I noticed that the tutorial suggests mapping port 8080 of the VM to localhost. The problem is that if ssh disconnects, the port is gone and Jupyter notebook gets stuck.
To bypass this, I changed the Jupyter config to allow all origins and I connect directly to the VM's public IP address at port 8080 - and that continues to work regardless of the SSH.
If that sounds like a better approach, I can share the config I changed for Jupyter (maybe there's a better way to do it)
This refers to docs/start_gcp.md
I noticed that the tutorial suggests mapping port 8080 of the VM to localhost. The problem is that if ssh disconnects, the port is gone and Jupyter notebook gets stuck. To bypass this, I changed the Jupyter config to allow all origins and I connect directly to the VM's public IP address at port 8080 - and that continues to work regardless of the SSH.
If that sounds like a better approach, I can share the config I changed for Jupyter (maybe there's a better way to do it)
Cheers