Tutorial 5

More Docker Practice

An RStudio Container

Note

If you are on an ARM based M-chip mac, you need to pass in the --platform=linux/amd64 flag to the docker run command. https://ubc-dsci.github.io/reproducible-and-trustworthy-workflows-for-data-science/lectures/110-containerization-2.html#specifying-the-image-architectureplatform

  1. Run the RStudio container and connect it to your local computer’s port 8787
docker run --rm -e PASSWORD="pizza" -p 8787:8787 rocker/rstudio:4.4.2

  1. This should be your first time using the rocker/rsudio Docker image, and so it will need to fetch the image from the internet first. Think about how that process compares to using renv, is it faster or slower? What sort of things does Docker need to fetch that renv does not?

  2. Explain in your own words what each of the flags are doing in the docker run command. You can use the official docker run documentation: https://docs.docker.com/reference/cli/docker/container/run/

  3. Open and log into the RStudio container and try to run some R code in a script

  4. Save this script

  5. Close the container and run it again

  6. Notice how all your work inside the container is lost

An RStudio Tidyverse Container

  1. Do the same process as before, but this time use the rocker/tidyverse:4.4.2 container.
  2. Without turning off this container, open a new terminal and run a separate rocker/rstudio:4.4.2 container
  • You will need to change the local port mapping
  1. Make sure you are able to log into and run both RStudio sessions in the container
  • Note how only one of the RStudio sessions has tidyverse installed and the other doesn’t

Integrating Docker into a repository

  1. Create a repository and clone it down

  2. Create a bash script that lists (using a bash command) the files of your current working directory, the root of the repo. Name it list.sh

  3. Create a dockerfile that uses the rocker/rstudio container, using version 4.2.2.

  4. Copy the list.sh into the container

  5. Run the script in the container

  6. Document how to run the container and run the script in the container. We must follow the steps until we are able to run and see the output. Put your steps in the README.md file

  7. Create a github action that also runs this bash script