JupyterHub Configuration with OrbStack on Mac (all in Docker)

Prerequisites

You must have docker running on your computer

• On MacOS, OrbStack is recommanded

Installation Steps

1. Create Directory Structure

git clone https://forge.metabarcoding.org/MetabarcodingSchool/OBIJupyterHub.git

File Structure

Your ~/OBIJupyterHub directory should contain:

~/OBIJupyterHub/
├── Dockerfile                  # Image for students (already created)
├── Dockerfile.hub              # Image for JupyterHub (new)
├── jupyterhub_config.py        # Configuration
├── docker-compose.yml          # Orchestration
└── start-jupyterhub.sh         # Startup script

2. Start JupyterHub

./start-jupyterhub.sh

3. Access JupyterHub

Open your browser and go to: http://localhost:8888

You can log in with any username and password: metabar2025

Useful Commands

View JupyterHub logs

docker-compose logs -f jupyterhub

View all containers (hub + students)

docker ps | grep jupyterhub

Stop JupyterHub

docker-compose down

Restart JupyterHub (after config modification)

docker-compose restart jupyterhub

Rebuild after Dockerfile modification

# For student image
docker build -t jupyterhub-student:latest -f Dockerfile .
docker-compose restart jupyterhub

# For hub image
docker-compose up -d --build

View logs for a specific student

docker logs jupyter-<username>

Replace <username> by the actual user name of the student.

Clean up after lab

# Stop and remove all containers
docker-compose down

# Remove student containers
docker ps -a | grep jupyter- | awk '{print $1}' | xargs docker rm -f

# Remove volumes (WARNING: deletes student data)
docker volume ls | grep jupyterhub-user | awk '{print $2}' | xargs docker volume rm

# Clean everything (containers + volumes + network)
docker-compose down -v
docker ps -a | grep jupyter- | awk '{print $1}' | xargs docker rm -f
docker volume prune -f

Managing Shared Data

Directory Structure for Each Student

Each student will see this directory structure in their JupyterLab (everything under work/ is persistent):

work/                              # Personal workspace root (persistent)
├── [student files]               # Their own files and notebooks
├── R_packages/                   # Personal R packages (writable by student)
├── shared/                       # Shared workspace (read/write, shared with all)
└── course/                       # Course files (read-only, managed by admin)
    ├── R_packages/               # Shared R packages (read-only, installed by prof)
    ├── bin/                      # Shared executables (in PATH)
    └── [course materials]        # Your course files

R Package Priority: 1. R checks work/R_packages/ first (personal, writable) 2. Then work/course/R_packages/ (shared, read-only, installed by prof) 3. Then system libraries

Important: Everything is under work/, so all student files are automatically saved in their persistent volume.

User Accounts

Admin Account: - Username: admin - Password: admin2025 (change in docker-compose.yml: JUPYTERHUB_ADMIN_PASSWORD) - Can write to course/ directory

Student Accounts: - Username: any name - Password: metabar2025 (change in docker-compose.yml: JUPYTERHUB_PASSWORD) - Read-only access to course/ directory

Installing R Packages (Admin Only)

From your Mac (recommended):

chmod +x install-r-packages-admin.sh

# Install packages
./install-r-packages-admin.sh reshape2 plotly knitr

This script: - Installs packages in the course/R_packages/ directory - All students can use them (read-only) - No need to rebuild the image

From admin notebook:

Login as admin and create an R notebook:

# Install packages in course/R_packages (admin only, available to all students)
course_lib <- "/home/jovyan/work/course/R_packages"
dir.create(course_lib, recursive = TRUE, showWarnings = FALSE)

install.packages(c('reshape2', 'plotly', 'knitr'), 
                 lib = course_lib,
                 repos = 'http://cran.rstudio.com/')

Note: Admin account has write access to the course directory.

