A job scheduler utility for kue, backed by redis and built for node.js with support of schedules restore on system restart.
Scheduling API is heavily inspired and borrowed from agenda and others.
Note!: expiry key notification are now enabled by default, if provided kue options has a permission to do so unless explicit disabled by passing skipConfig
option when creating kue
instance
Note!: kue-scheduler v0.6.0 is a refactored version of previous kue-scheduler to allow redis data structure and schedule queue best practice. API is the same but some of internal working may not work as previous ones
-
Redis 2.8.0 or higher.
-
If
kue-scheduler
failed to enable keyspace notification(s) automatic, then you have to enable them usingredis-cli
$ redis-cli config set notify-keyspace-events Ex
$ npm install --save kue kue-scheduler
Use this if you want to maintain different(multiple) job instances on every run
Example schedule a job to run every two seconds from now
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('every', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run every 2 seconds
Queue.every('2 seconds', job);
//somewhere process your scheduled jobs
Queue.process('every', function(job, done) {
...
done();
});
Use this if you want to maintain a single job instance on every run.
Example schedule a job to run every two seconds from now
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('unique_every', data)
.attempts(3)
.backoff(backoff)
.priority('normal')
.unique('unique_every');
//schedule it to run every 2 seconds
Queue.every('2 seconds', job);
//somewhere process your scheduled jobs
Queue.process('unique_every', function(job, done) {
...
done();
});
Example schedule a job to run every 10 seconds from now
see http://momentjs.com/timezone/ for timezones names.
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('unique_every', { timezone: 'Europe/Amsterdam' })
.attempts(3)
.backoff(backoff)
.priority('normal')
.unique('unique_every');
//schedule it to run every 10 seconds
Queue.every('*/10 * * * * *', job);
//somewhere process your scheduled jobs
Queue.process('unique_every', function(job, done) {
...
done();
});
Use this if you want to maintain different(multiple) job instances on every run
Example schedule a job to run only once two seconds from now
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('schedule', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run once 2 seconds from now
Queue.schedule('2 seconds from now', job);
//somewhere process your scheduled jobs
Queue.process('schedule', function(job, done) {
...
done();
});
Use this if you want to maintain a single job instance on every run.
Example schedule a job to run only once two seconds from now
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('unique_schedule', data)
.attempts(3)
.backoff(backoff)
.priority('normal')
.unique('unique_schedule');
//schedule it to run once 2 seconds from now
Queue.schedule('2 seconds from now', job);
//somewhere process your scheduled jobs
Queue.process('unique_schedule', function(job, done) {
...
done();
});
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run now
Queue.now(job);
//somewhere process your scheduled jobs
Queue.process('now', function(job, done) {
...
done();
});
kue-scheduler
support all kue
options with addition of the following
restore:boolen
- tellskue-scheduler
to try to restore schedules in case of restarts or other causes. By default its not enable. When enable userestore error
andrestore success
queue events to communicate with the scheduler.
Example
var options = {
prefix: 'w',
skipConfig: false,
redis: {
port: 6379,
host: '127.0.0.1',
db: 2
},
restore: true
};
...
Queue = kue.createQueue(options);
worker:boolen
- tellskue-scheduler
to listen and process job. Default totrue
. If set tofalse
you need anotherkue-scheduler
instance to process the scheduled jobs from other process
Example
//create scheduler instance(process)
var scheduler = kue.createQueue({
restore:true,
worker:false
});
//in separate process create an instance that will process works
var worker = kue.createQueue({
restore:true,
worker:true
});
worker.process(...);
...
skipConfig:boolen
- tellskue-scheduler
to skip enabling enabling key expiry notification. Example
//create scheduler instance(process) with skipConfig
var scheduler = kue.createQueue({
skipConfig:true
});
Note! if you experience the following error: ReplyError: ERR unknown command 'config'
, which will happen if you're using a redis instance with the config command disabled (AWS Elasticache for example) you must call createQueue with the skipConfig
option and manually ensure that the notify-keyspace-events
configuration key is set to Ex
.
Clear all kue
and kue-scheduler
redis data. Clean up if performed atomically with fail all or success all guarantee.
Example
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//perform cleanup
Queue.clear(function(error,response){
...
});
Enable redis key expiry notifications
.
