Are you logging your NodeJS code?

Logs are essential part of an application, it provides you deep level access of output comes from user inputs. When application is on production, logs are important to identify the problem if something goes wrong. If you are developer, have you asked yourself a question, “Am I logging or not?”

Why Logs are important?

No matter how careful we are while developing application, it is quite difficult to make it 100% bug free. Even after spending time to find defects in test cycle, we still won’t be able to catch all of them.

These remaining error may cause unexpected exceptions on production environment. In some case your application might crash in certain user journeys. It is always a good idea to keep an extra eye on application flow.

In order to achieve success, we use application logs to check why the application is behaving differently. For this, we have to set up our application to record information about its events and errors. This is what we call logger, it helps us to identify problems with an application running in production.

Best Practices

1. Avoid using console.log()

There are some important things we need to consider while configuring logs in application. Adding logs with console.log() won’t stay for longer period of time. They are available till the time user refresh the page.

console.log() can be use for temporary logging since it provides in memory logs which can be easily read through terminal since it uses stdout. Similarly console.error or console.warn can also be used but you can’t store these logs into any file or database.

Hence, console.log doesn’t provide us enough options to use it as primary logging system. You should consider proper library to use as logging system.

2. Third party libraries

Dedicated libraries unlike console.log provides us adequate options to define and configure logging system.

• Levels: It offers different levels of logging, example you can use several levels of logging like, info, warn, debug, and error. These helps to filter the problems.
• Appearance: You can use different colors and appearance to distinguish your logs.
• Data types: Not only appearance, you can also make your logs different in type. Some libraries support JSON format as well.

Winston and Bunyan are two of the most popular logging libraries available for Node applications.

3. Source, Timestamps, Context

Logging is not only to log text when exception or success comes, there are 3 important aspects which makes them more useful when it comes to debugging.

• Source: While debugging the application through logs it is important to know what is the source of particular log. Hence, it is important to keep, hostname, method, module name.
• Timestamps: Timestamps plays very important role while debugging an application, because it helps you to identify at what timeframe error happened. In the world of micro services it is important to keep timestamps because requests are not sequential but async.
• Context: These types of errors/exceptions comes based on users inputs. Example, when user is trying to register in an application but registration failed due to he/she is already registered user or provided email address is wrong etc. So, application behavior was expected but user was not able to register.

4. Logging levels

Use different levels of logs to distinguish them, so developer can easily understand while debugging the application.

• Emergency: the system is unusable
• Alert: action must be taken immediately
• Critical: critical conditions
• Error: error conditions
• Warning: warning conditions
• Notice: normal but significant conditions
• Informational: informational messages
• Debug: debug-level messages

You can still alter these levels upon your needs.

5. What shouldn’t be part of logs

There are some obvious mistakes developers does while configuring loggings. One shouldn’t log any PII data while logging requests or exceptions. I would like to share some bad and good examples of logging.

Bad Example

const express = require('express');
const winston = require('winston');
const app = express();

// configuring logger
const logger = winston.createLogger({
  transports: [
    new winston.transports.Console(),
    new winston.transports.File({ filename: 'combined.log' })
  ]
});

app.post('/user/add', (req, res) => {
  try {
    modal.addUser({
      email: req.email,
      password: req.pwd,
    }).then(() => {
      logger.log({
        level: 'info',
        message: `${req.email} with password ${pwd} has been successfully registered`
      });
      res.send(200);
    });
  } catch (err) {
    logger.log({
      level: 'error',
      message: `${req.email} with password ${pwd} wasn't registered`
    });
  }
});

Good Example

const express = require('express');
const winston = require('winston');
const app = express();

// configuring logger
const logger = winston.createLogger({
  transports: [
    new winston.transports.Console(),
    new winston.transports.File({ filename: 'combined.log' })
  ]
});

app.post('/user/add', (req, res) => {
  try {
    modal.addUser({
      email: req.email,
      password: req.pwd,
    }).then((response) => {
      logger.log({
        level: 'info',
        message: `Success: ${response.data.id} user has been successfully registered`
      });
      res.send(200);
    });
  } catch (err) {
    logger.log({
      level: 'error',
      message: `An exception occurred while registering new user: ${err}`
    });
  }
});

Summary

All I can say that, it is important to have proper logging system setup in application. This can make developers life easier to debug any problem on production environment with less turnaround time.