How to deploy Jupyter Notebook on Zenifra
Jupyter Notebook can be deployed on Zenifra as an HTTP project from the official image. This guide shows the recommended configuration, the required variables, and how to validate the result.
Prerequisites
- A verified account and access to the console.
- Permission to create projects in the organization (
project.create). - Credit or a saved card for the selected plan; see Payment models.
Quick option: template
Jupyter Notebook is in the public Templates catalog. Use Deploy to create the application with the recommended configuration, or Customize to review plan, storage, and variables before confirming. The steps below show a manual configuration from the official image.
Step 1: create the HTTP project
In the console, click + Create Project, choose HTTP, and fill in:
| Field | Value |
|---|---|
| Project source | OCI Image |
| Image | quay.io/jupyter/scipy-notebook:2025-03-14 |
| Registry | Public, no authentication |
| Plan | Basic is enough for testing and light notebooks. For heavy processing, choose a larger plan and check the console for GPU options. |
| Port | 8888 |
| Instances | 1. Jupyter does not distribute sessions across instances. |
| Persistent storage | Yes, 5 GB, directory /home/jovyan/work |
Pick a name between 6 and 32 characters, using lowercase letters, numbers, and hyphens; it also defines the *.clients.zenifra.com address. For testing, hourly billing is the cheapest; for continuous use, compare it with monthly.
Storage
Recommended if you want to keep notebooks and files across restarts. Save notebooks inside work.
Step 2: configure environment variables
Add the variables below in the Environment variables step:
JUPYTER_TOKEN=<long-random-token>The token works as the access password. Use a long random value: anyone with the token can run code in your project. See Environment variables.
Step 3: create and validate
Review the estimated cost and click Create Project. When it finishes, the console opens the project page with the public address.
Open the project URL, enter the token, and create a notebook inside the work folder. Run a simple cell such as import pandas; pandas.__version__.
If the application does not respond, check the project logs, the configured port, and the troubleshooting guide.
Production recommendations
- Restrict access by IP with access control when your plan allows it.
- Use hourly billing and stop the project when it is not in use.
- Keep database credentials in environment variables, never inside notebooks.
To remove the project, open it in Projects and click Delete. Deletion is permanent and erases storage.
Next steps
Last updated on