Skip to content

xpepermint/mongodb-cron

Repository files navigation

Build Status NPM Version

mongodb-cron

MongoDB collection as crontab

This package offers a simple API for scheduling tasks and running recurring jobs on MongoDB collections. Any collection can be converted into a job queue or crontab list. It uses the officially supported Node.js driver for MongoDB. It's fast, minimizes processing overhead and it uses atomic commands to ensure safe job executions even in cluster environments.

This is a light weight open source package for NodeJS written with TypeScript. It's actively maintained, well tested and already used in production environments. The source code is available on GitHub where you can also find our issue tracker.

Installation

This is a module for Node.js and can be installed via npm. It depends on the mongodb package and uses promises.

$ npm install --save mongodb mongodb-cron

Example

Below, is a simple example to show the benefit of using this package in your Node.js projects.

Let's start by initializing the database connection.

import { MongoClient } from 'mongodb';

const mongo = await MongoClient.connect('mongodb://localhost:27017');
const db = mongo.db('test');

Continue by initializing and starting a the worker.

import { MongoCron } from 'mongodb-cron';

const collection = db.collection('jobs');
const cron = new MongoCron({
  collection, // a collection where jobs are stored
  onDocument: async (doc) => console.log(doc), // triggered on job processing
  onError: async (err) => console.log(err), // triggered on error
});

cron.start(); // start processing

We can now create our first job.

const job = await collection.insert({
  sleepUntil: new Date('2016-01-01'), // ISO 8601 format (can include timezone)
});

When the processing starts the onDocument handler (defined earlier) is triggered. We have a very basic example here so please continue reading.

Documentation

The MongoCron class converts a collection into a job queue. Jobs are represented by the documents stored in a MongoDB collection. When cron is started it loops through the collection and processes available jobs one by one.

A job should have at least the sleepUntil field. Cron processes only documents where this field exists, other documents are ignored.

One-time Jobs

To create a one-time job we only need to define the required field sleepUntil. When this filed is set to some date in the past, the processing starts immediately.

const job = await collection.insert({
  sleepUntil: new Date(),
});

When the processing of a document starts the sleepUntil field is updated to a new date in the future. This locks the document for a certain amount of time in which the processing must complete (lock duration is configurable). This mechanism prevents possible race conditions and ensures that a job is always processed by only one process at a time.

When the processing ends, the sleepUntil field is set to null.

If cron is unexpectedly interrupted during the processing of a job (e.g. server shutdown), the system automatically recovers and transparently restarts.

Deferred Execution

We can schedule job execution for a particular time in the future by setting the sleepUntil field to a future date.

const job = await collection.insert({
  ...
  sleepUntil: new Date('2016-01-01'), // start on 2016-01-01
});

Recurring Jobs

By setting the interval field we define a recurring job.

const job = await collection.insert({
  ...
  interval: '* * * * * *', // every second
});

The interval above consists of 6 values.

* * * * * *
┬ ┬ ┬ ┬ ┬ ┬
│ │ │ │ │ └── day of week (0 - 7) (0 or 7 is Sun)
│ │ │ │ └──── month (1 - 12)
│ │ │ └────── day of month (1 - 31)
│ │ └──────── hour (0 - 23)
│ └────────── minute (0 - 59)
└──────────── second (0 - 59)

A recurring job will repeat endlessly unless we limit that by setting the repeatUntil field. When a job expires it stops repeating by removing the processable field.

const job = await collection.insert({
  ...
  interval: '* * * * * *',
  repeatUntil: new Date('2020-01-01'),
});

Auto-removable Jobs

A job can automatically remove itself from the collection when the processing completes. To configure that, we need to set the autoRemove field to true.

const job = await collection.insert({
  ...
  autoRemove: true,
});

API

new MongoCron({ collection, condition, onStart, onStop, onDocument, onError, nextDelay, reprocessDelay, idleDelay, lockDuration, sleepUntilFieldPath, intervalFieldPath, repeatUntilFieldPath, autoRemoveFieldPath })

The core class for converting a MongoDB collection into a job queue.

Option Type Required Default Description
autoRemoveFieldPath String No autoRemove The autoRemove field path.
collection Object Yes - MongoDB collection object.
condition Object No null Additional query condition.
idleDelay Integer No 0 A variable which tells how many milliseconds the worker should wait before checking for new jobs after all jobs has been processed.
intervalFieldPath String No interval The interval field path.
lockDuration Integer No 600000 A number of milliseconds for which each job gets locked for (we have to make sure that the job completes in that time frame).
nextDelay Integer No 0 A variable which tells how fast the next job can be processed.
onDocument Function/Promise No - A method which is triggered when a document should be processed.
onError Function/Promise No - A method which is triggered in case of an error.
onIdle Function/Promise No - A method which is triggered when all jobs in a collection have been processed.
onStart Function/Promise No - A method which is triggered when the cron is started.
onStop Function/Promise No - A method which is triggered when the cron is stopped.
repeatUntilFieldPath String No repeatUntil The repeatUntil field path.
reprocessDelay Integer No 0 A variable which tells how many milliseconds the worker should wait before processing the same job again in case the job is a recurring job.
sleepUntilFieldPath String No sleepUntil The sleepUntil field path.
import { MongoClient } from 'mongodb';

const mongo = await MongoClient.connect('mongodb://localhost:27017/test');

const cron = new MongoCron({
  collection: db.collection('jobs'),
  onStart: async () => {},
  onStop: async () => {},
  onDocument: async (doc) => {},
  onIdle: async (doc) => {},
  onError: async (err) => {},
  nextDelay: 1000,
  reprocessDelay: 1000,
  idleDelay: 10000,
  lockDuration: 600000,
  sleepUntilFieldPath: 'cron.sleepUntil',
  intervalFieldPath: 'cron.interval',
  repeatUntilFieldPath: 'cron.repeatUntil',
  autoRemoveFieldPath: 'cron.autoRemove',
});

cron.start():Promise

Starts the cron processor.

cron.stop():Promise

Stops the cron processor.

cron.isRunning():Boolean

Returns true if the cron is started.

cron.isProcessing():Boolean

Returns true if cron is processing a document.

cron.isIdle():Boolean

Returns true if the cron is in idle state.

Processing Speed

Processing speed can be reduced when more and more documents are added into the collection. We can maintain the speed by creating indexes.

await collection.createIndex({
  sleepUntil: 1, // the `sleepUntil` field path, set by the sleepUntilFieldPath
}, {
  sparse: true,
});

Don't forget to adjust the index definition when using your custom query condition.

Best Practice

  • Make your jobs idempotent and transactional. Idempotency means that your job can safely execute multiple times.
  • Run this package in cluster mode. Design your jobs in a way that you can run lots of them in parallel.

Licence

Copyright (c) 2016 Kristijan Sedlak <xpepermint@gmail.com>

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.