One very important use case of Docker is to run existing packaged dependencies for your application in a container. Your application might want to run its own MongoDB or Postgres cluster, or need to communicate with Apache Zookeeper. We can run these dependencies as their own separate Docker containers.
In the upcoming tutorials, we'll cover how to set up a container for your own application. We'll also show you how you can combine and orchestrate multiple containers including both your application containers and the pre-built dependency containers!
For this part of the tutorial, we'll be working on running Postgres with Docker. Postgres, more formally known as PostgreSQL, is a powerful, open-source object-relational database management system (ORDBMS) that is widely used for storing, retrieving, and managing structured data. It is highly regarded for its robustness, scalability, and ability to support complex queries, large datasets, and advanced data types, making it suitable for a wide range of applications, from small websites to large-scale enterprise systems.
Docker containers are made up of layers of images, which are the files of configuration data necessary to be installed for the container.
For example, a Docker container running Postgres might have these image layers:
postgres:10.10(application image layer)- [intermediate layers]
alpine:3.10(Linux base image layer)
Notice how the bottommost layer is the layer for the operating system, usually specifying a Linux image. Each successive layer will include more specific dependencies as we move up towards the application image layer.
You might hear the terms container and image being used frequently with respect to Docker. They, however, are not the same.
A container is a running environment for an image.
A container will contain its own:
- a virtual filesystem
- environment configurations
- application image
Let's demonstrate an example. Head over to the official Dockerhub website and search for Postgres. Then, go to the Docker official image for Postgres. Here, you can see information for Postgres, including different versions and configuration details.
Notice in the top right hand side, there is a command to easily download or pull the Docker image to your machine. If you would like, run this command
docker pull postgres
to pull the latest version of the Postgres image to your machine. You can even specify a specific version if you would like as such
docker pull postgres:<version number>.
You can then run your image by executing the command
docker run postgres
which will run the Postgres image on your machine. If you get an error in the terminal saying that you must specify POSTGRES_PASSWORD, you can run
docker run -e POSTGRES_PASSWORD=mypassword -d postgres
where you would replace mypassword with a password of your choice. Now, on a different terminal, run the following command:
docker ps
to list all current running Docker processes. You should see Postgres in the list. We will go over more use cases of this command, but keep it in mind!
Next, let's move on to creating your own Dockerfile to create am image for your own application!