I can't run ray on kubeflow

How severe does this issue affect your experience of using Ray?

  • Low: It annoys or frustrates me for a moment.

i am running ray on kubeflow,i have follow the instruction of this doc.
https://docs.ray.io/en/latest/cluster/kubernetes/k8s-ecosystem/kubeflow.html

i follow all the step besides install kubeflow v1.9

here when i run the final code

i get error ConnectionError: ray client connection timeout

it seeme like i can’t connect to raycluster which install on my original machine.

every thins is fine, i figure out what’s going on. using vpn and set no_proxy fix my problem.

by the way,using ray in kubeflow, is that i am using my original resources like cpu,memory,gpu but code it in a few notebook? is there any using cases that suit this kind of situation? i think it might be easier that using it just real environment not a container because when i using kubeflow and bulid a notebook it is hard to do apt install to install some tools.the only thin i can do is pip install