/workshop-docker-enei-2016

This repository holds the code related with the Hands-on Docker Workshop in ENEI@UA 2016

Apache License 2.0Apache-2.0

Hands-on Docker Workshop ENEI 2016

This repository holds the code related with the Hands-on Docker Workshop in ENEI@UA 2016

The following steps were taken from the official Self-paced Docker Tutorials and Labs and the official Self-paced Docker Training.

Table of contents

  1. Introduction
  2. First steps
  3. Containers
  4. Networking
  5. Dockerfile
  6. Docker Compose
  7. Final exercise

Introduction

This hands-on guide was written to offer you a quick and simple way to learn Docker. Please follow every step carefully and if you have any doubt please call for mentor’s help.

First things first! You will follow this Setup to install Docker in your machine.

First Steps

From this step and beyond you’ll work with docker client. First let’s use the Hello World example.

$ docker run hello-world

Check Docker Client and Server version

$ docker version

Docker Hub

Docker Hub is a cloud hosted service for building and shipping application or service containers. It provides a centralized resource for container image discovery, distribution and change management, user and team collaboration, and workflow automation throughout the development pipeline.

Open the web browser and access to Docker Hub.

Docker Hub Search Header

Exercise

Locate the search bar (image above) and search for the tomcat repository. Try to get to know Docker Hub a little better.

Local Docker Images

Images are kept on the docker daemon host machine. To list what docker images are present:

$ docker images

Exercise

Use the command docker search to find all the images related with tomcat.

Containers

Let's create a container with the ubuntu:14.04 image that will run the echo command:

$ docker run ubuntu:14.04 echo "hello world"

Now you can test some commands by yourself:

## list container processes
$ docker run ubuntu:14.04 ps ax

Note that the 2nd execution was faster, because you already had a local ubuntu 14.04 image.

Flags

  • Flag -i tells docker to connect to STDIN on the container
  • Flag -t specifies to get a pseudo-terminal
  • Flag -d tells docker to detache from the container (i.e. background)

Let's test -i and -t flags with a terminal process e.g. bash

$ docker run -i -t ubuntu:14.04 /bin/bash

You're now inside the container. Execute the following commands that will add a new user admin and install vim.

# Add a user
$ adduser admin
# Add admin user to sudo group
$ adduser admin sudo
# Change to user admin
$ su admin
# Install vim
$ sudo apt-get install vim
# Test vim (to quit vim :q!)
$ vim
# After testing, exit the container. Why do you have to run the exit command twice?
$ exit
$ exit

Re-run the last docker run command and test the following commands.

# Try to change to admin user
$ su admin
# User doesn't exist, why?
# Try to start vim
$ vim
# vim not installed, but why?

Now run a container with an ubuntu:14.04 image with the initial process /bin/bash

$ docker run -it ubuntu:14.04 /bin/bash
  $ ps -ef
  # Note that your bash process is PID 1. Once this process is finished container will stop.
  # Hint: To quit without killing the container press CTRL+P+Q
$ docker ps

You'll see that the container is still running. CTRL+P+Q didn't kill the container, instead it just detached from the container.

Listing your active containers

$ docker ps

Use flag -a to list all containers (includes containers that are stopped)

$ docker ps -a

Take note on the STATUS column. See the difference? Stopped containers have a message like Exited () ....

Let's test detached -d flag, by executing a container that runs ping for 100 times.

$ docker run -d ubuntu:14.04 ping 127.0.0.1 -c 100

You'll get a ID, which is the container Long ID. Check the container output with docker logs command. You'll only need the first 6 digit from the container ID from the previous command output.

$ docker logs <ID>

You can simplify by giving a name the container. For that just use the --name flag.

$ docker run -d --name test-ping ubuntu:14:04 ping 127.0.0.1 -c 100

Follow the log if you use the -f flag (CTRL+C to exit)

$ docker logs -f test-ping

Go to https://docs.docker.com/engine/reference/run/ and learn more about docker run.

Exercise

Fiddle around with docker run, docker stop, docker start, docker ps and docker logs.

Networking

In this part we'll use the tomcat:7 image.

