Docker configuration for Rails

I am creating a Docker configuration with three separate containers, oriented for a production environment:

• The Rails application (under Puma)
• A PostgreSQL database
• An nginx frontend and SSL proxy

The three images are base upon alpine variants of official images:

• ruby:alpine
• postgres:alpine
• nginx:stable-alpine

Note this is not a Docker tutorial. I assume you have a Docker environment already running. Follow the superb official instructions if you need to set it up.

Build and first start instructions

Note I ommit the sudo in all docker commands: prepend it if you need it in your environment.

1. Copy the contents of the app directory to the root of your Rails application.

2. Build the Rails image (this is the longest step):

    docker build -t rails_app /path/to/your/app
   

3. Build the PostgreSQL image and start the container:

    docker build -t rails_db postgres
    docker run -d --name rails_db rails_db
   

   Allow some time for the database to be created and started before launching rake db:migrate to populate it later.

4. Enter the nginx directory.

   • There are two files, cert.key and cert.crt: these are sample self-signed certificates, replace them with your certificates (with the same file names and without passphrase).
   • Review default.site. In particular, you may want to change server_name localhost; to a better value.

5. From the nginx directory, build the nginx image:

    docker build -t rails_nginx .
   

6. Populate the DB with a one shot run of rake in the Rails image with the following command:

    docker run -t --rm --link rails_db:db rails_app rake db:migrate
   

7. Start the Rails application server, replacing YOUR_SECRET with a random string of your choice.

    docker run -d --link rails_db:db --name rails_app -e SECRET_KEY_BASE=YOUR_SECRET rails_app
   

8. Start the nginx server:

    docker run -d -P --link rails_app:rails --volumes-from rails_app --name rails_nginx rails_nginx
   

9. Check mapped ports by running docker ps (in this case, 32769 for HTTP and 32768 for HTTPS):

    # docker ps
    CONTAINER ID        IMAGE               COMMAND                  CREATED             STATUS              PORTS                                           NAMES
    e66a6c7b7c48        rails_nginx         "nginx -g 'daemon off"   20 minutes ago      Up 20 minutes       0.0.0.0:32769->80/tcp, 0.0.0.0:32768->443/tcp   rails_nginx
    de5b5b34f3dd        rails_app           "/bin/sh -c -l 'bundl"   21 minutes ago      Up 21 minutes       3000/tcp                                        rails_app
    0e6284aed5cd        rails_db            "docker-entrypoint.sh"   28 minutes ago      Up 28 minutes       5432/tcp                                        rails_db
   

10. Verify you can access the application by visiting https://localhost:port/, e.g. https://localhost:32768/

Stopping

Run

docker stop rails_nginx rails_app rails_db

Subsequent starts

Run

docker start rails_db rails_app rails_nginx

Running on fixed ports

If you want to use specific ports, you need to specify them when first starting the nginx image.

Remove the container, if you started it with -P as instructed above, with

docker rm -f rails_nginx

To start the nginx container in the standard ports, run

docker run -d -p 80:80 -p 443:443 --link rails_app:rails --volumes-from rails_app --name rails_nginx rails_nginx

This will make the container listen in all host interfaces. Change the ports before the colons (:) if you want specific but different ports.