Docker for Ruby. Docker Containers for Ruby Development Enviroment. It is heavily inspired by Laradock project.

• Getting Started
  • Environment Values
  • Files and Folders
  • Docker-Sync-Specific Setup
  • Create a Project
  • Run Rails App
• App With Database
  • Create App
  • Setup Database
  • Run App

Create .env file by copying from env.example

cp env.example .env

This setup suggests with below folders structure. You should have a main folders that holds 2 folders:

1. apps - where you setup your multiple ruby/rails projects
2. dockery - where you checkout Dockery project. All commands (docker-sync, docker-compose, synch.sh) should be run in this folder.

<main_folder>
|
|--apps/ (contains multple Ruby/Rails projects)
|  |-- blog/
|  |-- other-awesome-project/
|
|--dockery/ (Dockery project)

This structure is define by below .env entry

APPLICATIONS_PATH=../apps

If you wish to have different place to hold your projects you'll have to change that entry.

ruby workspace has 2 versions of volume mapping; cached and dockery-sync. This is because for Mac, it requires performance boost with Docker Sync. Refer to Docker-Sync-Specific Setup for more details. Basically copy either docker-compose.cached.yml or docker-compose.docker-sync.yml to docker-compose.override.yml in order to get your ruby workspace volumes mapped to your host machine.

Docker on Mac is known to have performance issue with its file syncing. To improve the speed, we can use Docker Sync. docker-sync.yml is Docker Sync configuration file which maps applications and bundle folders.

Run below command to start Docker Sync service

./sync.sh

Note: Under the hood, the script runs docker-sync clean and docker-sync start to start fresh with docker-sync service. If the folders to be managed by docker-sync are missing, this script will create them for you.

If you wish to stop the docker-sync service, run docker-sync stop.

Refer to this Docker Sync commands wiki for other available commands.

docker-compose.dockery-sync.yml is the override file which overrides the volumes configurations from docker-compose.yml to Docker Sync managed volumes.

There are 2 ways to use this override file.

• 1st option

First option is to use docker-compose -f flag as shown below:

docker-compose -f docker-compose.yml -f docker-compose.docker-sync.yml up -d ruby

• 2nd option

Second option, is to copy docker-compose.docker-sync.yml to docker-compose.override.yml. docker-compose by default will read both docker-compose.yml as default and docker-compose.override.yml as override configurations.

cp docker-compose.docker-sync.yml docker-compose.override.yml

docker-compose up -d ruby

Each time you changed something in .env that affects dockery-sync source folder , i.e. $MY_RUBY_VERSION, you will need to run ./sync.sh to resync your project folder.

If just want to start fresh and the project folder is empty, you can generate new rails project with below command.

docker-compose run ruby rails new blog

This will create blog app within the apps folder.

Set the application folder name that you wish to run. For example, if you wish to run blog application, you set that project folder name to APP key in .env file

APP=blog

Assume you already have an app named blog in apps folder.

docker-compose up -d ruby

You could also override the APP env value inline with the command.

APP=blog docker-compose up -d ruby

The Rails app will be available at http://localhost:8000

Run below command to create new rails blog project with postgresql database.

docker-compose run ruby rails new blog --database=postgresql

Note: You can switch postgresql to mysql if that is your choice of database.

There is a helper script, rails-new that will do the same but will create following database.yml.

docker-compose run ruby rails-new blog postgresql

# database.yml
development: &default
  adapter: <%= ENV.fetch("DB_ADAPTER") %>
  host: <%= ENV.fetch("DB_HOST") %>
  port: <%= ENV.fetch("DB_PORT") %>
  username: <%= ENV.fetch("DB_USER") %>
  password: <%= ENV.fetch("DB_PASSWORD") %>
  database: <%= ENV.fetch("APP_NAME") + "_development" %>
  encoding: utf8
  min_messages: warning
  pool: <%= Integer(ENV.fetch("DB_POOL", 5)) %>
  reaping_frequency: <%= Integer(ENV.fetch("DB_REAPING_FREQUENCY", 10)) %>
  timeout: 5000

test:
  <<: *default
  database: <%= ENV.fetch("APP_NAME") + "_test" %>

production: &deploy
  encoding: utf8
  min_messages: warning
  pool: <%= [Integer(ENV.fetch("MAX_THREADS", 5)), Integer(ENV.fetch("DB_POOL", 5))].max %>
  timeout: 5000
  url:  <%= ENV.fetch("DATABASE_URL", "") %>

Start the postgresql service in the background.

docker-compose run up -d postgres

Bundle install.

docker-compose run ruby bash -c 'cd blog && bundle check || bundle install'

Setup the database.

docker-compose run ruby bash -c 'cd blog && bin/rails db:setup'

Migrate database.

docker-compose run ruby bash -c 'cd blog && bin/rails db:migrate'

Run the app.

APP=blog docker-compose up -d ruby postgres

The Rails app will be available at http://localhost:8000