Assign profiles to whitelisted users - jupyter-notebook

Question

I have been exploring the profile list feature of the kubespawner, and am presented with a list of available notebooks when I login. All good. Now I have the use case of User A logging in and seeing notebooks 1 and 2, with User B seeing notebooks 2 and 3.
Is it possible to assign certain profiles to specific users?

Answer 1

I do not think Jupyterhub enables you do to that based on this https://zero-to-jupyterhub.readthedocs.io/en/latest/user-environment.html
I think a way to achieve this would be having multiple jupyterhub instances configured with different list of notebook images. Based on something like AD group, you redirect your user to required instance so they get specific image options.

Answer 2

You can dynamically configure the profile_list to provide users with different image profiles.
Here's a quick example:
#gen.coroutine
def get_profile_list(spawner):
"""get_profile_list is a hook function that is called before the spawner is started.
Args:
spawner (_type_): jupyterhub.spawner.Spawner instance
Yields:
list: list of profiles
"""
# gets the user's name from the auth_state
auth_state = yield spawner.user.get_auth_state()
if spawner.user.name:
# gets the user's profile list from the API
api_url = # Make a request
data_json = requests.get(url=api_url, verify=False).json()
data_json_str = str(data_json)
data_json_str = data_json_str.replace("'", '"')
data_json_str = data_json_str.replace("True", "true")
data_python = json.loads(data_json_str)
return data_python
return [] # return default profile
c.KubeSpawner.profile_list = get_profile_list
And you can have your API return some kind of configuration similar to this:
[
{
"display_name": "Google Cloud in us-central1",
"description": "Compute paid for by funder A, closest to dataset X",
"spawner_override": {
"kubernetes_context": "<kubernetes-context-name-for-this-cluster">,
"ingress_public_url": "http://<ingress-public-ip-for-this-cluster>"
}
},
{
"display_name": "AWS on us-east-1",
"description": "Compute paid for by funder B, closest to dataset Y",
"spawner_override": {
"kubernetes_context": "<kubernetes-context-name-for-this-cluster">,
"ingress_public_url": "http://<ingress-public-ip-for-this-cluster>",
"patches": {
"01-memory": """
kind: Pod
metadata:
name: {{key}}
spec:
containers:
- name: notebook
resources:
requests:
memory: 16Mi
""",
}
}
},
]
Credit: https://multicluster-kubespawner.readthedocs.io/en/latest/profiles.html