From here to there, quickly and directly.
In this tutorial exercise, we will skip the steps of installing Git, Docker, docker-compose, and Portainer as they are covered in depth elsewhere already. Please see the links at the end of this README for more information.
Further, this codebase and documentation is meant to be used in conjunction with the video posted by University of Calgary
Executing the basic command without any parameters will display built-in help:
docker
Running an example container image:
docker run hello-world
Review local images:
docker image ls
Run explicitly tagged (version) of an image (default value is 'latest'):
docker run hello-world:latest
Look at images again and notice there is no new image - because it is the same one:
docker image ls
Help works with all commands and sub commands. Use tab-complete for command-completion in most shells:
docker container
List containers and non-running containers:
docker container ls
docker ps
docker ps -a
Use docker image ID or name to manage the image:
docker container inspect <container_name or ID>
Run a container with a specific name:
docker run --name test1 hello-world
List all containers (useful for finding names and status of the containers):
docker ps -a
Run and login to a container:
docker run --name test2 -it alpine sh
ls -la
pwd
exit
Review progress to this point:
docker image ls
docker container ls -a
docker network ls
Run a container with the same name again - it fails:
docker run --name test2 hello-world
Use a new name, and add the auto-remove flag to remove the container when it finishes running:
docker run --rm --name test3 hello-world
docker ps -a
Clean up:
docker container rm test2
docker container rm <from list>
docker ps -a
docker system prune
docker container ls -a
Full clean - do not do this on production systems:
docker system prune -af
docker image ls -a
Clone the project folder and files, and then review them:
ls -la
cat Dockerfile
cat app/app.py
cat requirements.txt
Build the image:
docker build -t myapp:latest .
Run the image as a container service with exposed port in interactive mode:
docker run --rm -p 8080:8080 myapp:latest
Browse the localhost by accessing http://localhost:8080/ - Use CTRL+C to stop the service when finished:
CTRL+C
Run the image as a container with a chosen name in detached/daemon mode:
docker run --rm -d -p 8080:8080 --name test1 myapp:latest
Verify that it is still running and attach to the logs for debugging:
docker ps
docker logs -f test1
USe CTRL+C to stop the log view:
CTRL+C
Log into the still running container to test and debug directly and then stop it:
docker exec -it test1 sh
ls -la
exit
docker stop test1
Run the image as a container in detached/daemon mode and bind a persistent volume:
docker run --rm -d -p 8080:8080 --name test1 -v ${PWD}/app_data:/var/app_data/ myapp:latest
View the web page (http://localhost:8080/). Login to the container to review the /var/app_data directory.
docker exec -it test1 sh
cd /var/app_data
ls -la
cat counter.txt
exit
Now, stop the container. Since we used the --rm flag, it will auto remove itself.
docker stop test1
docker ps -a
Re run the container, view the page (http://localhost:8080/), and ensure that you see the counter has indeed incremented further and not simply reset itself.
docker run --rm -d -p 8080:8080 --name test1 -v ${PWD}/app_data:/var/app_data/ myapp:latest
Log back into the container and change the value of the counter manually.
docker exec -it test1 sh
cd /var/app_data
echo '99999' > counter.txt
exit
View the web page (http://localhost:8080/) again to see that the value has indeed increased dramatically. Now open the project folder and look at app_data/counter.txt file and see that the value is indeed reflected there as well. This shows that what happened in the container persists on the host filesystem.
Do not forget to stop the container. Since we used the --rm flag, it will auto remove itself once stopped.
docker stop test1
Many times, we will want to create more than one service for a particular deployment, or deploy the same stack in multiple environments. Think in terms of a website that is comprised of a database server, a content engine, and a proxy.
We are now going to use the docker-compose.yml file to develop and test our application further because it simplifies and ensures consistancy of our working and deployment environments. The docker-compose.yml can also allow deployment to multiple hosts simultaneously on swarm enabled hosts.
Make sure that you have docker compose plugin installed on your host. Now, bring the container up using docker compose, like so:
docker compose up
This command essentially replicates the lengthy docker run --rm -d -p 8080:8080 --name test1 -v ${PWD}/app_data:/var/app_data/ myapp command by putting the configuration information into thr docker-compose.yml file in the same directory.
Browse your webpage (http://localhost:8080/) to see that the application is running and properly incrementing our counter.
Hitting CTL-C will disconnect you from the log, but it will not shut down the container. Shut down the container when we have finished using it.
docker compose down
Using the down subcommand will tell docker to remove our container, transient volumes, networks, etc. Leaving a clean slate for our next run.
We can also run the stack in detached (aka daemon or background) mode, by simply adding the -d to the end of the command like so:
docker compose up -d
To see see what containers are running,
docker ps
To connect and view the log file,
docker compose logs -f
You can disconnect by pressing CTL-C (on Windows and Linux - use COMMAND-C on Mac), and the stack will continue to run in the background, but the logging spew will cease.
You can connect to a specific container like so:
docker exec -it myapp sh
As with any Linux container or shell, you can exit it by simply typing exit and pressing the enter key.
Shut down the entire project when you are done.
docker compose down
If we want to add a persistant database to our project, we would merely add the service definition in our docker-compose.yml file just like we did for the myapp application. Once the new service is defined, the same commands can now be used to build and deploy both containers as a complete stack.
Networks, ports, and more can all be added through the docker-compose.yml file as you continue to develop your application stack.
Portainer is an amazing management and development tool that affords us easy, web-enabled access to all of our containers, images, networks, and so much more.
Use the documentation links below to install for your particular environment. For myself and this demo, I am using a Windows11 system with WSL2, so my installation command is the following:
docker run -d -p 8000:8000 -p 9443:9443 -p 9000:9000 --name portainer --restart=always -v /var/run/docker.sock:/var/run/docker.sock -v portainer_data:/data cr.portainer.io/portainer/portainer-ce:latest
Since this document is intended to be used with the video, and this tool is largely a visual system, I will not detail the usage here that is demonstrated in that video. That said, once this is deployed, you can look at http://localhost:9000/ to see an amazing new world of access to your containers and projects.
- Getting started wth Docker
- Getting started with docker-compose
- Install Portainer
- hello-world
- Python
- Alpine Linux
- Docker
- Docker Hub - home of many official images, and more
- Take 3 - free Business Edition Portainer
- Toolbox