Connecting to the CSUSB JupyterHub Server

Updated: 10-05-2022

• This instruction is for those who want to their local VS-Code to (remotely) use the CSUSB JupyterHub server. This is different from running a Jupyter stack that contains VS Code.

• This instruction is based on https://blog.jupyter.org/connect-to-a-jupyterhub-from-visual-studio-code-ed7ed3a31bcb.

Required items

We need to the followings.

1. (Server) A running container.

2. (Server) A valid token, see below.

3. (Local) VS-code with the Jupyter extension installed.

For item 1, see xxx, and for item 3, see xxx.

Below, we will explain how to generate a token and how to connect to the CSUSB JupyterHub using the token.

Generate a JupyterLab Token

1. Click on the File tab.

   

2. Select Hub Control Panel.

3. Click on Token.

   

4. You may assign any name for your token.

   

   Note

   Tokens are more or less your password. Anyone having your token can access your server. Do not share this information in public. If not sure, revoke it when done.

Connect to a remote server

1. Open your VS-Code.

2. Open an ipynb file.

3. Use !hostname to check your current kernel. Currently, it is local.

   

4. Open the palette and type Jupyter:; see how to here https://docs.github.com/en/codespaces/codespaces-reference/using-the-vs-code-command-palette-in-codespaces.

   

5. Choose Jupyter: Specify Jupyter Server for Connections.

6. Choose Existing

   

7. Type the following in the box https://csusb-jupyter.nrp-nautilus.io/user/<your email address>?token=<your token>.

   

   Note

   For the first type of the URI, you can copy it from CSUSB JupyterHub. In your server, click the URL box and copy the necessary part.

   

8. Lastly, run !hostname to check the current kernel. If it is still connected to the last one, see the following images.

   

Click on the current kernel, in the photo it is main (Python 3.9.13) located at the upper right corner.

Now, choose the (Remote) Jupyter Kernel

We are good to go.