Failed to list clusters: might the server extension not be installed/enabled?

Hello,

I’m having troubles to get my dashboard setup on a SageMaker notebook instance.

Tests I’ve done:

- Checked serverextension list:
(JupyterSystemEnv) sh-4.2$ jupyter serverextension list
config dir: /home/ec2-user/.jupyter
    jupyterlab_git  enabled 
    - Validating...
Error loading server extension jupyterlab_git
      X is jupyterlab_git importable?
config dir: /home/ec2-user/anaconda3/etc/jupyter
    dask_labextension  enabled 
    - Validating...
      dask_labextension 1.0.3 OK
    ipyparallel.nbextension  enabled 
    - Validating...
      ipyparallel.nbextension  OK
    jupyter_server_proxy  enabled 
    - Validating...
      jupyter_server_proxy  OK
    jupyterlab  enabled 
    - Validating...
      jupyterlab 0.32.1 OK
    nbdime  enabled 
    - Validating...
      nbdime 1.1.0 OK
    nb_conda  enabled 
    - Validating...
      nb_conda 2.2.1 OK

• copy-pasted the url below in the extension search box

Nothing much happens doing that.

The pop up error “Failed to list clusters: might the server extension not be installed/enabled?” show up when disabling/enabling the lab extension.

Tried: https://github.com/dask/dask-labextension/issues/61#issuecomment-515786252

Any ideas?

[jacobtomlinson]

A quick note if you don’t mind your dashboard being exposed to the internet on some obscure URL you can always proxy it with serveo. Bit of a hacky workaround but thought I’d share anyway.

Once you have your Dask cluster running in Sagemaker you can open a terminal and run ssh -R 80:localhost:8787 serveo.net where localhost:8787 is the host and port of your dashboard relative to the notebook server. It will probably ask you to accept some host keys but then it will print a URL where your dashboard will be available.

In this example the subdomain virgo was randomly assigned but you can also override this if you check the serveo docs.

[dschick]

okay,

jupyter serverextension enable dask_labextension

solved the problem, although I am running jupyter notebook 6.0.2