[Rails 6.0, Docker] I tried to summarize the Docker environment construction and commands necessary to create a portfolio

I would like to summarize the commands and environment construction that I needed to create a portfolio with Rails. It is necessary at the portfolio creation level.

Overview

1. Frequently used commands for handling docker-compose
2. Description of various files (Dockerfile, docker-compose.yml, database.yml)

1. Frequently used commands for handling docker-compose

rails command (rails g model user, rails db: migrate, etc.)

terminal

$ docker compose run --rm web <rails command>

The --rm option is a command to delete the container after execution. If you do not attach this, an unnecessary container will be created and it will put pressure on the memory, so be sure to attach it.

Connect to standard I / O (what is displayed on the terminal when developing locally)

terminal

$ docker attach <web server container name>

You can check the communication status by connecting to the standard input / output on the server. This command is also used when debugging with binding.pry. Detach with proper processing. (If you ctrl + C, the container will drop)

Ctrl + P, Ctrl + Q

Connect to database (mysql)

terminal

$ docker exec -it <db container name> bash

From inside the container

terminal

$ mysql -u root -p

Create a Docker image, start and run the container.

terminal

$ docker-compose stop
#Dockerfile or docker-compose.after yml modification
$ docker-compose up -d --build

For when modifying Dockerfile or docker-compose.yml.

Delete command

Even with the --rm option, containers and images that you do not need to develop will accumulate. Here is the command to delete them.

Bulk delete dead containers, untagged images, unused volumes, unused networks

terminal

docker system prune

• The volume in use is not deleted, so it's okay.
• Not recommended if multiple projects are running

Bulk deletion of stopped containers

terminal

docker container prune

Delete unused images in bulk

terminal

docker image prune

Description of various files

Dockerfile First, the Dockerfile. First, read the official Document. https://docs.docker.com/compose/rails/

Dockerfile

FROM ruby:2.7.1

#Debian installation
RUN curl -sS https://dl.yarnpkg.com/debian/pubkey.gpg | apt-key add - \
    && echo "deb https://dl.yarnpkg.com/debian/ stable main" | tee /etc/apt/sources.list.d/yarn.list

RUN apt-get update -qq && apt-get install -y nodejs yarn imagemagick mariadb-client vim

RUN mkdir /myapp
WORKDIR /myapp
COPY Gemfile /myapp/Gemfile
COPY Gemfile.lock /myapp/Gemfile.lock
RUN bundle install
COPY . /myapp

COPY entrypoint.sh /usr/bin/
RUN chmod +x /usr/bin/entrypoint.sh
ENTRYPOINT ["entrypoint.sh"]
EXPOSE 3000

CMD ["rails", "server", "-b", "0.0.0.0"]

Yarn installation

From rails 6.0, the JavaScript compiler has been changed to webpacker, and yarn, the package manager required to install webpacker, is installed.

Install ImageMagick

Installed because I wanted to use MiniMagic for image upload. If you want to use MiniMagick with your Rails app, you need to install ImageMagick in the environment where your Rails app is running. The following is from the minimagic ReadMe

ImageMagick or GraphicsMagick command-line tool has to be installed. You can check if you have it installed by running

Installation of mariadb-client

Install if you want to use rails db: console. There is an article that says mysql-client, but be careful because it is now unified to mariadb-client. It may not be necessary because you can access the database with the above command.

install vim

Install the editor because you will need it. After deploying the app to the production environment, I wanted to edit credentials.yml.enc with the following command, but I used vim at that time.

terminal

$ docker-compose run -e EDITOR=vim web rails credentials:edit

docker-compose.yml Next is docker-compose.yml

docker-compose.yml

version: "3"
services:
  db:
    image: mysql:8.0
    restart: always
    environment:
      MYSQL_ROOT_PASSWORD: password
    ports:
      - 4306:3306
    
    command: --default-authentication-plugin=mysql_native_password
    volumes:
      - mysql-data:/var/lib/mysql

  web:
    build: .
    tty: true
    stdin_open: true
    #A server is already running when rails s.Error may appear. To prevent this, rm-f tmp/pids/server.It is pid.
    command: bash -c "rm -f tmp/pids/server.pid && bundle exec rails s -p 3000 -b '0.0.0.0'"
    volumes:
      - .:/myapp
      - bundle:/usr/local/bundle
    ports:
      - "3100:3000"
    depends_on:
      - db

volumes:
  bundle:
    driver: local
  mysql-data:
    driver: local

Port number setting

docker-compose.yml

db
  ports:
    - 4306:3306
web
  ports:
    - 3100:3000

The port number of the host machine is on the left and the port number of the container is on the right. If MySQL is (LISTEN) on the host or you want to use localhost: 3000, set the port numbers used by the machine to 4306 and 3100, respectively, because the ports used are duplicated. If you don't wear it, you can use 3307 or 10000.

Changed mysql authentication method

docker-compose.yml

command: --default-authentication-plugin=mysql_native_password

The default authentication method has changed from MySQL 8.0 (caching_sha2_password), but since the Rails app does not yet support it, change it back to mysql_native_password.

Enable standard input

docker-compose.yml

tty: true
stdin_open: true

Very important for debugging. If you do not do this, binding.pry etc. will not be possible. tty: true A setting that enables pseudo-terminals, which corresponds to the -t option of docker run. Even if it is enabled, it cannot be entered, so stdin_open: true below is always included as a set. stdin_open: true A setting that corresponds to the -i option of docker run and keeps the standard input connected. This makes it possible to execute commands.

Control container boot order

docker-compose.yml

depends_on:
      - db

Control the order in which containers are started with depends_on (in this case, db → web)

Creating a Data Volume container

docker-compose.yml

volumes:
  bundle:
    driver: local
  mysql-data:
    driver: local

Normally, if you recreate the container, all the data in the bundle installed files and database will be deleted, so you will have to build again or register the data. It's too annoying to use a Data Volume container. Data Volume container is a container that has only data and uses a mechanism to share between containers. The settings to create this with the names bundle and mysql-data are as above.

Mount to volume

docker-compose.yml

volumes:
  - mysql-data:/var/lib/mysql
volumes:
  - .:/myapp
  - bundle:/usr/local/bundle

Persist data using the Data Volume container created earlier. Now, even if you destroy the container and rebuild it, the data in the database will remain as it is, and Gem will be installed and ready to use.

database.yml Finally database.yml

database.yml

default: &default
  adapter: mysql2
  encoding: utf8mb4
  pool: <%= ENV.fetch("RAILS_MAX_THREADS") { 5 } %>
  username: <%= ENV.fetch("MYSQL_USERNAME", "root") %>
  password: <%= ENV.fetch("MYSQL_PASSWORD", "password") %>
  host: <%= ENV.fetch("MYSQL_HOST", "db") %>

development:
  <<: *default
  database: myapp_development

Character code

database.yml

encoding: utf8mb4

I wanted to register pictograms in the database, so the character code is utf8mb4 instead of utf8.

Please let me know if you have any other useful settings or commands ^^

reference

https://docs.docker.com/compose/rails/ Introduction to Docker / Kubernetes Practical Container Development https://qiita.com/neko-neko/items/abe912eba9c113fd527e https://qiita.com/chisaki0606/items/68e21d9a31f1eaaeac00 https://qiita.com/nsy_13/items/9fbc929f173984c30b5d