Kernel Crashes with #9741
Comments
philosofool
added
bug
|
@philosofool Thanks for filing. The main error looks like it's happening on the Jupyter server side. In general that might not be something that we can help to resolve at all if the server is crashing or cancelling the request. Does this kernel crash repro with just running The UI issue is odd. Any chance the notebook is something that you can share so that I could look at a repro? |
|
One suspicion from talking with the team. Is it possible that you are running out of memory on your system? The symptoms here both with the kernel crash and then the subsequent electron issues (Desktop Slack, UI oddness). |
|
Closing from lack of response. Please reopen if you can provide the necessary information. Thanks. |
|
I experienced this error today and can confirm it occurred due to an out of memory condition |
|
I experienced this error today. |
|
Just putting this here to help everyone that's going through this with Mac's Silicon (Yes, I know) . Problem is with the new versions of tensorflow-metal, downgrading to 0.3 worked for me. |
|
Same issue on win10 |
|
Had this error today on my mac (M1 Max with 64GB Memory, Python version 3.9.6) when I played with the MNIST dataset using PyTorch. However, when I switch back to cpu my code worked fine. |
|
I'm having this error trying to run a jupyter notebook on a remote server through vs code. I have no problems if I use the terminal in vs code to run the first cell of the notebook, but any time I try to connect a notebook I get similar problems to above. |
|
@Karlheinzniebuhr @Astroneko404 @jackransomlovell Just noticed that some new comments were coming in here. This particular issue of the Kernel crashing can be manifested from different root causes. So I don't want to re-open this old issue as there is a good chance your issue might not be related to the initial issue filed. If you are running into this now, please open up a new issue with Jupyter logs and repro info in the Jupyter issue page on github and we'll help take a look. |
|
same here |
|
I was having this error on M1 Mac too. The problem is I was using tensorflow 2.9 with tensorflow-metal 0.6.0. Downgrading tensorflow-metal to 0.5.0 according to https://developer.apple.com/metal/tensorflow-plugin/ solved it. |
@IanMatthewHuff I have been experiencing many crashes per day, some times even many per hour, and in my case the issue is that the system is running out of memory. There is a balloon driver for dynamic RAM... Having had a hint that lack of memory could be the issue would have been very useful for me. What about informing the user that the problem may be caused elsewhere(not in vscode), and offer the example of the system running out of memory? |
Ok, and how does one solve the out of memory condition? Sorry if question.is_noob :P |
Unfortunately, there's no uniform solution to out of memory errors and they are not really about VSCode in most cases. You will need to look at the code path where this error occurs and try to figure out which step causes it. I usually add some |
|
Visual Studio Code (1.76.0, undefined, desktop)
warn 17:49:25.630: StdErr from Kernel Process 2023-03-04 17:49:25.629906: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread p info 17:49:32.601: Cancel all remaining cells true || Error || undefined error 17:50:16.115: Disposing session as kernel process died ExitCode: 3, Reason: c:\Users\radu2\anaconda3\envs\tf\lib\site-packages\traitlets\traitlets.py:2544: FutureWarning: Supporting extra quotes around strings is deprecated in traitlets 5.0. You can use 'hmac-sha256' instead of '"hmac-sha256"' if you require traitlets >=5. info 17:50:16.115: Dispose Kernel process 19404.
warn 17:52:44.316: StdErr from Kernel Process 2023-03-04 17:52:44.316814: I tensorflow/core/common_runtime/proce info 17:52:52.000: Cancel all remaining cells true || Error || undefined error 17:53:20.119: Disposing session as kernel process died ExitCode: 3, Reason: c:\Users\radu2\anaconda3\envs\tf\lib\site-packages\traitlets\traitlets.py:2544: FutureWarning: Supporting extra quotes around strings is deprecated in traitlets 5.0. You can use 'hmac-sha256' instead of '"hmac-sha256"' if you require traitlets >=5. info 17:53:20.119: Dispose Kernel process 2084. |
Celula de cod 4.1.3save_path = 'test' # Modifica calea cu locul unde doresti sa salvezi fisierul audio #Calculez MFCC Plotare MFCCsfig = plt.figure() Canceled future for execute_request message before replies were done
|
|
Please could you file a separate issue, |
Applies To
What happened?
After running a cell that raised a a Python error while constructing a Keras model, re-running the cell crashes the kernel and has other undesired effects.
After the crash, all cells are surrounded by white (see image below.) This renders all output text invisible.
VSCode needs to restarThe notebook needs to be closed and reopened to solve the white border issue. An error message below the cell that caused the crash reads:Somewhat oddly, this also causes my desktop Slack client to move to a blank window. Slack > History > Back twice fixes this.
White border on cells:
VS Code Version
Version: 1.66.2 Commit: dfd34e8260c270da74b5c2d86d61aee4b6d56977 Date: 2022-04-11T07:49:24.808Z Electron: 17.2.0 Chromium: 98.0.4758.109 Node.js: 16.13.0 V8: 9.8.177.11-electron.0 OS: Linux x64 5.13.0-7614-generic snap
Jupyter Extension Version
v2022.3.1000901801
Jupyter logs
Coding Language and Runtime Version
Python 3.9.10
Language Extension Version (if applicable)
v2022.4.1
Anaconda Version (if applicable)
4.11.0
Running Jupyter locally or remotely?
Remote
The text was updated successfully, but these errors were encountered: