↓ Ir para o conteúdo principal

← todas as notas

📎 Webclip

Performance Best Practices: Running and Monitoring Express.js in Production

The article says a performant Express.js application should aim for minimal downtime, predictable resource usage, effective scaling, and higher developer productivity. It recommends a basic project structure, async code, structured logging, metrics, and production tooling like dotenv, systemd, PM2, Nginx, Redis, and monitoring.

Reading notes
#

  • An intuitive file structure helps keep files small, makes reviews and refactoring easier, and reduces merge conflicts.
  • The suggested layout separates config, controllers, providers, services, models, routes, database setup, and tests.
  • Setting NODE_ENV=production is presented as a quick performance improvement.
  • Gzip compression should be enabled, either in Express or at the Nginx layer.
  • Synchronous functions should be avoided because they block the thread; Promises and async functions are preferred.
  • If a CPU-heavy task cannot be avoided, it should run in child processes or forks instead of blocking the main thread.
  • Logs should be structured and collected centrally instead of using console.log().
  • The Sematext Express agent is shown as a way to send logs and request data to a central logging system.
  • Errors should be handled with try-catch, a catch-all error middleware, and process-level listeners for uncaught exceptions and unhandled rejections.
  • Memory leaks should be watched through process and heap metrics, especially when memory grows steadily without matching request growth.
  • Pure functions are recommended because they do not mutate outer state and make code easier to reason about.
  • Function arguments are easier to manage when passed as objects, since names do not depend on parameter order.
  • Tests should be written first or early, using Mocha and Chai in the example.
  • dotenv is used to load environment variables from a .env file.
  • Systemd can run the Node.js app as a service and restart it on failure.
  • PM2 can also restart the app and run it in cluster mode.
  • The Node.js cluster module can fork workers across CPU cores and restart failed workers.
  • Nginx is recommended as a reverse proxy rather than exposing Node.js directly on ports 80 or 443.
  • Nginx can also handle load balancing, caching, and gzip compression.
  • Redis can cache API responses and reduce repeated database work.
  • Infrastructure-wide monitoring is needed because Nginx or Redis can fail even when the Express app itself is still running.
  • Alerts are important so failures are detected before users report them.