Kernel restarting: The kernel appears to have died. It will restart automatically.

Applies To

• Notebooks (.ipynb files)
• Interactive Window and/or Cell Scripts (.py files with #%% markers)

What happened?

Hi Folks, Whenever I run my .ipynb, the kernel is interrupted at the cell 31. I’ve tried with many search such as:

uninstall & reinstall 
    ipykernel
    ipython
    jupyter_client
    jupyter_core
    traitlets
    ipython_genutils

conda install ipykernel --update-deps --force-reinstall
python -m pip install traitlets==4.3.3 --force-reinstall
python -m pip install pyzmq==19.0.2 --force-reinstall

…from the following sources: stackoverflow, github, github But still no hope. I got the same issue with VSCode Jupyter notebook on pure Win10.

Any help please!? With thanks - BK

VS Code Version

Version: 1.64.2 (user setup) Commit: f80445acd5a3dadef24aa209168452a3d97cc326 Date: 2022-02-09T22:02:28.252Z Electron: 13.5.2 Chromium: 91.0.4472.164 Node.js: 14.16.0 V8: 9.1.269.39-electron.0 OS: Windows_NT x64 10.0.18362

Jupyter Extension Version

WSL: Ubuntu-18.04. v2022.1.1301854968, Win10

Jupyter logs

Info 22:18:06: Dispose Kernel process
Error 22:18:06: Raw kernel process exited code: undefined
Error 22:18:06: Error in waiting for cell to complete [Error: Canceled future for execute_request message before replies were done
	at t.KernelShellFutureHandler.dispose (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:51:432353)
	at /home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:66:307599
	at Map.forEach (<anonymous>)
	at y._clearKernelState (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:66:307585)
	at y.dispose (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:66:301017)
	at /home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:51:194444
	at t.swallowExceptions (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:16:2495)
	at dispose (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:51:194423)
	at t.RawSession.dispose (/home/bruce/.vscode-server/extensions/ms-toolsai.jupyter-2022.1.1301854968/out/client/extension.js:113:303296)
	at processTicksAndRejections (internal/process/task_queues.js:93:5)]
Info 22:18:06: Cancel all remaining cells due to dead kernel

Coding Language and Runtime Version

Python 3.9.7

Language Extension Version (if applicable)

No response

Anaconda Version (if applicable)

conda 4.10.3

Running Jupyter locally or remotely?

Local

[savagedroes]

Hey guys, I just had a similar problem. I tried to configure my GPU yesterday with TensorFlow and it appeared to have worked when trying this code in jupyter notebook: import TensorFlow as tf print(tf.version) print(len(tf.config.list_physical_devices(‘GPU’))>0)

but when I tried to train a convolutional neural net today my kernel stopped. When reading the error message I found in my anaconda prompt: Could not locate zlibwapi.dll. Please make sure it is in your library path!

I looked for solutions to this problem and found this forum: https://forums.developer.nvidia.com/t/zlib-dll-for-latest-cudnn-in-official-install-guide-is-missing/197630

where the last comment said: In NVIDIA website, they referred to zlibwapi.dll- I was able to locate this file in “C:\Program Files\Microsoft Office\root\Office16\ODBC Drivers\Salesforce\lib” (I installed using Microsoft 365 x64 in windows 11) and copy pasted this file into “C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin” I was able to run Tensorflow 2.8.0 thereafter

So I simply did this and it worked!

[DonJayamanne]

Kernel dies when I try to run inference using Numpyro

Please could you file an issue for this. Also, please test whether the same code works in regular Jupyter Notebook or Jupyter lab. Its most likely its failing there as well. If not, please do file an issue and we’ll take a look and try to get you unbloked