This lab requires Docker to be installed on the machine running it. Instructions for installing Docker on various platforms can be found on their website:
Docker Desktop is not needed so long as the command line tools are installed. However, it may be useful to have a GUI to Docker for managing containers and images.
Once Docker is installed, the lab image can either be downloaded from Docker Hub or built from the repository's files. See the below section for more information.
It is important to note that Linux users will have to run Docker commands as root or using the sudo command!
A pre-built image is availabe on Docker Hub if you wish to skip the building process. This image is automatically built, so it stays up to date with the repository. A set of directions is provided by Docker Hub to pull and run the image.
Building the image requires Docker to be installed and running on the user's computer. These instructions are applicable to both Windows and Linux operating systems. To assist with building, "docker_setup.ps1" (Windows) and "docker_setup.sh" (Linux) have be provided to automatically build and run the lab container.
Lab 4 utilizes a Jupyter notebook that is accessible from a web browser. Upon running the Docker container for the first time, a URL associated with a port running on the localhost will be provided. Following this will be the string of text "/tree?token=". The characters that follow this make up the token that is used to login to the container's webpage. If you plan on accessing the container in the future, it is recommended to save this token.
Once connected to the notebook's webpage, select "poisoning_lab.ipynb" and then click the "Open" button. This will load the lab, which can now be ran from the web browser.
Note: When running the lab, a warning related to TensorFlow may be displayed. However, these can be ignored and shouldn't affect the functionality of the lab.
Accessing Docker can vary depending on the IDE and operating system you use. Visual Studio Code has a number of plugins that make accessing a Docker container fairly straightforward.
Without any plugins or GUI, accessing an already existing container requires first starting it in the background, then attaching a terminal session to it.
$ docker start lab4
$ docker attach lab4
Running a new container under the name "lab4" requires deletion of the old one. This can be done by simply deleting the container using Docker's command line tools.
$ docker container rm lab4
After this, the docker run command provided as part of the lab's scripting can be ran again to generate a fresh
container instance.