# Configuring jupyterhub with cadabra2

+1 vote

Hi, I recently installed cadabra2 on xubuntu and also installed jupyter-lab and configured it to run cadabra2. I'm new to linux so it took some time to figure out what's what...

The thing is - the linux computer I'm using isn't my default workstation - a windows 10 pc - and so I want to run and jupyterhub on xubuntu and access cadabra notebook via my web browser (LAN access from the pc to the xubuntu computer).

I got jupyterhub to work using the instructions from this page. However when I start a new cadabra notebook, it's stuck on "kernel is starting", and then fails.

I don't know how to see the log file of the actual failure. I also don't know if the problem is the missing PYTHONPATH specification, and if so - I don't know how to configure jupyterhub to include it before it begins - it starts as a service.

Any help would be appreciated.

OK... I think I solved the issue by updating the jupyterhub.service configuration. I had to locate the jupyterhub.service file. In my system it was in the folder /etc/systemd/system/

i.e.

sudo gedit /etc/systemd/system/jupyterhub.service

Underneath the existing "Environment=" line, I added another line

Environment=export PYTHONPATH=\${PYTHONPATH}:/usr/local/lib/python3.9/site-packages

After saving the file, I had to re-load the configuration into the system with

sudo systemctl daemon-reload

finally - restart the service:

sudo systemctl restart jupyterhub.service

Now I can run cadabra2 commands in jupyterhub from any computer in my LAN (I configured a firewall rule to allow http traffic only from LAN IPs).

by (260 points)