Tutorial: Setting Up a Collaborative Jupyter Lab

This tutorial outlines the steps to set up a collaborative Jupyter Lab environment for multiple users. Make sure to follow each step carefully.

A. Initial Preparation

1. Create a Tmux Session:

   • Open your terminal and run the command: tmux new -t your_session_name (replace "your_session_name" with your desired name).

   • What is Tmux?

     • It keeps your terminal session active even if the SSH connection is lost.

     • It allows you to access the terminal session from different locations.

   • Important Tmux commands:

     • To exit the session: Press Ctrl+b then d.

     • To re-enter the session: tmux a -t your_session_name.

   • For more information on tmux, visit: https://tmuxcheatsheet.com/

2. Create a Conda Environment:

   • Ask all users for the required Python version.

   • Create a conda environment with the command: conda create -n your_env_name python=python_version (replace "your_env_name" and "python_version" accordingly).

     • Example: conda create -n collaboration python=3.9

   • Activate the environment: conda activate your_env_name

3. Create a Working Directory:

   • Create a new directory to store your collaborative project. You can use the initials of each team member or person (e.g., NF, MH, RDWP).

   • Command: mkdir your_directory_name

   • Enter the directory: cd your_directory_name

B. Installing and Configuring Jupyter Lab

4. Install Jupyter Lab and Jupyter-Collaboration:

   • Ensure that the conda environment you created earlier is active.

   • Install Jupyter Lab and Jupyter-Collaboration: pip install jupyter jupyter-collaboration

   • Alternatively: Install them when creating the conda environment by adding jupyter jupyter-collaboration to the end of the conda create command.

5. Run Collaborative Jupyter Lab:

   • Run the following command in the terminal: jupyter lab --collaborative --allow-root --no-browser --ip=0.0.0.0 --port your_desired_port

   • Replace "your_desired_port" with the port number you want to use.

6. Save the Jupyter Token:

   • After running the command above, a link containing a token will appear.

   • IMPORTANT: Save this token securely. Other people can access your Jupyter Lab if they know this token.

C. Tunnel Configuration and Jupyter Lab Access

7. Provide Jupyter Lab Details:

   • Inform the tunnel administrator (in this case, Naufal Faza) of the following information:

     • The name of the PC where Jupyter Lab is running.

     • The port number used.

     • The requested subdomain (if any).

     • Example: uas-ai.digilabdte.com, the subdomain is uas-ai.

8. Access Jupyter Lab:

   • After the tunnel administrator has completed the configuration, you will be given a link to access Jupyter Lab.

   • The link will look like this: your_requested_subdomain.digilabdte.com/lab?token=your_token_value.

   • Don't forget to enter the token you obtained in step 6.

9. Start Collaborating:

   • Inside Jupyter Lab, you can access the terminal and monitor resource usage with nvtop.

   • Make sure to coordinate with your team via group chat to avoid conflicts when running programs. Remember that Jupyter Lab is like Google Docs, so collaboration is real-time.