100-days-mlops-kodekloud

Set Up and Configure Jupyter Notebook Server

Problem

A teammate has configured a JupyterLab server for the xFusionCorp Industries data science team, but the server does not behave correctly. Inspect the configuration, diagnose the issues, and start the server.

1. JupyterLab is already installed in the virtual environment at /root/code/ml-env/. The team’s configuration file is at /root/code/jupyter_lab_config.py and is visible in the file explorer.

2. When JupyterLab is started, the Jupyter UI button at the top of the lab must open the notebook interface.

3. For this to work, the running server must meet the following requirements:

   • it listens on port 8888;
   • it binds on 0.0.0.0 (the lab proxy cannot reach a server that is only bound on 127.0.0.1);
   • the notebook root directory is /root/notebooks/, and that directory exists on disk.

4. Open the configuration file, identify every setting that prevents the requirements above from being met, and correct it. Create any missing directories.

5. Start JupyterLab from the virtual environment using the corrected configuration:

   source /root/code/ml-env/bin/activate
   jupyter lab --config=/root/code/jupyter_lab_config.py --allow-root --no-browser &
   

Make sure JupyterLab is running before using the button at the top of the lab.

Solution

1. Update jupyter_lab_config.py:

    # Jupyter configuration file for the xFusionCorp Industries data science team
   
    # --- xFusionCorp team overrides (review before starting the server) ---
    c.ServerApp.token = ''
    c.ServerApp.password = ''
    c.ServerApp.disable_check_xsrf = True
    c.ServerApp.notebook_dir = '/root/notebooks/'
    c.ServerApp.port = 8888
    c.ServerApp.ip = '0.0.0.0'
   

   According to the requirements:

   • replaced ServerApp.port from 8000 to 8888
   • replaced ServerApp.ip from 1.1.1.1 to 0.0.0.0

2. Create the required directory if doesn’t exists

    mkdir -p /root/notebooks
   

3. Start JupyterLab with the given commands:

    source /root/code/ml-env/bin/activate
    jupyter lab --config=/root/code/jupyter_lab_config.py --allow-root --no-browser &
   

4. Open the Jupyter UI from top right option

This site is open source. Improve this page.