Rails Application Development using Bitnami Docker Images

We increasingly see developers adopting two strategies for development. Using a so called “micro services” architecture and using containers for development. At Bitnami, we have developed tools and assets that dramatically lowers the overhead for developing with this approach.

If you’ve never tried to start a project with containers before, or you have tried it and found the advice, tools, and documentation to be chaotic, out of date, or wrong, then this tutorial may be for you.

In this tutorial we walk you through using the Bitnami docker images during the development lifecycle of a Ruby on Rails application.

Why Docker?

We think developers are adopting containers for development because they offer many of the same advantages as developing in VMs, but with lower overhead in terms of developer effort and development machine resources. With Docker, you can create a development environment for your code, and teammates can pull the whole development environment, install it, and quickly get started writing code or fixing bugs.

Docker development environments are more likely to be reproducible than VMs because the definition of each container and how to build it is captured in a Dockerfile.

Docker also has a well known and standard API so tools and cloud services are readily available for docker containers.

The Bitnami Approach

When we designed and built our development containers, we kept the following guiding principles in mind:

1. Infrastructure should be effort free. By this, we mean, there are certain services in an application that are merely configured. For example, databases and web servers are essential parts of an application, but developers should depend on them like plumbing. They should be there ready to use, but developers should not be forced to waste time and effort creating the plumbing.

2. Production deployment is a late bound decision. Containers are great for development. Sometimes they are great for production, sometimes they are not. If you choose to get started with Bitnami containers for development, it is an easy matter to decide later between monolithic and services architectures, between VMs and Containers, between Cloud and bare metal deployment. This is because Bitnami builds containers specifically with flexibility of production deployment in mind. We ensure that a service running in an immutable and well tested container will behave precisely the same as the same service running in a VM or bare metal.

Assumptions

First, we assume that you have the following components properly setup:

• Docker Engine
• Docker Compose
• Docker Machine

The Docker documentation walks you through installing each of these components.

We also assume that you have some beginner-level experience using these tools.

Note:

If your host OS is Linux you may skip setting up Docker Machine since you'll be able to launch the containers directly in the host OS environment.

Further, we also assume that your application will be using a database. In fact, we assume that it will be using MariaDB. Of course, for a real project you may be using a different database, or, in fact, no database. But, this is a common set up and will help you learn the development approach.

Download the Bitnami Orchestration File for Rails development

We assume that you're starting the development of the Ruby on Rails application from scratch. So lets begin by creating a directory for the application source where we'll be bootstrapping a Rails application:

$ mkdir ~/workdir/myapp
$ cd ~/workdir/myapp

Next, download our Docker Compose orchestration file for Rails development:

$ curl -L "https://raw.githubusercontent.com/bitnami/bitnami-docker-rails/master/docker-compose.yml" > docker-compose.yml

We encourage you to take a look at the contents of the orchestration file to get an idea of the services that will be started for Rails development.

Run

Lets put the orchestration file to the test:

$ docker-compose up

This command reads the contents of the orchestration file and begins downloading the Docker images required to launch each of the services listed therein. Depending on the network speeds this can take anywhere from a few seconds to a couple minutes.

After the images have been downloaded, each of the services listed in the orchestration file is started, which in this case are the mariadb and myapp services.

As mentioned earlier, the mariadb service provides a database backend which can be used for the development of a data-driven Rails application. The service is setup using the bitnami/mariadb docker image and is configured with the default credentials.

The second service thats started is named myapp and uses the Bitnami Rails development image. The service mounts the current working directory (~/workdir/myapp) at the /app location in the container and provides all the necessary infrastucture to get you started developing a data-driven Rails application.

Once the WEBrick application server has been started, visit port 3000 of the Docker Machine in your favourite web browser and you'll be greeted by Rails welcome page.

Lets inspect the contents of the ~/workdir/myapp directory:

~/workdir/myapp # ls
Gemfile             app/                db/                 public/
Gemfile.lock        bin/                docker-compose.yml  test/
README.rdoc         config/             lib/                tmp/
Rakefile            config.ru           log/                vendor/

You can see that we have a new Rails application bootstrapped in the ~/workdir/myapp directory of the host and is being served by the WEBrick application server running inside the Bitnami Rails development container.

Since the application source resides on the host, you can use your favourite IDE for developing the application. Only the execution of the application occurs inside the isolated container environment.

That’s all there is to it. Without actually installing a single Rails component on the host you have a completely isolated and highly reproducible Rails development environment which can be shared with the rest of the team to get them started building the next big feature without worrying about the plumbing involved in setting up the development environment. Let Bitnami do that for you.

In the next sections we take a look at some of the common tasks that are involved during the development of a Rails application and how we go about executing those tasks.

Executing commands

You may recall that we've not installed a single Rails component on the host and that the entire Rails development environment is running inside the myapp service container. This means that if we wanted to execute rake or any other Rails command, we'd have to execute it inside the container.

This may sound like a complex task to achieve. But don't worry, Docker Compose makes it very simple to execute tasks inside a service container using the exec command. The general form of the command looks something like the following:

$ docker-compose exec <service> <command>

This instructs Docker Compose to execute the command specified by <command> inside the service container specified by <service>. The return value of the docker-compose command will reflect that of the specified command.

With this information lets try listing the available rake tasks:

$ docker-compose exec myapp bundle exec rake -T

Next, lets try to get some information about our development environment by executing the about task:

$ docker-compose exec myapp bundle exec rake about

How about loading the Rails console?

$ docker-compose exec myapp rails console

You get the idea..

Before we wrap up this subject, lets take a look at one of the most common tasks that's performed during the development lifecycle of a Rails application. Yes, we're going to use rails generate to generate a scaffold.

$ docker-compose exec myapp rails generate scaffold User name:string email:string

The above command will create the User model with name and email properties. Before we can start using this new scaffold we need to apply the migrations, to the app_developmemt database, that implement the User model.

$ docker-compose exec myapp bundle exec rake db:migrate

Sure enough, we're executing the db:migrate rake task

Note

Database migrations are automatically applied during the start up of the myapp service container. This means that the myapp service could also be restarted to apply the database migrations.

$ docker-compose restart myapp

Thats it! Visit the /users resource of the Rails application and you should be able to interact with the newly created User model.

Installing Gems

The functionality of a Rails application can be augmented using Ruby gems. Thousands of gems developed by the Rails development community are available on Rubygems.org and can be used to quickly add functionality to our Rails applications. In this section, we look at adding new gems for our application.

As a Rails developer you must already be aware that additional gems required by a Rails application should be specified in the Gemfile of the Rails application.

For demonstration purposes we'll add the latest version of the httparty gem to our Gemfile using:

$ echo "gem 'httparty'" >> Gemfile

After making changes to the Gemfile, all we need to do is restart the myapp service using:

$ docker-compose restart myapp

When the myapp service is restarted, it checks to see if any new gems need to be installed using bundle check. If this is found to be the case, then bundle install command is invoked to install the missing gems.

That all there is to it. We hope that you find our Rails development image useful in your quest to world domination. Happy hacking!