Debugging & Production Logging

📖 Concept

Effective debugging and structured logging are critical for maintaining Node.js applications. In production, logs are your primary window into application behavior.

Debugging tools:

Tool	Use Case
`console.log()`	Quick dev debugging (remove before committing!)
`debugger` statement	Breakpoints in Node.js inspector
`node --inspect`	Chrome DevTools debugger
VS Code debugger	IDE-integrated debugging
`debug` package	Conditional debug logging (e.g., Express uses it)

Production logging with Winston: Winston is the most popular logging library for Node.js. It supports multiple log levels, transports (console, file, database), and structured JSON logging.

Log levels (severity):

error: 0  →  Application errors, crashes
warn:  1  →  Unexpected behavior, deprecations
info:  2  →  Important events (server start, user actions)
http:  3  →  HTTP request/response logs
debug: 4  →  Detailed debugging (disabled in production)

Structured logging best practices:

Use JSON format — parseable by log aggregation tools (ELK, Datadog)
Include context — request ID, user ID, timestamp
Log at appropriate levels — don't log everything at info
Never log sensitive data — passwords, tokens, credit cards
Use correlation IDs — trace a request across microservices

🏠 Real-world analogy: Debugging is like being a detective. console.log is asking witnesses (prints). The Node.js debugger is reviewing security camera footage (step-by-step execution). Production logging is like a flight recorder (black box) — capturing everything needed to reconstruct events after they happen.

💻 Code Example

codeTap to expand ⛶

1// Debugging & Production Logging
2
3// === 1. Winston Logger Setup ===
4const winston = require("winston");
5
6const logger = winston.createLogger({
7  level: process.env.LOG_LEVEL || "info",
8  format: winston.format.combine(
9    winston.format.timestamp(),
10    winston.format.errors({ stack: true }),
11    winston.format.json()
12  ),
13  defaultMeta: { service: "my-api" },
14  transports: [
15    // Console (colorized for development)
16    new winston.transports.Console({
17      format:
18        process.env.NODE_ENV === "development"
19          ? winston.format.combine(winston.format.colorize(), winston.format.simple())
20          : winston.format.json(),
21    }),
22    // File transports (production)
23    new winston.transports.File({ filename: "logs/error.log", level: "error", maxsize: 5242880, maxFiles: 5 }),
24    new winston.transports.File({ filename: "logs/combined.log", maxsize: 5242880, maxFiles: 5 }),
25  ],
26});
27
28// 2. Request logging middleware
29function requestLogger(req, res, next) {
30  const requestId = req.headers["x-request-id"] || require("crypto").randomUUID();
31  req.requestId = requestId;
32  res.setHeader("X-Request-Id", requestId);
33
34  const start = Date.now();
35
36  res.on("finish", () => {
37    const duration = Date.now() - start;
38    const logData = {
39      requestId,
40      method: req.method,
41      url: req.originalUrl,
42      statusCode: res.statusCode,
43      duration: `${duration}ms`,
44      ip: req.ip,
45      userAgent: req.headers["user-agent"],
46      userId: req.user?.id,
47    };
48
49    if (res.statusCode >= 500) {
50      logger.error("Request failed", logData);
51    } else if (res.statusCode >= 400) {
52      logger.warn("Client error", logData);
53    } else {
54      logger.info("Request completed", logData);
55    }
56  });
57
58  next();
59}
60
61// 3. Usage in application code
62logger.info("Server starting", { port: 3000, env: process.env.NODE_ENV });
63logger.warn("Cache miss", { key: "user:42", source: "redis" });
64logger.error("Database query failed", { query: "SELECT...", error: "Connection refused" });
65
66// ❌ NEVER log sensitive data
67// logger.info("User login", { email, password }); // BAD!
68// ✅ Log safely
69// logger.info("User login", { email, userId: user.id }); // GOOD
70
71// 4. Node.js Debugger
72// Start with: node --inspect src/server.js
73// Open: chrome://inspect in Chrome
74// Or use VS Code's built-in debugger
75
76// .vscode/launch.json
77const vscodeLaunchConfig = {
78  version: "0.2.0",
79  configurations: [
80    {
81      type: "node",
82      request: "launch",
83      name: "Debug Server",
84      program: "${workspaceFolder}/src/server.js",
85      env: { NODE_ENV: "development" },
86      restart: true,
87    },
88    {
89      type: "node",
90      request: "launch",
91      name: "Debug Tests",
92      program: "${workspaceFolder}/node_modules/.bin/jest",
93      args: ["--runInBand", "--no-cache"],
94    },
95  ],
96};
97
98// 5. Conditional debug logging (debug package)
99// const debug = require("debug");
100// const dbDebug = debug("app:db");
101// const httpDebug = debug("app:http");
102// dbDebug("Query executed:", query); // Only shows if DEBUG=app:db
103// Start with: DEBUG=app:* node server.js
104
105// 6. Performance profiling
106function measurePerformance() {
107  console.time("operation");
108  // ... do work
109  console.timeEnd("operation"); // "operation: 123ms"
110
111  // More precise
112  const { performance, PerformanceObserver } = require("perf_hooks");
113  const start = performance.now();
114  // ... do work
115  const duration = performance.now() - start;
116  logger.debug("Operation completed", { durationMs: duration.toFixed(2) });
117}
118
119module.exports = { logger, requestLogger };

🏋️ Practice Exercise

Exercises:

Set up Winston with console, file, and error-specific transports
Create request logging middleware that includes request ID, method, URL, status, and duration
Add correlation IDs — generate a unique ID per request and include it in all logs
Configure VS Code debugger for both running the server and running tests
Profile a slow endpoint using console.time, perf_hooks, and Node.js --prof flag
Set up structured logging that can be parsed by ELK Stack or Datadog

⚠️ Common Mistakes

Using console.log in production — it's synchronous, has no log levels, and can't be routed to files or services; use Winston or Pino
Logging sensitive data — passwords, tokens, credit card numbers must NEVER appear in logs; audit your logs regularly
Not including correlation IDs — without request IDs, you can't trace a request through multiple log entries or services
Logging too much in production — excessive logging impacts performance and storage costs; use info level in production, debug in development
Not setting up log rotation — without rotation, log files grow indefinitely and fill the disk; use Winston's maxsize and maxFiles options

💼 Interview Questions

🎤 Mock Interview

Mock interview is powered by AI for Debugging & Production Logging. Login to unlock this feature.

Was this topic helpful?

← PreviousIntegration & API Testing Next →Performance Profiling & Optimization