready' : 'Not ready');
});

Common follow-ups: What's the risk of making the liveness probe check external dependencies rather than just the process's own health?;How do you tune failureThreshold and periodSeconds to avoid both false positives and slow detection?

name: CI
on: [push]
jobs:
  test:
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
      - run: npm ci
      - run: npm run lint
      - run: npm test
      - run: docker build -t myapp:${{ github.sha }} .

Common follow-ups: What's the difference between continuous delivery and continuous deployment, given both are abbreviated CD?;How would you add a rollback mechanism in case a deployed change causes issues?

const fn = new lambda.Function(this, 'ApiHandler', {
  runtime: lambda.Runtime.NODEJS_20_X,
  handler: 'index.handler',
  code: lambda.Code.fromAsset('dist'),
});
new apigateway.LambdaRestApi(this, 'Api', { handler: fn });

Common follow-ups: What's the tradeoff of using the CDK versus a purely declarative tool like Terraform?;How do you handle infrastructure state drift when using IaC alongside occasional manual console changes?

// Stateful (breaks horizontal scaling)
app.use(session({ secret: 'x' }));

// Stateless: session in shared Redis
app.use(session({ store: new RedisStore({ client: redisClient }), secret: process.env.SESSION_SECRET }));

Common follow-ups: How does sticky sessions at the load balancer offer a partial alternative to making the app fully stateless?;What's the difference between horizontal scaling and Node's built-in cluster module?

upstream node_app {
  server 10.0.0.1:3000;
  server 10.0.0.2:3000;
}
server {
  listen 80;
  location / { proxy_pass http://node_app; }
}

Common follow-ups: How does the load balancer itself avoid becoming a single point of failure?;What's the difference between Layer 4 and Layer 7 load balancing, and when does it matter for a Node.js app?

aws elbv2 modify-listener \
  --listener-arn $LISTENER_ARN \
  --default-actions Type=forward,TargetGroupArn=$GREEN_TARGET_GROUP_ARN

Common follow-ups: How does blue-green deployment compare to canary release in terms of risk exposure?;What's the cost implication of running two full production environments simultaneously?

const client = new SecretsManagerClient({ region: 'us-east-1' });
const secret = await client.send(new GetSecretValueCommand({ SecretId: 'prod/database' }));
const { password } = JSON.parse(secret.SecretString);

Common follow-ups: What's the specific security risk of environment variables versus a dedicated secrets manager?;How do you handle secret rotation without requiring a full redeploy?

{
  "TargetValue": 60.0,
  "PredefinedMetricSpecification": { "PredefinedMetricType": "ECSServiceAverageCPUUtilization" },
  "ScaleOutCooldown": 60,
  "ScaleInCooldown": 300
}

Common follow-ups: Why is scale-in cooldown typically set longer than scale-out cooldown?;How does auto-scaling a stateless HTTP API differ from auto-scaling a background-job worker fleet?

server {
  listen 443 ssl;
  location /static/ { root /var/www; }
  location / { proxy_pass http://localhost:3000; }
}

Common follow-ups: How does the Node.js app read the real client IP when requests are proxied through Nginx?;What's the tradeoff of terminating TLS at Nginx versus in the Node.js process directly?