jupyterhub-ansiblespawner

Spawn JupyterHub single user notebook servers using Ansible

BSD-3-CLAUSE License

Downloads
61
Stars
4
Committers
3

AnsibleSpawner

Spawn JupyterHub single user notebook servers using Ansible.

This spawner runs Ansible playbooks to start, manage and stop JupyterHub singleuser servers. This means any Ansible module can be used to orchestrate your singleuser servers, including Docker and many public/private clouds, and other infrastructure platforms supported by the community. You can do things like create multiple storage volumes for each user, or provision additional services on other containers/VMs.

Prerequisites

Python 3.6 or above and JupyterHub 1.0.0 or above are required.

Installation

Configuration

Example jupyterhub_config.py spawner configuration.

ansible_path = "/path/to/"
c.JupyterHub.spawner_class = "ansible"
c.AnsibleSpawner.inventory = ansible_path + "inventory.yml.j2"
c.AnsibleSpawner.create_playbook = ansible_path + "create.yml"
c.AnsibleSpawner.update_playbook = ansible_path + "update.yml"
c.AnsibleSpawner.poll_playbook = ansible_path + "poll.yml"
c.AnsibleSpawner.destroy_playbook = ansible_path + "destroy.yml"
c.AnsibleSpawner.playbook_vars = {
    "container_image": "docker.io/jupyter/base-notebook",
    "ansible_python_interpreter": "python3",
}
c.AnsibleSpawner.start_timeout = 600
c.JupyterHub.hub_connect_ip = "10.0.0.1"

Examples

Example playbooks and configurations can be found under examples.

Development

Pytest is used to run automated tests that require Docker and Podman. These platforms were chosen because they are self-contained and can be installed in Travis, whereas testing with public cloud platforms requires secure access credentials.

If you only have one of these you can limit tests by specifying a marker. For example, to disable the Docker tests:

pytest -vs -m "not docker"

To view test coverage run pytest with --cov=ansiblespawner --cov-report=html, then open htmlcov/index.html.

setuptools-scm is used to manage versions. Just create a git tag.

Package Rankings
Top 32.4% on Pypi.org
Badges
Extracted from project README
PyPI version GitHub Workflow
Related Projects