Run a tomcat container with -P flag (it will map all exposed ports on the containers to the host)

$ docker run -d --name tomcat -P tomcat:7

Check what port is mapped to container 8080 port with docker ps, take a look at the PORTS column.

$ docker ps

Open your browser and go to http://VM_IP:PORT Stop the container

$ docker stop tomcat

Remove the container

$ docker rm tomcat

Let's clean up, by removing all stopped containers.

$ docker rm `docker ps -aq`

Linking containers

Re-execute the tomcat container.

$ docker run -d --name tomcat -P tomcat:7

Run a new container and try to ping tomcat container.

$ docker run -it ubuntu:14.04 /bin/bash
    $ ping tomcat
    # It fails! Exit the container.
    $ exit

The --link flag allows to connect two containers. Its syntax is <name or id>:<alias> .

$ docker run -it --link tomcat:tomcat ubuntu:14.04 /bin/bash
    $ ping tomcat
    # Success!!

Dockerfile

Dockerfile is a text document that contains instructions (commands) that will assemble an image. Using docker build an automated build is launched that will execute several command-line instructions in succession.

If you're on a Dockerfile folder, you just need to execute

$ docker build -t <name>:<tag> .

Using -t flag the default image tag is latest and -f flag default instruction file is Dockerfile (you can ommit -f). The . means that docker build will use the current folder as context folder (if you copy files it will refer to the context path). You can use -t flag multiple times to tag the image to different repositories.

The following table has the main Dockerfile instructions. A valid Dockerfile will have a FROM instruction as its first instruction (MAINTAINER instruction is not counted).

Instruction Description
FROM Sets the base image for subsequent instructions.
RUN Execute any commands in a new layer on top of current base image.
USER Sets the user name or UID to use when running the image (including execution instructions from the same Dockerfile).
ADD Copies new files, directories or remote URLs and adds to the filesystem of the container.
COPY Copies new files or directories and adds to the filesystem of the container.
CMD There can only be one CMD instruction per Dockerfile. CMD provides defaults for executing container.
ENTRYPOINT Allows to configure a container that will run as an executable.
ONBUILD Adds a trigger instruction to the image to be executed at a later time, when the image is used as the base for another build.

The following example provides an ubuntu:14.04 image that will create an admin user, add it to the sudo group and set it as the default user. It also adds the src/binary file to the root filesystem and sets it as the image initial execution command.

FROM ubuntu:14.04
MAINTAINER Tiago Pires <tiago.pires@mindera.com>

RUN adduser --system --shell /bin/bash --group --disabled-password admin && \
    adduser admin sudo

USER admin

ADD src/binary /binary

CMD ["/binary"]

Go to https://docs.docker.com/engine/reference/builder/ and learn more about Dockerfile instructions.

Exercise

Create a Dockerfile from a custom image (you decide which image) that will create an user and set it as default. Initial command should be sh or bash shell.

Docker Compose

Docker Compose is a YAML file that helps to define services, networks and volumes. The default file is docker-compose.yml. Each container has its own service configuration, much like command-line parameters to docker run.

Docker Compose isn't part of Docker Engine, so installation is required.

Run sudo -i then the two commands below:

$ curl -L https://github.com/docker/compose/releases/download/1.8.1/docker-compose-`uname -s`-`uname -m` > /usr/local/bin/docker-compose
$ chmod +x /usr/local/bin/docker-compose
# Compose is now installed!

In the previous Dockerfile example you added the src/binary file to the image. In result, each time you recompile the file you also will need to re-build the image.

The following docker-compose.yml example uses that same Dockerfile, but instead of copying the file it will mount the src/binary on /binary. That means that you can make changes to the file without having to re-build the image.

version: '2'

services:
  myimage:
    build: .
    volumes:
      - ./src/binary:/binary
    command: ./binary

Go to https://docs.docker.com/compose/compose-file/ and learn more about Docker Compose.

Exercise

Transform the previous Dockerfile exercise to a docker-compose.yml and instead of copying the binary, share it as a volume.

Final Exercise

So now that your Docker knowledge is up to date, you can challenge your self by doing a final exercise.

You'll be guided through the creation and customization of a voting app. Click here to continue.