Introduction

Objective

• Tried teaching Docker to data scientists
• Had to back up every few minutes to explain underlying concepts
  • “What does it mean to ‘mount’ a filesystem?”
  • “What does it mean to ‘background’ a job?”
  • “What is a ‘port’?”
• Answer those questions step-by-step so that a web service running in Docker will make sense

What This Is

• Notes and working examples that instructors can use to perform a lesson
  • Do not expect novices with no prior Unix experience to be able to learn from them on their own
• Musical analogy
  • This is the chord changes and melody
  • We expect instructors to create an arrangement and/or improvise while delivering
• Please see the license for terms of use, the Code of Conduct for community standards, and these guidelines for notes on contributing

Scope

• Intended audience
  • Ning did a bachelor’s degree in economics and now works as a data analyst for the Ministry of Health
  • They are comfortable working with Unix command-line tools, writing data analysis programs in Python, and downloading data from the web manually
  • Ning wants to understand what happens when they install a package or run a pipeline in the cloud
  • Their work schedule is unpredictable and highly variable, so they need to be able to learn a bit at a time

Prerequisites

• Unix shell commands covered in this Software Carpentry lesson:
  • pwd; ls; cd; . and ..; rm and rmdir; mkdir; touch; mv; cp; tree; cat; wc; head; tail; less; cut; echo; history; find; grep; zip; man
  • current working directory; absolute and relative paths; naming files; editing with nano
  • standard input; standard output; standard error; redirection; pipes
  • * and ? wildcards; shell variable with $ expansion; for loop
• Python for command-line scripting
  • variables; numbers and strings; lists; dictionaries; for and while loops; if/else; with; defining and calling functions; sys.argv, sys.stdin, and sys.stdout; simple regular expressions; reading JSON data; reading CSV files using Pandas or Polars
  • pip install
  • python -m venv or conda create

Learning Outcomes

1. Explain the difference between shell variables and environment variables and write shell scripts that use each.
2. Create a virtual environment and explain what this actually does.
3. Create requirements.txt file for pip and explain version pinning.
4. Explain what a filesystem is (disk partitions, inodes, symbolic links) and use df, ln, similar commands to explore with them.
5. Explain what a process is and use commands like ps and kill to explore and manage them.
6. Explain what a job is and use commands like jobs, bg, and fg to manage them.
7. Explain what cron jobs are and how to create them.
8. Explain the difference between a container and a virtual machine.
9. Create and manage Docker images.
10. Explain what ports are and write Python code that uses sockets and HTTP.
11. Explain what certificates are and how they are used to support HTTPS.
12. Explain what key pairs are and how they are stored, and create and manage key pairs.
13. Explain what IP addresses are and how they are resolved.
14. Explain how traditional password authentication works and describe its weaknesses.

Setup

• Download the latest release
• Unzip the file in a temporary directory to create:
  • ./site/*.*: files and directories used in examples
  • ./src/*.*: shell scripts and Python programs
  • ./out/*.*: expected output for examples

Acknowledgments

My thanks to everyone who helped make this tutorial possible:

• Stefan Arentz
• Julia Evans
• Robert Kern
• Matt Panaro
• Jean-Marc Saffroy