Installing Docker Community Edition on Ubuntu Bionic Beaver

• compute
• docker
• getting
• started

Docker is a tool designed to create, deploy, and run applications using containers.

Containers allow to pack an application with all requirements necessary, such as libraries and other dependencies, and ship it out as one package.

This ensures that an application can run on any Linux machine, regardless of the specific settings of the host machine, which might be different from the development environment.

Installing Docker Community Edition

Docker Community Edition (CE) is the classic open source Docker environment.

1. Before installing the latest version of Docker on your server, make sure to remove any old version that might have already been installed.

   sudo apt-get remove docker docker-engine docker.ioCopy code

   Note:

   apt-get might report that none of these packages is installed. This is fine, and you can continue with the installation.

2. Update the apt package cache to make sure all required software is referenced.

   sudo apt-get updateCopy code

3. As Docker uses HTTPS for their repository, install the required packages so apt can download them via a secure connection.

   sudo apt-get install apt-transport-https ca-certificates curl software-properties-commonCopy code

4. Add Docker’s official SSH key to apt.

   curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -Copy code

5. Verify the fingerprint of the key (9DC8 5822 9FC7 DD38 854A E2D8 8D81 803C 0EBF CD88), by looking at the last 8 characters.

   sudo apt-key fingerprint 0EBFCD88Copy code

6. Add the repository for the stable version of Docker. For a x86_64 based server, use the following command.

   sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"Copy code

   Note:

   On ARM-based servers, replace amd64 either with armhf or arm64, depending on the architecture of the CPU.

7. Update the apt packet cache.

   sudo apt-get updateCopy code

8. Install the latest version of Docker.

   sudo apt-get install docker-ceCopy code

9. Verify that Docker has been installed by running the Hello World container

   sudo docker run hello-worldCopy code

Managing Docker as a non-root user

The docker daemon always runs as root user and binds to a Unix socket instead of a TCP port. This Unix socket is owned by the root user by default, and other users can only access it by using sudo.

If you don’t want to use sudo with the docker command, create a Unix group called docker and add users to it. The daemon makes the ownership of the Unix socket read/writable by the docker group when it starts.

1. Create the docker group.

   sudo groupadd dockerCopy code

2. Add your user to the docker group.

   sudo usermod -aG docker $USERCopy code

3. Log out and log back in to re-evaluate your group membership.

   Note:

   If you are executing these commands on a virtual machine, it may be necessary to restart the VM for the changes to take effect.

4. Verify that you can run docker commands without sudo by downloading a test-image and running it in a container.

   docker run hello-worldCopy code

If you were using docker CLI using sudo before adding your user to the docker group, you might encounter an error message, telling you that your ~/.docker/ directory was created with incorrect permissions:

WARNING: Error loading config file: /home/user/.docker/config.json -stat /home/user/.docker/config.json: permission deniedCopy code

This problem can either be fixed by removing the ~/.docker/ directory or by changing the ownership and permissions using the following commands

$ sudo chown "$USER":"$USER" /home/"$USER"/.docker -R$ sudo chmod g+rwx "/home/$USER/.docker" -RCopy code

Configuring Docker to start on boot

Ubuntu uses systemd to manage services. Docker comes with a pre-built systemd configuration, and you can enable the service to start automatically on boot.

sudo systemctl enable dockerCopy code

If you don’t want Docker to start automatically, disable it in the same way.

sudo systemctl disable dockerCopy code

Building your first container

In the past, all software dependencies required to run your application had to be installed on either the physical machine or a VM.

Managing all these machines can become quite hard and if you ship your application to one of your customers and they have a different system configuration and your application might not work as expected.

Let’s say you want to develop a Python web application. Docker allows you to grab a portable Python runtime as an image, no installation necessary.

You develop your application, and your build can include the base Python image right alongside your app code, ensuring that your app, its dependencies, and the runtime, perfectly designed for each other are shipped together.

Creating the Dockerfile

Portable images are defined in a file called Dockerfile and define what goes on in the environment inside your container.

Resources of the server (like network interfaces or disk drives) are virtualized inside this environment, and it is isolated from the rest of your system.

This means you have to map ports to the outside world and define precisely about what files you want to copy into that environment.

Thanks to this, you can expect that the build of your application defined in the Dockerfile behaves precisely the same, no matter wherever it runs.

