3

When I try to open an existing or new notebook on my Google Vertex Notebooks instance (on Python 3.7.6), no code cells run and it says "kernel unknown" at the top left. When I try to run a code cell, an asterisk appears as if it were running, but nothing else happens. Everything else works fine (e.g. opening the terminal, editing markdown cells in a notebook, viewing files). The issue started happening immediately after I tried to update conda to fix conflicting package errors. What can I do to fix the issue and be able to run code cells again?

(Please let me know if I'm leaving out important information)

screenshot of the issue: error screenshot

merv
  • 67,214
  • 13
  • 180
  • 245
KD89042
  • 31
  • 1
  • 2
  • Hello, I think on the launcher you are able to select the kernel your notebook will use, did you updated the current notebook default configuration? which commands you used?? – Betjens Dec 20 '21 at 10:20
  • I think for this kind of issues you can create a case so google support can investigate further. But first, you will need to get the exact steps to reproduce the issue like commands used, if this is reproducible with every new instances you create. Also, if you have found a workaround for it? ( like creating a new notebook instance on GCP from scratch). After having that all figure it out I think you can open a case on [google support for notebooks issues](https://issuetracker.google.com/issues/new?component=958003) and it will be handled by google. – Betjens Dec 20 '21 at 10:25
  • I don't really understand what Betjens wants me to do (eg how do I get the exact commands used) so I don't have any updates on that front. It's still not working so I just created a new notebook instance as he suggested. – KD89042 Dec 23 '21 at 00:38
  • What I was looking to know was if you created a notebook and apply some changes (like running a codeblock that might have affected the notebook config) on your default kernel that might have change your current notebook configuration. If not, then if you created a new notebook instance and selected the conda kernel or default kernel to perform your operations, did it solve your issue? – Betjens Dec 27 '21 at 12:14

0 Answers0