Repository files navigation

If you're viewing this athttps://github.com/collectiveidea/delayed_job,you're reading the documentation for the master branch.View documentation for the latest release(4.1.13).

Delayed::Job (or DJ) encapsulates the common pattern of asynchronously executinglonger tasks in the background.

It is a direct extraction from Shopify where the job table is responsible for amultitude of core tasks. Amongst those tasks are:

• sending massive newsletters
• image resizing
• http downloads
• updating smart collections
• updating solr, our search server, after product changes
• batch imports
• spam checks

Follow us on Twitter to get updates and notices about new releases.

delayed_job 3.0.0 only supports Rails 3.0+.

delayed_job supports multiple backends for storing the job queue.See the wikifor other backends.

If you plan to use delayed_job with Active Record, adddelayed_job_active_record to yourGemfile.

gem'delayed_job_active_record'

If you plan to use delayed_job with Mongoid, adddelayed_job_mongoid to yourGemfile.

gem'delayed_job_mongoid'

Runbundle install to install the backend and delayed_job gems.

The Active Record backend requires a jobs table. You can create that table byrunning the following command:

rails generate delayed_job:active_recordrake db:migrate

For Rails 4.2+, seebelow

In development mode, if you are using Rails 3.1+, your application code will automatically reload every 100 jobs or when the queue finishes.You no longer need to restart Delayed Job every time you update your code in development.

In Rails 4.2+, set the queue_adapter in config/application.rb

config.active_job.queue_adapter=:delayed_job

See therails guide for more details.

If you are using the protected_attributes gem, it must appear before delayed_job in your gemfile. If your jobs are failing with:

 ActiveRecord::StatementInvalid: PG::NotNullViolation: ERROR:  null value in column "handler" violates not-null constraint

then this is the fix you're looking for.

Delayed Job 3.0.0 introduces a new column to the delayed_jobs table.

If you're upgrading from Delayed Job 2.x, run the upgrade generator to create a migration to add the column.

rails generate delayed_job:upgraderake db:migrate

Call.delay.method(params) on any object and it will be processed in the background.

# without delayed_job@user.activate!(@device)# with delayed_job@user.delay.activate!(@device)

If a method should always be run in the background, you can call#handle_asynchronously after the method declaration:

classDevicedefdeliver# long running methodendhandle_asynchronously:deliverenddevice=Device.newdevice.deliver

#handle_asynchronously and#delay take these parameters:

• :priority (number): lower numbers run first; default is 0 but can be reconfigured (see below)
• :run_at (Time): run the job after this time (probably in the future)
• :queue (string): named queue to put this job in, an alternative to priorities (see below)

These params can be Proc objects, allowing call-time evaluation of the value.

For example:

classLongTasksdefsend_mailer# Some other codeendhandle_asynchronously:send_mailer,:priority=>20defin_the_future# Some other codeend# 5.minutes.from_now will be evaluated when in_the_future is calledhandle_asynchronously:in_the_future,:run_at=>Proc.new{5.minutes.from_now}defself.when_to_run2.hours.from_nowendclass <<selfdefcall_a_class_method# Some other codeendhandle_asynchronously:call_a_class_method,:run_at=>Proc.new{when_to_run}endattr_reader:how_importantdefcall_an_instance_method# Some other codeendhandle_asynchronously:call_an_instance_method,:priority=>Proc.new{|i|i.how_important}end

If you ever want to call ahandle_asynchronously'd method without Delayed Job, for instance while debugging something at the console, just add_without_delay to the method name. For instance, if your original method wasfoo, then callfoo_without_delay.

Delayed Job uses special syntax for Rails Mailers.Do not call the.deliver method when using.delay.

# without delayed_jobNotifier.signup(@user).deliver# with delayed_jobNotifier.delay.signup(@user)# delayed_job running at a specific timeNotifier.delay(run_at:5.minutes.from_now).signup(@user)# when using parameters, the .with method must be called before the .delay methodNotifier.with(foo:1,bar:2).delay.signup(@user)

You may also wish to consider usingActive Job with Action Mailerwhich provides convenient.deliver_later syntax that forwards to Delayed Job under-the-hood.

