Structured logging with Pino for Node.js

Why structured logging?

Flat logs (console.log) are hard to analyze in production. Structured logging generates JSON that can be parsed, filtered, and analyzed by tools like Elasticsearch, Datadog, or Grafana.

Pino

Pino is the fastest logger for Node.js. Designed for production, it generates JSON directly with minimal performance impact.

import pino from 'pino';

const logger = pino({
  level: process.env.LOG_LEVEL || 'info',
});

logger.info({ userId: 123 }, 'User logged in');
// {"level":30,"time":...,"msg":"User logged in","userId":123}

Log levels

• fatal: system cannot continue.
• error: requires immediate attention.
• warn: unexpected but not critical.
• info: general system information.
• debug: detailed debugging info.
• trace: very detailed traces.

Context and bindings

const childLogger = logger.child({ requestId: 'abc-123' });
childLogger.info('Processing payment');

Best practices

• Never log sensitive data (passwords, tokens).
• Use child loggers for request context.
• Configure level per environment.
• No synchronous logging in production.

Want structured logging in your app? At Vynta we configure Pino for production.