This is a support package for the winston library that adds meta data for the logs and structures them into json. The package is tested with winston 2.x version.
npm install https://github.com/browserstack/chitragupta-node
The informative structured logging can be achieved in 2 steps:
- Setting up the logger
- Setting up the contexts in server/process/worker
const winston = require('winston');
const { Chitragupta } = require('chitragupta');
// Configure your logger's formatter
var logger = new winston.Logger({
level: 'info',
transports: [
new (winston.transports.Console)({
formatter: Chitragupta.jsonLogFormatter
})]
});
const winston = require('winston');
const { Chitragupta } = require('chitragupta');
const jsonLogFormatter = winston.format.printf( ({ level, message, ...meta}) => {
const options = {level: level, message: message, meta: meta};
return Chitragupta.jsonLogFormatter(options);
});
const logger = winston.createLogger({
transports: [
new winston.transports.Console({
format: jsonLogFormatter,
})
],
});
For web servers built with node http package
const http = require("http");
// Initialize and add meta data for the server
function requestHandler(request, response) {
logger.log('info', 'processing request');
response.end();
}
var server = http.createServer(function(request, response) {
// Chitragupta.setupServerLogger(loggerObject, request, response, userId, requestHandler);
Chitragupta.setupServerLogger(logger, request, response, 123, requestHandler);
});
server.listen(8080);
OR
function requestHandler(request, response) {
logger.log('info', 'processing request', {'data': {'request': {'method': 'POST'}}});
response.end();
}
For processes/services
function process_some_crazy_stuff(a, b, c) {
logger.log('info', a);
logger.log('info', b);
logger.log('info', c);
}
// Chitragupta.setupProcessLogger(uniqueNameOfTheProcess, functionToBeCalled, all, the, args, that, you, would, like, to, pass);
Chitragupta.setupProcessLogger('processing_crazy_stuff', process_some_crazy_stuff, 1, true, 45);
OR
function process_some_crazy_stuff(a, b, c) {
logger.log('info', a, {'data': {'name': 'process_name', 'execution_id': 'some_unique_id'}});
logger.log('info', b, {'data': {'name': 'some_other_process_name', 'execution_id': 'some_other_unique_id'}});
}
For workers
const processFromQueue = (n) => {
getCurrentQueue()
.then(someAwesomeStuff)
.catch(e => {
logger.log('info', 'Exceptions');
setTimeout(getWorkFromRedis, 1000);
});
}
const getWorkFromRedis = (n, first_run) => {
// Use the following function to set the worker job id if required
// Chitragupta.setWorkerJobId('unique-job-id', functionToBeCalled, all, the, args, that, you, would, like, to, pass);
Chitragupta.setWorkerJobId(jobId, processFromQueue, n);
}
// Chitragupta.setupWorkerLogger(uniqueNameOfTheWorker, functionToBeCalled, all, the, args, that, you, would, like, to, pass);
Chitragupta.setupWorkerLogger('redisWorker', getWorkFromRedis, 1, true);
OR
const processFromQueue = (n) => {
getCurrentQueue()
.then(someAwesomeStuff)
.catch(e => {
logger.log('info', 'Exceptions', {'data': {'worker_name': 'worker_name', 'thread_id': 'some_thread_id'}});
setTimeout(getWorkFromRedis, 1000);
});
}
- For the logs where
log.kind
is missing, first 40 chars of dynamic_data is populated to uniquely identify each log in a restricted fashion. - For the logs where no contexts are set, ie. server, process or worker are not set, server format category and version is populated.
request.request_id
and therequest_start_time
can be set manually if required.
Run eslint using the following command
# Please install the eslint supported node version using `nvm install v12.14.1`
npx eslint .
Bug reports and pull requests are welcome on GitHub at https://github.com/browserstack/chitragupta-node.
The package is available as open source under the terms of the MIT License.