Docker is a popular containerization tool used to provide software applications with a filesystem that contains everything they need to run. Using Docker containers ensures that the software will behave the same way, regardless of where it is deployed, because its run-time environment is ruthlessly consistent.

In this tutorial, we’ll provide a brief overview of the relationship between Docker images and Docker containers. Then, we’ll take a more detailed look at how to run, start, stop, and remove containers.

We can think of a Docker image as an inert template used to create Docker containers. Images typically start with a root filesystem and add filesystem changes and their corresponding execution parameters in ordered, read-only layers. Unlike a typical Linux distribution, a Docker image normally contains only the bare essentials necessary for running the application. The images do not have state and they do not change. Rather, they form the starting point for Docker containers.

Images come to life with the docker run command, which creates a container by adding a read-write layer on top of the image. This combination of read-only layers topped with a read-write layer is known as a union file system. When a change is made to an existing file in a running container, the file is copied out of the read-only space into the read-write layer, where the changes are applied. The version in the read-write layer hides the original file but doesn’t remove it. Changes in the read-write layer exist only within an individual container instance. When a container is deleted, any changes are lost unless steps are taken to preserve them.

Each time you use the docker run command, it creates a new container from the image you specify. This can be a source of confusion, so let’s take a look with some examples:

To restart an existing container, we’ll use the start command with the -a flag to attach to it and the -i flag to make it interactive, followed by either the container ID or name. Be sure to substitute the ID of your container in the command below:

docker start -ai 11cc47339ee1

We find ourselves at the container’s bash prompt once again and when we cat the file we previously created, it’s still there.

cat /tmp/Example1.txt

''Output''

''Example1 ''

We can exit the container now:

exit

This output shows that changes made inside the container persist through stopping and starting it. It’s only when the container is removed that the content is deleted. This example also illustrates that the changes were limited to the individual container. When we started a second container, it reflected the original state of the image.

We’ve created two containers, and we’ll conclude our brief tutorial by deleting them. The docker rm command, which works only on stopped containers, allows you to specify the name or the ID of one or more containers, so we can delete both with the following:

docker rm 11cc47339ee1 kickass_borg

Output

11cc47339ee1 kickass_borg

Both of the containers, and any changes we made inside them, are now gone.

We’ve taken a detailed look at the docker run command to see how it automatically creates a new container each time it is run. We’ve also seen how to locate a stopped container, start it, and connect to it. If you’d like to learn more about managing containers, you might be interested in the guide, Naming Docker Containers: 3 Tips for Beginners.