Scheduler

For handling background processes in your application use Q3 Scheduler. We recommended running the scheduler on an independent server since it uses MongoDB as a persistence layer. This way, your web requests and your jobs don't need to compete for resources.

Scheduler.queue

To add one-time jobs, call the queue method. It takes two arguments: the name and the payload (optional). The payload can contain any data, though it will process the buckets and session properties differently.

If the payload contains buckets, the receiving function will have file attachments in its second parameter position. Likewise, if the payload contains session, it will hydrate the q3-core-session module before executing the file.

const Scheduler = require('q3-core-scheduler');

Scheduler.queue('file', {
  buckets: [],
  session: {},
  id: 1,
}).then(() => {
  // will eventually call the file below at chores/file.js
});

module.exports = (data, attachments) => {
  console.log('Regular data:', data);
  console.log('File buffers:', attachments);
};

Scheduler.seed

When the scheduler receives a target directory, it will walk the files in it looking for recurring jobs. Recurring jobs have files ending in @${interval}.js. Acceptable interval values include annually, biannually, quarterly, monthly, weekly, daily, bihourly, hourly, semihourly, biminutely and minutely.

const Scheduler = require('q3-core-scheduler');

// tells the scheduler to look for jobs in this directory
Scheduler.seed(__dirname).then(() => {
  console.log('Recurring jobs saved');
});

Scheduler.start

Given an executable directory and an interval, the queue manager can run. It will work by priority sequence (high to low) and lock any jobs in progress, processing just one at a time. After execution, the queue manager will stamp a completition date (iso) and duration (ms) for benchmarking and debugging purposes.

const Scheduler = require('q3-core-scheduler');

// tells the scheduler to execute files in this directory
// every 10 milliseconds
Scheduler.start(__dirname, 10).then(() => {
  console.log('Queue manager started');
});

Events

Scheduler implements a custom EventEmitter and exposes its on listener publicly. During the queueing process, each job will emit two of three events.

1. Scheduler.on("start"): Every job emits this event before it executes;
2. Scheduler.on("finish"): Successful jobs emit this event before the queue starts to look for the next job;
3. Scheduler.on("stall"): Failed jobs emit this when an error occurs during execution.

Alternatively, when a new job gets added, the emitter will fire the queued event.