Deploying Sinatra Apps with Heroku

1.) In Terminal, login to Heroku: heroku login.

2.) Initialize a git repository for your project : git init (Careful! Do not nest Git repositories).

3.) Add a specific Ruby version number to the Gemfile if you have not done so already. Example: ruby 2.3.3

4.) Add specific production and development methods to app_name.rb (optional). Example:

# app_name.rb
require 'sinatra'
require 'sinatra/reloader' if development?

5.) Switch to use a production web server in the Gemfile. Whichever server you declare, it should be able to handle concurrent requests. Webricks, the default server, is single-threaded, and therefore cannot handle concurrent requests. Using Webricks in production could result in application timeouts.

# Gemfile
group :production do
  gem "puma"
end

6.) Add a config.ru file to the project root folder. This file tells the web server how to start the app.

# config.ru
require '.app_name'
run Sinatra::Application 

7.) Add a Procfile to the project root folder. This file is a mechanism for declaring what commands are run by your application’s dynos on the Heroku platform.

# Procfile
web: bundle exec puma -t 5:5 -p ${PORT:-3000} -e ${RACK_ENV:-development}

8.) In Terminal, run bundle install .

9.) Test on the local Heroku server. In Terminal, run heroku local, and view at localhost:5000/route_name.

Now, the app is running on the Puma server not the Webricks server.

10.) Create a Heroku app. In Terminal, run heroku apps:create $app-name.

11.) Push your project to the new Heroku app. In Terminal, run git push heroku master. If you want to keep your project on a local branch, run git push heroku branch-name:master instead.

Now, your app is available to view from any device or browser.

12.) View the app using URL in the terminal output by running heroku open.

Example URL: https://example-app-name.herokuapp.com/example-route