I am trying to start a ray cluster under Azure via the Azure cloud shell as described in the example
with the setup file example-full.yaml
However, I have commented out the conda setup commands as ray up always seemed to fail at those (could not find environment py37_tensorflow).
With this I can start a head node and connect to it via ray attach.
Then, I try to run ray status and get No Cluster Status with a ModuleNotFoundError:
I tried to install azure-common manually (via pip), but it is already installed.
I get the same error if I try to call ray.init(address=‘auto’) in a python script.
Furthermore I noticed that this setup only creates a head node but no worker nodes as described in the tutorial.
How can I get ray status and init to work? Is the problem with the worker node potentially liked to this issue?