Closed jabbera closed 9 months ago
@jabbera thanks for opening the issue. I saw something similar in the past when the kernel crashes (https://github.com/deshaw/jupyterlab-execute-time/issues/84) and I believe that was already fixed in https://github.com/deshaw/jupyterlab-execute-time/pull/100 but the fix is only effective in JupyterLab 4.1+ (currently in release candidate phase). Which version of JupyterLab did you use?
In general, can you provide reproduction instructions, or does it happen sporadically without a clear trigger?
Finally, do you have windowing enabled in JupyterLab settings (Settings → Settings Editor → Notebook → Windowing mode, at the very bottom)?
FYI, there is a newer version (3.1.1) available.
JupyterLab 4.0.10 I haven't been able to pin down the cause but it is happening during a long running memory intensive op. There is the potential the kernel crashed without providing feedback. Windowing mode is set to Defer
I'm happy to wait till 4.1 comes out and validate that the issue goes away!
I am seeing this with defer + 4.1.0rc1
I can reproduce when re-running a cell which was already running. There is an error in the console:
ExecuteTimeWidget.ts:311 Uncaught TypeError: Cannot read properties of undefined (reading 'execution_failed')
at ExecuteTimeWidget.ts:311:34
so this is a regression from the PR I mentioned. I will open a PR to fix it.
@jabbera if you would like to test the fix from https://github.com/deshaw/jupyterlab-execute-time/pull/112, you can download extension-artifacts
from the associated CI build and install the patched version using pip
.
This was released
Sorry of the description is odd, but I didn't know how to explain it.
https://github.com/deshaw/jupyterlab-execute-time/assets/301208/ea199936-435b-4415-9a98-d24ef1884c9a
jupyterlab_execute_time 3.1.0 pyhd8ed1ab_0 conda-forge