This tutorial covers the basics of Docker, a container that wraps code in a complete filesystem that contains code, runtime, system tools, and system libraries. Docker containers ensure that code will function the same, regardless of the environment -- in essence, a code incubator.
This tool is a hot product on the market right now.
Let me explain some terms used in Docker:
These terms are the most used terms in the Docker world, and they are the main terms that I will use in this tutorial.
I will also cover the creation and packaging of a Java application that runs on Tomcat Server.
Docker only runs above Linux but you can use it with Mac OS or Windows OS using Kitematic. Kitematic creates VirtualBox Machines that let you run Docker on them.
You can install Kitematic for Mac OS and Windows from here https://kitematic.com/ and for Linux you can type this command:
1curl -sSL https://get.docker.com/ | sh
If you are using Mac or Windows, I suggest that you not use the Kitematic interface early on. It will be better if you start creating and manipulating everything from the command line. In this way you will learn more.
You start by creating your Dockerfile that describes the steps in order to create the base image. Here is an example (comments on each line describe the code):
1# All Dockerfiles need to start from a base Linux image. 2# Docker has a hub (http://hub.docker.com) where you can see all Images submitted by users and 3# where you can submit your own images. This is like a github for Docker images. 4# We will start from java:8-jre. This is an ubuntu machine with java 8 installed on it. 5# The keyword here is FROM. 6FROM java:8-jre 7 8# The KEYWORD ENV it let us specify some Linux environment variables 9# Here we will set CATALINA_HOME (the home of the tomcat server) 10ENV CATALINA_HOME /usr/local/tomcat 11ENV PATH $CATALINA_HOME/bin:$PATH 12 13 14# RUN lets you run Linux commands inside the image. 15# The \ symbol allows you to continue the command onto the next line 16# The && allows you to string multiple commands together. This is preferred over having a RUN statement for each command 17# Here we will navigate into /usr/local then download tomcat from the server 18# then decompress it and rename it to /tomcat. 19# The final path to tomcat will be /usr/local/tomcat. 20RUN cd /usr/local/ \ 21 && wget http://mirrors.m247.ro/apache/tomcat/tomcat-8/v8.0.48/bin/apache-tomcat-8.0.48.tar.gz \ 22 && tar xzf apache-tomcat-8.0.48.tar.gz \ 23 && mv apache-tomcat-8.0.48/ tomcat/ 24 25 26# the EXPOSE command will tell your future container to expose 8080 port to the outside 27# 8080 is the default port for tomcat. 28EXPOSE 8080 29 30# CMD sets the first command that will run when you create containers from the resulting image. 31# Here, we will start tomcat once the container is called 32CMD ["/usr/local/tomcat/bin/catalina.sh", "run"]
After you create this Dockerfile you need to create your image. Docker will run an intermediate container for each step in your Dockerfile during the build process, eventually creating a final image that you can export and import to other systems. You need to navigate through your directory to where the Dockerfile was saved. Use this command line prompt:
1docker build -t tomcat .
A couple quick definitions:
docker build - This command builds a new image from the Dockerfile code at PATH. In our case, the PATH is '.' (current folder). The option -t refers to a Name of (or tag of) the image (format name:tag). In this case, just the name "tomcat".
After all steps have been executed, you can test your code's functionality by typing
docker images and you will see your image there named
tomcat after the tomcat image we just built:
1$ docker images 2REPOSITORY TAG IMAGE ID CREATED SIZE 3tomcat latest 24b60eaf8397 7 seconds ago 334 MB 4java 8-jre e44d62cf8862 8 months ago 311 MB
You can now start a container from an image by running this:
1docker run -p 8080:8080 --name=tomcat tomcat
Let's break down this simple, yet complex command.
docker run - Runs a command in a new container.
-p parameter - Establishes a connection between container's port(s) to the host. By using port1:port2 you make port1 match with port2. After that, you can use port2 in order to access the service that is opened on port1 inside the container. Thus, the container is now connected with outside environments. Usually applications that use some ports in order to run (like tomcat on port 8080, mySQL on port 3306, Apache on port 80, and so) will be restricted in a container to that specific container, unable to communicate with the host or outside network. That means if you start a mySQL container using port 3306, you wouldn't be able to access it with another application outside that container. Using the
-p 8080:8080 option allows communication between the container on the container's port 8080 and the host's port 8080.
--name= - The assigned name for the container. This will be what you reference when you run other docker commands against the container. Here we are naming the container
tomcat - The image the container is based on. In this example, we are specifying running the Tomcat image we just created as a container.
After you run a container you can check that it exists by typing the following sequence:
This command will show you all running containers. If this commands returns nothing at all, then your container creation failed.
1docker ps -a
The above command displays all containers (running or not).
After you started a container you can stop it by using docker stop [name]. You can restart the container by typing docker start [name].
Another important command is docker exec, which lets you run a command inside your Docker running container. Note that this only works in running containers, not stopped ones.
For a complete list of commands you can go to Docker references. I will recommend that you read up on many of those commands if you find Docker containers interesting.
I hope that this brief tutorial on Dockers has helped you!
If you have any questions, please comment below or find me on Twitter, at @grvmariobyn.
Thank you. Marius.