@cuckoointernet/logger-node v1.0.0

logger-node

A simple and fast JSON logging library for Node.js services

Usage

import { Logger } from "@cuckoointernet/logger-node";

const logger = new Logger("my-package");

Log simple messages at different levels

logger.debug("Hello Cuckoo!");
logger.info("Hello Cuckoo!");
logger.warn("Hello Cuckoo!");
logger.error("Hello Cuckoo!");
logger.fatal("Hello Cuckoo!");

Log additional data via the second argument

// Data in object supplied is automatically merged into the log record
logger.info("Hello Cuckoo!", { colour: "yellow" });

Serialisers

Data provided to the second argument undergoes additional processing if they match certain keys. For instance, if you pass an object with an error key it will be run through a serialiser that is able to process stack information in a better way. The standard serialisers are:

Log JavaScript Error objects

// Alternatively you can supply an instance of Error to log its exception details via the second argument
logger.warn("Sad Cuckoo...", new Error("Wings were clipped!"));

// To log an Error *and* other data at the same time, use the 'error' field name
logger.error("Sad Cuckoo...", {
  error: new Error("Wings were clipped!"),
  colour: "yellow",
});

Return a JavaScript Error after logging

When logging at levels error and fatal you can return a JavaScript Error that has the same message as the log record and then throw:

// The message of the error thrown will be "Mission failed"
throw logger.error("Mission failed").returnError();

// You can log additional data via the second argument as per usual
throw logger
  .fatal("Mission failed", { reason: "Ran out of fuel..." })
  .returnError();

Child Loggers

A child logger can be created from an existing one to specialize a logger for a sub-component of your application, i.e. to create a new logger with additional bound fields that will be included in its log records.

const parentLogger = new Logger("parent", "debug");

// The child logger inherits the same name and log level as the parent
const childLogger = parentLogger.createChildLogger({
  subPackage: "child",
  anotherChildField: "whatever-you-want",
});

// All log records will contain the two additional fields setup at initialisation, ie: subPackage & anotherChildField
childLogger.info("Hello from child");

Log Levels

Setting a logger instance to a particular level results in only log records of that level and above being written. You can configure it via the options below:

1. If not specified the logger defaults to info level:

const logger = new Logger("my-package");

2. Set via logLevel constructor parameter:

const logger = new Logger("my-package", "debug");

3. Set via LOG_LEVEL environment variable:

// process.env.LOG_LEVEL = "debug"
const logger = new Logger("my-package");

The available log levels and best practices guidance on when to use them are as follows:

• fatal (60): The service/app is going to stop or become unusable now
• error (50): Fatal for a particular request, but the service/app continues servicing other requests
• warn (40): A note on something that should probably be looked at
• info (30): Detail on regular operation
• debug (20): Anything else, i.e. too verbose to be included in "info" level

If you want to prevent the logger from printing any messages you can set the log level to silent. This is sometimes useful, for example when running tests to reduce noise in the terminal.

Log Records

The structure of log records is outlined below:

{
  // User supplied data
  name: "my-package",
  msg: "Hello Cuckoo!",

  ...any additional data supplied via second argument to logger methods (see examples above)

  // Record metadata (added automatically)
  logLevel: "info",
  level: 30,
  time: "2022-02-03T19:02:57.534Z",
  hostname: "banana.local",
  pid: 123,

  // AWS metadata (added automatically if applicable)
  @requestId: <id>
}