With the base image defined, we need to run various commands to configure our image. There are many commands to help with this, the main commands two are COPY and RUN.

RUN <command> allows you to execute any command as you would at a command prompt, for example installing different application packages or running a build command. The results of the RUN are persisted to the image so it's important not to leave any unnecessary or temporary files on the disk as these will be included in the image.

COPY <src> <dest> allows you to copy files from the directory containing the Dockerfile to the container's image. This is extremely useful for source code and assets that you want to be deployed inside your container.

Task

A new index.html file has been created for you which we want to serve from our container. On the next line after the FROM command, use the COPY command to copy index.html into a directory called /usr/share/nginx/html

Protip

If you're copying a file into a directory then you need to specify the filename as part of the destination.

With our files copied into our image and any dependencies downloaded, you need to define which port application needs to be accessible on.

Using the EXPOSE <port> command you tell Docker which ports should be open and can be bound too. You can define multiple ports on the single command, for example, EXPOSE 80 433 or EXPOSE 7000-8000

Task

We want our web server to be accessible via port 80, add the relevant EXPOSE line to the Dockerfile.

With the Docker image configured and having defined which ports we want accessible, we now need to define the command that launches the application.

The CMD line in a Dockerfile defines the default command to run when a container is launched. If the command requires arguments then you need to use an array, for example ["cmd", "-a", "arga value", "-b", "argb-value"], which will be combined together and the command cmd -a arga value -b argb-value would be run.

Task

The command to run NGINX is nginx -g daemon off;. Set this as the default command in the Dockerfile.

Protip

An alternative approach to CMD is ENTRYPOINT. While a CMD can be overridden when the container starts, a ENTRYPOINT defines a command which can have arguments passed to it when the container launches. This makes launching a container behave more launching a normal process.

In this example, NGINX would be the entrypoint with -g daemon off; the default command.

After writing your Dockerfile you need to use docker build to turn it into an image. The build command takes in a directory containing the Dockerfile, executes the steps and stores the image in your local Docker Engine. If one fails because of an error then the build stops.

Task

Using the docker build command to build the image. You can give the image a friendly name by using the -t <name> option.

Protip

You can use docker images to see a list of the images on your local machine.

With the image successfully created, you can now launch the container in the same way we described in the first scenario.

Task

Launch an instance of your newly built image using either the ID result from the build command or the friendly name you assigned it.

NGINX is designed to run as a background service so you should include the option -d. To make the web server accessible, bind it to port 80 using p 80:80

For example:

docker run -d -p 80:80 <image-id|friendly-tag-name>

You can access the launched web server via the hostname docker. After launching the container, the command curl -i http://docker will return our index file via NGINX and the image we built.

Protip

You can check the container is running using docker ps