DJ 3 introduces Resque-style named queues while still retaining DJ-stylepriority. The goal is to provide a system for grouping tasks to be worked byseparate pools of workers, which may be scaled and controlled individually.

Jobs can be assigned to a queue by setting thequeue option:

object.delay(:queue=>'tracking').methodDelayed::Job.enqueuejob,:queue=>'tracking'handle_asynchronously:tweet_later,:queue=>'tweets'

You can configure default priorities for named queues:

Delayed::Worker.queue_attributes={high_priority:{priority: -10},low_priority:{priority:10}}

Configured queue priorities can be overriden by passing priority to the delay method

object.delay(:queue=>'high_priority',priority:0).method

You can start processes to only work certain queues with thequeue andqueuesoptions defined below. Processes started without specifying a queue will run jobsfromany queue. To effectively have a process that runs jobs where a queue is notspecified, set a default queue name withDelayed::Worker.default_queue_name andhave the processes run that queue.

script/delayed_job can be used to manage a background process which willstart working off jobs.

To do so, addgem "daemons" to yourGemfile and make sure you've runrails generate delayed_job.

You can then do the following:

RAILS_ENV=production script/delayed_job startRAILS_ENV=production script/delayed_job stop# Runs two workers in separate processes.RAILS_ENV=production script/delayed_job -n 2 startRAILS_ENV=production script/delayed_job stop# Set the --queue or --queues option to work from a particular queue.RAILS_ENV=production script/delayed_job --queue=tracking startRAILS_ENV=production script/delayed_job --queues=mailers,tasks start# Use the --pool option to specify a worker pool. You can use this option multiple times to start different numbers of workers for different queues.# The following command will start 1 worker for the tracking queue,# 2 workers for the mailers and tasks queues, and 2 workers for any jobs:RAILS_ENV=production script/delayed_job --pool=tracking --pool=mailers,tasks:2 --pool=*:2 start# Runs all available jobs and then exitsRAILS_ENV=production script/delayed_job start --exit-on-complete# or to run in the foregroundRAILS_ENV=production script/delayed_job run --exit-on-complete

Rails 4:replace script/delayed_job with bin/delayed_job

Workers can be running on any computer, as long as they have access to thedatabase and their clock is in sync. Keep in mind that each worker will checkthe database at least every 5 seconds.

You can also invokerake jobs:work which will start working off jobs. You cancancel the rake task withCTRL-C.

If you want to just run all available jobs and exit you can userake jobs:workoff

Work off queues by setting theQUEUE orQUEUES environment variable.

QUEUE=tracking rake jobs:workQUEUES=mailers,tasks rake jobs:work

The following syntax will restart delayed jobs:

RAILS_ENV=production script/delayed_job restart

To restart multiple delayed_job workers:

RAILS_ENV=production script/delayed_job -n2 restart

Rails 4:replace script/delayed_job with bin/delayed_job

Jobs are simple ruby objects with a method called perform. Any object which responds to perform can be stuffed into the jobs table. Job objects are serialized to yaml so that they can later be resurrected by the job runner.

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}endendDelayed::Job.enqueueNewsletterJob.new('lorem ipsum...',Customers.pluck(:email))

To set a per-job max attempts that overrides the Delayed::Worker.max_attempts you can define a max_attempts method on the job

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefmax_attempts3endend

To set a per-job max run time that overrides the Delayed::Worker.max_run_time you can define a max_run_time method on the job

NOTE: this can ONLY be used to set a max_run_time that is lower than Delayed::Worker.max_run_time. Otherwise the lock on the job would expire and another worker would start the working on the in progress job.

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefmax_run_time120# secondsendend

To set a per-job default for destroying failed jobs that overrides the Delayed::Worker.destroy_failed_jobs you can define a destroy_failed_jobs? method on the job

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefdestroy_failed_jobs?falseendend

To set a default queue name for a custom job that overrides Delayed::Worker.default_queue_name, you can define a queue_name method on the job

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefqueue_name'newsletter_queue'endend

On error, the job is scheduled again in 5 seconds + N ** 4, where N is the number of attempts. You can define your ownreschedule_at method to override this default behavior.

