We in our organisation have deployed z2jh setup through helm. It is hosted behind our app domain and used by data science people to create ML pipelines. Everything is working fine except the following issue:
While working on notebook, it randomly gives Error starting kernel, Kernel died unexpectedly. Most of the time there is no issue, people are able to run notebooks but every now and then kernel dies issue occurs. I have tried to gather as much data as I could. Any help is appreciated. We are sort of stuck here and it is very frustrating for users using the notebook. Sharing few screenshots here.
Check the logs (on k8s) for your singleuser servers to see if there are any clues. It’s also worth monitoring your resource usage (CPU, memory, storage, etc) to see if anything coincides with the failure.
Thanks @manics for you reply. I did check logs and resource usage. I could find nothing in logs and resource usage was also optimal ~20-30%
jupyter notebook kernel died randomly and it shows cpu usage 100
The kernel for XXX.ipynb appears to have died. It will restart automatically.