1. Create a new directory to develop your environment

   mkdir my-applicationCopy code

2. Enter into the directory

   cd my-applicationCopy code

3. Create a new file, called Dockerfile, copy & paste the following content into it and save the file

   # Use an official Python runtime as a parent imageFROM python:2.7-slim
   # Set the working directory to /appWORKDIR /app
   # Copy the current directory contents into the container at /appADD . /app
   # Install the Flask frameworkRUN pip install --trusted-host pypi.python.org Flask
   # Make port 80 available to the world outside this containerEXPOSE 80
   # Define environment variableENV NAME World
   # Run app.py when the container launchesCMD ["python", "app.py"]Copy code

Creating the Application

Now as you have created your Dockerfile, it is time to create your application. Open a text editor, copy paste the following content and save the file as app.py.

from flask import Flaskimport osimport socket

app = Flask(__name__)
@app.route("/")def hello():
    html = "<h1>Hello {name}!</h1>" \           "<b>Hostname:</b> {hostname}<br/>"    return html.format(name=os.getenv("NAME", "world"), hostname=socket.gethostname(),)
if __name__ == "__main__":    app.run(host='0.0.0.0', port=80)Copy code

Building the Container

Now as you have created a Dockerfile and the application, it is time to build the container.

1. Check if you are in the correct directory and that both files are present

   lsDockerfile        app.pyCopy code

2. Build the image. The -t tag allows you to set a friendly name for the image. The application will recover the environment variable we have set in the dockerfile and display the containers host name.

   docker build -t helloworld .Copy code

   Docker will now download the python base image and process the tasks defined in the Dockerfile.

   Sending build context to Docker daemon  3.584kBStep 1/7 : FROM python:2.7-slim2.7-slim: Pulling from library/python683abbb4ea60: Pull complete8af590abc616: Pull completef6cf6c5b7946: Pull completee61a0272532e: Pull completeDigest: sha256:8ccd3a1e1836e2d1d1abda0c7f56b726fe70e2f4148190ce38456021c5ffc744Status: Downloaded newer image for python:2.7-slim---> 02ca219cf841Step 2/7 : WORKDIR /appRemoving intermediate container b4186d6dccd7---> ddc94a21efbbStep 3/7 : ADD . /app---> 84df6bca17f6Step 4/7 : RUN pip install --trusted-host pypi.python.org Flask---> Running in e992f7fcc7d8Collecting Flask  Downloading https://files.pythonhosted.org/packages/7f/e7/08578774ed4536d3242b14dacb4696386634607af824ea997202cd0edb4b/Flask-1.0.2-py2.py3-none-any.whl (91kB)  [...]Removing intermediate container e992f7fcc7d8---> 5da35ac6d845Step 5/7 : EXPOSE 80---> Running in 445b607f5795Removing intermediate container 445b607f5795---> 1b5471afba0cStep 6/7 : ENV NAME World---> Running in 2bb6c11c5a22Removing intermediate container 2bb6c11c5a22---> 9c0b3c60260cStep 7/7 : CMD ["python", "app.py"]---> Running in d01c685cfda3Removing intermediate container d01c685cfda3---> b584aacca9d8Successfully built b584aacca9d8Successfully tagged helloworld:latestCopy code

   Your image is now available in your local docker registry:

   docker image ls
   REPOSITORY            TAG                 IMAGE IDhelloworld         latest              b584aacca9d8Copy code

Running the Application

Now it is time to run your application for the first time.

Use the -p option to map the port 4000 of your computer to port 80 of the container:

docker run -p 4000:80 helloworldCopy code

You will see a message similar to the following:

You can notice that the application is running on port 80.

Remember, this is the internal port of your container and can reach the application by typing http://localhost:4000 in the address bar of your browser:

The application displays the internal hostname of the container and retrieves the word “World” from the environment variable that you have set in the Dockerfile.

Sharing the Application

Docker lets you easily ship your image to a registry to share it with others.

We use the public docker registry in this example, but you can also set up a private registry to distribute the image on your infrastructure.

1. Login into the Docker registry:

   docker loginCopy code

   To associate a local image with a repository on a registry, you have to use the notation username/repository:tag.

   The tag is not mandatory, but recommended as it is the mechanism that registries use to give Docker images a version.

   To push your image you have to run docker tag with your username, repository, and tag, so your image uploads to the desired destination.

2. Tag the image.

   docker tag imagename username/repository:tagCopy code

3. Publish your image.

   docker push username/repository:tagCopy code

Once the image has been pushed to Docker Hub it will be publicly available and you can see the image with its pull command in your account.

Pulling and running an Image from a Remote Repository

As your image is available on the Docker Hub now, you can run it from anywhere.

docker run -p 4000:80 username/repository:tagCopy code

If the image is not available yet on the local machine, Docker will download it from the registry.

No matter on which machine you run it, it will always run in the same environment that you have configured when you have built the image.