NewsletterJob=Struct.new(:text,:emails)dodefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefreschedule_at(current_time,attempts)current_time +5.secondsendend

You can define hooks on your job that will be called at different stages in the process:

NOTE: If you are using ActiveJob these hooks arenot available to your jobs. You will need to use ActiveJob's callbacks. You can find details herehttps://guides.rubyonrails.org/active_job_basics.html#callbacks

classParanoidNewsletterJob <NewsletterJobdefenqueue(job)record_stat'newsletter_job/enqueue'enddefperformemails.each{ |e|NewsletterMailer.deliver_text_to_email(text,e)}enddefbefore(job)record_stat'newsletter_job/start'enddefafter(job)record_stat'newsletter_job/after'enddefsuccess(job)record_stat'newsletter_job/success'enddeferror(job,exception)Airbrake.notify(exception)enddeffailure(job)page_sysadmin_in_the_middle_of_the_nightendend

The library revolves around a delayed_jobs table which looks as follows:

create_table:delayed_jobs,:force=>truedo |table|table.integer:priority,:default=>0# Allows some jobs to jump to the front of the queuetable.integer:attempts,:default=>0# Provides for retries, but still fail eventually.table.text:handler# YAML-encoded string of the object that will do worktable.text:last_error# reason for last failure (See Note below)table.datetime:run_at# When to run. Could be Time.zone.now for immediately, or sometime in the future.table.datetime:locked_at# Set when a client is working on this objecttable.datetime:failed_at# Set when all retries have failed (actually, by default, the record is deleted instead)table.string:locked_by# Who is working on this object (if locked)table.string:queue# The name of the queue this job is intable.timestampsend

On error, the job is scheduled again in 5 seconds + N ** 4, where N is the number of attempts or using the job's definedreschedule_at method.

The defaultWorker.max_attempts is 25. After this, the job is either deleted (default), or left in the database with "failed_at" set.With the default of 25 attempts, the last retry will be 20 days later, with the last interval being almost 100 hours.

The defaultWorker.max_run_time is 4.hours. If your job takes longer than that, another computer could pick it up. It's up to you tomake sure your job doesn't exceed this time. You should set this to the longest time you think the job could take.

By default, it will delete failed jobs (and it always deletes successful jobs). If you want to keep failed jobs, setDelayed::Worker.destroy_failed_jobs = false. The failed jobs will be marked with non-null failed_at.

By default all jobs are scheduled withpriority = 0, which is top priority. You can change this by settingDelayed::Worker.default_priority to something else. Lower numbers have higher priority.

The default behavior is to read 5 jobs from the queue when finding an available job. You can configure this by settingDelayed::Worker.read_ahead.

By default all jobs will be queued without a named queue. A default named queue can be specified by usingDelayed::Worker.default_queue_name.

If no jobs are found, the worker sleeps for the amount of time specified by the sleep delay option. SetDelayed::Worker.sleep_delay = 60 for a 60 second sleep time.

It is possible to disable delayed jobs for testing purposes. SetDelayed::Worker.delay_jobs = false to execute all jobs realtime.

OrDelayed::Worker.delay_jobs can be a Proc that decides whether to execute jobs inline on a per-job basis:

Delayed::Worker.delay_jobs=->(job){job.queue !='inline'}

You may need to raise exceptions on SIGTERM signals,Delayed::Worker.raise_signal_exceptions = :term will cause the worker to raise aSignalException causing the running job to abort and be unlocked, which makes the job available to other workers. The default for this option is false.

Here is an example of changing job parameters in Rails:

# config/initializers/delayed_job_config.rbDelayed::Worker.destroy_failed_jobs=falseDelayed::Worker.sleep_delay=60Delayed::Worker.max_attempts=3Delayed::Worker.max_run_time=5.minutesDelayed::Worker.read_ahead=10Delayed::Worker.default_queue_name='default'Delayed::Worker.delay_jobs= !Rails.env.test?Delayed::Worker.raise_signal_exceptions=:termDelayed::Worker.logger=Logger.new(File.join(Rails.root,'log','delayed_job.log'))

You can invokerake jobs:clear to delete all jobs in the queue.

Good places to get help are:

• Google Groups where you can join our mailing list.
• StackOverflow