Example Rails Heroku cloud-app that uses Resque with Redis
Ruby JavaScript
Switch branches/tags
Nothing to show
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


Redis and Resque on Heroku

This is an extremely basic Ruby on Rails Application that demonstrates how a Rails 2.3 application hosted on Heroku can use Resque with Redis.

Adapted from Resque with Redis To Go by James R. Bracy and Defunkt's Resque

The main difference is that this application runs on Rails 2.3x rather than Rails 3 and this application has no database dependency. Mr. Bracy has a Rails 3.x version of this example.

Create the basic Rails application

  • New Rails app: rails heroku_redis_example
  • cd heroku_redis_example

Remove the database dependency

  • Uncomment this line from environment.rb: config.frameworks -= [ :active_record, :active_resource, :action_mailer ]
  • rm RAILS_ROOT/config/database.yml

Configure Gems with Bundler

  • Create and add this to Gemfile

    source 'http://rubygems.org'
    gem "rails", ">=2.3.8"
    gem 'resque'
    gem 'SystemTimer'
  • bundle install

  • bundle lock

  • Add a preinitializer file to work around Bundler issues: (From: http://stackoverflow.com/questions/2170697/bundler-isnt-loading-gems)

      # Require the preresolved locked set of gems.
      require File.expand_path('../.bundle/environment', __FILE__)
    rescue LoadError
      # Fallback on doing the resolve at runtime.
      require "rubygems"
      require "bundler"
    # Auto-require all bundled libraries.

Configure Resque

  • create RAILS_ROOT/config/resque.rb and add:

    ENV["REDISTOGO_URL"] ||= "redis://username:password@host:1234/"
    uri = URI.parse(ENV["REDISTOGO_URL"])
    Resque.redis = Redis.new(:host => uri.host, :port => uri.port, :password => uri.password)
    Dir["#{Rails.root}/app/jobs/*.rb"].each { |file| require file }

Configure Resque-web for debugging

  • Create RAILS_ROOT/config.ru

    # A mix of the config.ru at Defunkt's Resque and the original help
    # file mentioned above
    require "config/environment"
    require 'resque/server'
    require 'logger'
    use Rails::Rack::LogTailer
    use Rails::Rack::Static
    use Rack::ShowExceptions
    # Set the AUTH env variable to your basic auth password to protect Resque.
    AUTH_PASSWORD = 'secret'
      Resque::Server.use Rack::Auth::Basic do |username, password|
        password == AUTH_PASSWORD
    run Rack::URLMap.new \
      "/"       => ActionController::Dispatcher.new,
      "/resque" => Resque::Server.new

Add Resque Worker Class

  • Create a job in app/jobs/trogdor.rb

    class Trogdor
      @queue = :terrorize
      def self.perform(target)
        puts "Burninating the #{target}!"

Create Controller which will initiate Resque worker

In this step we write the code that will be initiated by a web request and itself initiate a worker.

  • RAILS_ROOT/script/generate controller trogdor

  • Create this method in the controller:

    def burninate
      Resque.enqueue(Trogdor, params[:target])
      render :text => "Telling Trogdor to burninate #{params[:target]}."

Configure the route to our controller

Here we tell Rails how to handle a certain web request in order to invoke the burninate method in TrogdorController

  • Set RAILS_ROOT/config/routes.rb to:

    ActionController::Routing::Routes.draw do |map|
      map.trogdor 'trogdor/burninate/:target', :controller => 'trogdor', :action => 'burninate'

Add Resque task to Rake

  • In lib/tasks/resque.rake:

    require 'resque/tasks'
    task "resque:setup" => :environment do
      ENV['QUEUE'] = '*'
    desc "Alias for resque:work (To run workers on Heroku)"
    task "jobs:work" => "resque:work"

Add project to Git

  • Add application to Git: git init
  • Add all files to Git: git add .
  • Commit to your Git repository: git commit -m 'Initial commit'

Add project to Heroku

  • Create the application with Heroku: heroku create
  • Add the Redis Heroku addon: heroku addons:add redistogo
  • Push your code to heroku: git push heroku master

Test the application

Verify Resque-web works:

Verify the application works:

Kickoff the worker using Heroku rake command syntax:

  • Start Resque worker: heroku rake resque:work --app young-sunrise-78 --trace

Observe for Maximum Win

  • Observe the output of the console command to heroku rake:

    ~/heroku_redis_example(master) $ heroku rake resque:work --app young-sunrise-78 --trace
    (in /disk1/home/slugs/258287_149fab6_faff/mnt)
    ** Invoke resque:work (first_time)
    ** Invoke resque:setup (first_time)
    ** Invoke environment (first_time)
    ** Execute environment
    ** Execute resque:setup
    ** Execute resque:work
    Burninating the countryside!


Now, let's imagine you have heroku rake running but you'd like to extend or correct some aspect of your Job. In order to do this you need to clear out the Job environment on Heroku.

  • ^C (Control C) to escape your heroku rake task
  • Make your changes to your code
  • Commit and push to the heroku remote
  • Go to Resque-web to verify that 0 of 0 workers are working. What you are doing is effectively waiting for the worker to timeout and die.
  • When you have 0 of 0 workers working, heroku rake resque:work again.

This new rake call will then load up the new code you just deployed. Otherwise, it would be using old code and not reflect your changes/improvements.


Once you are happy with your application and your background jobs it's time to bite the bullet and start worker jobs in Heroku officially. Simply starting workers manually with heroku rake is insufficient as this will create workers only for as long as it takes for Heroku to timeout.

To create a long-running Heroku worker:

  ~/heroku_redis_example(master) $ heroku workers 1 --app young-sunrise-78
    This action will cause your account to be billed at the end of the month
    For more information, see http://docs.heroku.com/billing
    Are you sure you want to do this? (y/n) y
  young-sunrise-78 now running 1 worker