Example
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//enable expiry key notifications
Queue.enableExpiryNotifications();
Runs a given job instance
every after a given interval
. If unique key
is provided only single instance job will exists otherwise on every run new job instance will be used.
interval
can either be a human-interval String
format or a cron String
format.
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('every', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run every 2 seconds
Queue.every('2 seconds', job);
//somewhere process your scheduled jobs
Queue.process('every', function(job, done) {
...
done();
});
Schedules a given job instance
to run once at a given time. when
can either be a Date instance
or a date.js String
such as tomorrow at 5pm
. If unique key
is provided only single instance job will exists otherwise on every run new job istance will be used.
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('schedule', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run once 2 seconds from now
Queue.schedule('2 seconds from now', job);
//somewhere process your scheduled jobs
Queue.process('schedule', function(job, done) {
...
done();
});
Schedules a given job instance
to run once immediately.
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//create a job instance
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
//schedule it to run now
Queue.now(job);
//somewhere process your scheduled jobs
Queue.process('now', function(job, done) {
...
done();
});
Remove either scheduled job with its expiry key and schedule data or non-scheduled job. A criteria may contain jobExpiryKey
, jobDataKey
or unique identifier
of the job in case of unique jobs
//using instance
Queue.remove(`<jobInstance>`, function(error, response) {
...
});
//using id
Queue.remove(`<jobId>`, function(error, response) {
...
});
//using criteria
Queue.remove({
unique: 'every_mail'
}, function(error, response) {
...
});
//using instance
Queue.remove(`<jobInstance>`, function(error, response) {
...
});
//using id
Queue.remove(`<jobId>`, function(error, response) {
...
});
//using criteria
Queue.remove({
unique: 'every_mail'
}, function(error, response) {
...
});
Enforce kue-scheduler
to restore schedules in cases that may cause your application to miss redis key expiry events. You may enable restore
using options but in some scenario you may invoke restore
by yourself.
Example
var kue = require('kue-scheduler');
var Queue = kue.createQueue();
//perform cleanup
Queue.restore(function(error, schedules){
...
});
Currently the only way to interact with kue-scheduler
is through its events. kue-scheduler
fires schedule error
, schedule success
, already scheduled
,lock error
, unlock error
, restore success
, restore error
and scheduler unknown job expiry key
events.
Use it to interact with kue-scheduler
to get notified when an error occur.
//listen on scheduler errors
Queue.on('schedule error', function(error) {
...
});
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
Queue.now(job);
Use it to interact with kue-scheduler
to obtained instance of current scheduled job.
Note: Use this event to attach instance level job events
//listen on success scheduling
Queue.on('schedule success', function(job) {
...
});
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
Queue.now(job);
Use it to interact with kue-scheduler
to be notified if the current instance of job is unique and already schedule to run.
Note: Use this event to attach instance level job events
//listen on already scheduled jobs
Queue.on('already scheduled', function(job) {
...
});
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
Queue.now(job);
Use it to interact with kue-scheduler
to get notified when a lock error occured.
//listen on scheduler errors
Queue.on('lock error', function(error) {
...
});
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
Queue.now(job);
Use it to interact with kue-scheduler
to get notified when unlock error occured.
//listen on scheduler errors
Queue.on('unlock error', function(error) {
...
});
var job = Queue
.createJob('now', data)
.attempts(3)
.backoff(backoff)
.priority('normal');
Queue.now(job);
Fired when kue-scheduler
receive unknown key event from redis. Use it to be notified on unknown key(s) events.
Queue
.on('scheduler unknown job expiry key', function(message) {
expect(Queue._isJobExpiryKey(message)).to.be.false;
});
Fired when kue-scheduler
successfully restore previous schedules.
Queue
.on('restore success', function() {
...
});
Fired when kue-scheduler
failed to restore previous schedules.
Queue
.on('restore error', function(error) {
...
});
##TODO
- should be able to extend existing unique jobs with new job data
- Fix clean to be able to clear every data including
q:<type>:jobs
which is saved during kue shutdown - Add schedule details on kue dashboard
-
Clone this repository
-
Install all development dependencies
$ npm install
- Then run test
$ npm test
It will be nice, if you open an issue first so that we can know what is going on, then, fork this repo and push in your ideas. Do not forget to add a bit of test(s) of what value you adding.
(The MIT License)
Copyright (c) 2011 lykmapipo && Contributors
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.