Kernel is dead after some time inactive, now unable to Start Lab

Course Q&A Generative AI with Large Language Models GenAI with LLMs Resources
1

Hi all,
I am in Week 1 of the course “Generative AI with Large Language Models” Coursera | Online Courses & Credentials From Top Educators. Join for Free | Coursera,

While working with the Jupyter Notebook I often have this problem: The kernel is dead after some time of inactivity, and I don’t know how to quickly restart the Kernel, as the button “Kernel”–> “Restart Kernel” doesnt work.

Then I have to wait for some time, relaunching the Workbench once more, Start Lab once more, and loading the models etc, which would cost several hours.
UPDATE: the dead kernel problem has made me unable to Start Lab, error message is “account is still in cleanup”. How to solve this ?
Questions are: (1) How to keep the kernel being not dead for at least 30 minutes even I switch to another browser’s tab ? (2) If tomorrow I come back to work this course, what I should do to be able to quickly load the Jupyter Notebook ?

2

I found the answer:
“Note: The AWS account, which was created for the lab, expires within 2 hours. During this period you can close all of the console windows and come back to your work later. After the expiration the current AWS account will go through a cleanup procedure (which will take up to 25 minutes), then the access to the new account will take longer (up to 20 minutes) and your previous work will not be saved. To save the notebook locally before the expiration you can download the notebook from the Amazon SageMaker Studio (right click on the notebook → “Download” command).”

1 Like