jupyterlab 500 error from dataproc cluster

Viewed 358

i have been trying to open jupyterlab from my dataproc cluster. i had been working on it last week and there seemed to be some issue with the kernel. so i signed out of jupyterlab and stopped the cluster. then i tried to open jupyterlab again from the link under web interfaces in cluster detail. however, since last wednesday, everytime i try this i get a 500 or 502 error. the error says "500. That's an error. That's all we know."

i'm not sure what to do here. i googled the issue and couldn't find anything that would help.

1 Answers

You can SSH into the master node with gcloud compute ssh <cluster-name>-m, then check the status of the following 3 services:

systemctl status jupyter
systemctl status knox
systemctl status google-dataproc-component-gateway

Restart them if needed:

sudo systemctl restart jupyter
sudo systemctl restart knox
sudo systemctl restart google-dataproc-component-gateway

Also, check the Knox log at /var/log/knox/gateway.log.

Related