I have an old IBM power8 server that I'm trying to get JupyterHub going on. There are no prebuilt Docker images, so I have to make my own.

The following builds and runs; however, when I go to enter my username and password, it never works. No information in the logs besides notifications of failed attempts.

Here are some quick instructions if you want to try it out yourself:

wget .zip
unzip 45219232bc4681038942763e825a071e43df77a9.zip -d ./
cd jhub_powerpc-45219232bc4681038942763e825a071e43df77a9/
openssl req -x509 -newkey rsa:4096 -keyout key.pem -out cert.pem -days 365 -nodes
docker build -t my-jupyter-ppc64le .
docker run -d -p 443:443 --name jhub my-jupyter-ppc64le

Here's the full Dockerfile below. The user taylor is a user on the server outside of the Docker container.

# Use a base image that supports ppc64le architecture
FROM ubuntu:20.04

# Set timezone
ENV TZ=America/New_York
RUN ln -fs /usr/share/zoneinfo/$TZ /etc/localtime && \
    echo $TZ > /etc/timezone && \
    DEBIAN_FRONTEND=noninteractive apt-get update && apt-get install -y tzdata

# Install system dependencies
RUN apt-get update && apt-get install -y \
    wget \
    bzip2 \
    curl \
    ca-certificates \
    git \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# Install NVM
ENV NVM_DIR=/root/.nvm
RUN curl -fsSL .39.4/install.sh | bash

# Install Node.js and npm using NVM
RUN bash -c "source $NVM_DIR/nvm.sh && nvm install 18 && nvm use 18 && nvm alias default 18"

# Set Node.js and npm in the PATH
ENV PATH="/root/.nvm/versions/node/v18.20.7/bin:$PATH"

# Verify Node.js and npm installation
RUN bash -c "source $NVM_DIR/nvm.sh && node -v && npm -v"

# Install yarn and configurable-http-proxy globally
RUN bash -c "source $NVM_DIR/nvm.sh && npm install -g yarn configurable-http-proxy"

# Ensure configurable-http-proxy is available globally
RUN echo "export PATH=\$PATH:/root/.nvm/versions/node/v18.20.7/bin" >> /root/.bashrc

# Download and install Miniconda
RUN wget -q .sh && \
    chmod +x Miniconda3-latest-Linux-ppc64le.sh && \
    ./Miniconda3-latest-Linux-ppc64le.sh -b && \
    rm Miniconda3-latest-Linux-ppc64le.sh

# Add Miniconda to PATH
ENV PATH="/root/miniconda3/bin:$PATH"

# Ensure conda is available
RUN conda --version

# Install mamba using conda
RUN conda install mamba -c conda-fe -y

# Install JupyterHub and dependencies using mamba
RUN mamba install --yes jupyterhub-singleuser jupyterlab nbclassic "notebook>=7.2.2" jupyterhub[localauth]

# Create SSL directory
RUN mkdir -p /etc/jupyterhub/ssl

# Copy SSL certificate and key (Make sure you have these files before building)
COPY cert.pem /etc/jupyterhub/ssl/cert.pem
COPY key.pem /etc/jupyterhub/ssl/key.pem

# Set correct permissions for SSL files
RUN chmod 600 /etc/jupyterhub/ssl/cert.pem /etc/jupyterhub/ssl/key.pem

# Create JupyterHub config directory
RUN mkdir -p /etc/jupyterhub

# Generate JupyterHub server configuration with LocalAuthenticator
# NB: change "taylor" to your system username you're using outside of Docker
RUN echo "c.Authenticator.admin_users = {'taylor'}" > /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.JupyterHub.authenticator_class = 'jupyterhub.auth.LocalAuthenticator'" >> /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.LocalAuthenticator.create_system_users = True" >> /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.JupyterHub.ip = '0.0.0.0'" >> /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.JupyterHub.port = 443" >> /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.JupyterHub.ssl_cert = '/etc/jupyterhub/ssl/cert.pem'" >> /etc/jupyterhub/jupyterhub_config.py && \
    echo "c.JupyterHub.ssl_key = '/etc/jupyterhub/ssl/key.pem'" >> /etc/jupyterhub/jupyterhub_config.py

# Expose SSL port
EXPOSE 443

# Ensure that the configurable-http-proxy path is set during JupyterHub startup
ENV PATH="/root/.nvm/versions/node/v18.20.7/bin:$PATH"

# Default command to start JupyterHub with SSL
CMD ["bash", "-c", "source $NVM_DIR/nvm.sh && exec npx configurable-http-proxy & exec jupyterhub --config /etc/jupyterhub/jupyterhub_config.py"]