Students can also install their own packages:

Students can install packages in their personal work/R_packages/:

# Install in personal library (each student has their own)
install.packages(c('mypackage'))  # Will install in work/R_packages/

Using R Packages (Students)

Students simply load packages normally:

library(reshape2)  # R checks: 1) work/R_packages/ 2) work/course/R_packages/ 3) system
library(plotly)

R automatically searches in this order: 1. Personal packages: /home/jovyan/work/R_packages/ (R_LIBS_USER) 2. Prof packages: /home/jovyan/work/course/R_packages/ (R_LIBS_SITE) 3. System packages

List Available Packages

# List all available packages (personal + course + system)
installed.packages()[,"Package"]

# Check personal packages
list.files("/home/jovyan/work/R_packages")

# Check course packages (installed by prof)
list.files("/home/jovyan/work/course/R_packages")

Deposit Files for Course

To put files in the course/ directory (accessible read-only):

# Create a temporary directory
mkdir -p ~/jupyterhub-tp/course-files

# Copy your files into it
cp my_notebooks.ipynb ~/jupyterhub-tp/course-files/
cp my_data.csv ~/jupyterhub-tp/course-files/

# Copy into Docker volume
docker run --rm \
  -v jupyterhub-course:/target \
  -v ~/jupyterhub-tp/course-files:/source \
  alpine sh -c "cp -r /source/* /target/"

Access Shared Files Between Students

Students can collaborate via the shared/ directory:

# In a notebook, to read a shared file
import pandas as pd
df = pd.read_csv('/home/jovyan/work/shared/group_data.csv')

# To write a shared file
df.to_csv('/home/jovyan/work/shared/alice_results.csv')

Retrieve Student Work

# List user volumes
docker volume ls | grep jupyterhub-user

# Copy files from a specific student
docker run --rm \
  -v jupyterhub-user-alice:/source \
  -v ~/submissions:/target \
  alpine sh -c "cp -r /source/* /target/alice/"

# Copy all shared work
docker run --rm \
  -v jupyterhub-shared:/source \
  -v ~/submissions/shared:/target \
  alpine sh -c "cp -r /source/* /target/"

User Management

Option 1: Predefined User List

In jupyterhub_config.py, uncomment and modify:

c.Authenticator.allowed_users = {'student1', 'student2', 'student3'}

Option 2: Allow Everyone (for testing)

By default, the configuration allows any user:

c.Authenticator.allow_all = True

⚠️ Warning: DummyAuthenticator is ONLY for local testing!

Kernel Verification

Once logged in, create a new notebook and verify you have access to: - Python 3 (default kernel) - R (R kernel) - Bash (bash kernel)

Customization for Your Labs

Add Additional R Packages

Modify the Dockerfile (before USER ${NB_UID}):

RUN R -e "install.packages(c('your_package'), repos='http://cran.rstudio.com/')"

Then rebuild:

docker build -t jupyterhub-student:latest -f Dockerfile .
docker-compose restart jupyterhub

Add Python Packages

Add to the Dockerfile (before USER ${NB_UID}):

RUN pip install numpy pandas matplotlib seaborn

Distribute Files to Students

Create a files_lab/ directory and add to the Dockerfile:

COPY files_lab/ /home/${NB_USER}/lab/
RUN chown -R ${NB_UID}:${NB_GID} /home/${NB_USER}/lab

Change Port (if 8000 is occupied)

Modify in docker-compose.yml:

ports:
  - "8001:8000"  # Accessible on localhost:8001

Advantages of This Approach

✅ Everything in Docker: No need to install Python/JupyterHub on your computer
✅ Portable: Easy to deploy on another server
✅ Isolated: No pollution of your system environment
✅ Easy to Clean: A simple docker-compose down is enough
✅ Reproducible: Students will have exactly the same environment

Troubleshooting

Error "Cannot connect to Docker daemon": - Check that OrbStack is running - Verify the socket exists: ls -la /var/run/docker.sock

Student containers don't start: - Check logs: docker-compose logs jupyterhub - Verify student image exists: docker images | grep jupyterhub-student

Port 8000 already in use: - Change port in docker-compose.yml

After config modification, changes are not applied:

docker-compose restart jupyterhub

I want to start from scratch:

docker-compose down -v
docker rmi jupyterhub-hub jupyterhub-student
# Then rebuild everything
./start-jupyterhub.sh