[BUG] In version 7.0.0, on executing a cell, the kernel status is stuck at connecting indefinitely. #1445

Activity

1. gsm1258 commented on Oct 14, 2025

   

   gsm1258

   on Oct 14, 2025

   More actions

   I'm encountering the same issue—Python 3.14.0 won't run, while versions 3.13, 3.12, and 3.10 all work fine.
2. ianthomas23 commented on Oct 14, 2025

   

   ianthomas23

   on Oct 14, 2025

   Collaborator

   More actions

   Please pin to ipykernel < 7 whilst this is dealt with.
3. ianthomas23

   closed this as completedin #1447on Oct 14, 2025
4. ianthomas23 commented on Oct 14, 2025

   

   ianthomas23

   on Oct 14, 2025

   Collaborator

   More actions

   Ipykernel 7.0.1 has been released on PyPI and works with Python 3.14. The new conda-forge package will follow shortly.

   👍React with 👍2Reacted by Balint Decsi and Soumadeep Sarkar
5. soumadeepsarkar1 commented on Oct 14, 2025

   

   soumadeepsarkar1

   on Oct 14, 2025

   Last edited by soumadeepsarkar1

   Author

   More actions

   @ianthomas23 Thanks a lot for the prompt resolution to this issue😄. I have installed ipykernel 7.0.1 and this issue has been fixed now.
6. earino

   added a commit that references this issue on Nov 14, 2025

   Add requirements.txt and improve dependency management

   ...

   7efc73f
7. csipapicsa commented on May 18, 2026

   

   csipapicsa

   on May 18, 2026

   More actions

   Same here. I ran into what looks like the same problem, but through VS Code Remote-SSH notebooks.

   In my case, both affected conda environments were on:

   ipykernel 7.2.0
   jupyter_client 8.8.0
   jupyter_core 5.9.1
   pyzmq 27.1.0

   The symptoms were:

   • VS Code notebook cells stuck indefinitely in Pending
   • kernel restart stuck/spinning
   • interrupt did not help
   • replacing/restarting the kernel from the VS Code UI did not reliably help
   • the same issue occurred in two separate conda environments and two separate notebooks
   • the environments worked normally from the shell, so the issue seemed specific to the notebook/kernel connection path

   Downgrading ipykernel fixed it:

   python -m pip install "ipykernel<7"

   This installed:

   ipykernel 6.31.0

   After the downgrade, I also had to make VS Code refresh the selected kernel:

   1. close/reopen the affected notebook or VS Code Remote-SSH window
   2. select another kernel temporarily
   3. switch back to the downgraded/fixed kernel
   4. run the notebook again

   After that, execution worked normally again.

   So this seems consistent with the earlier report that ipykernel 7.x can cause kernels to remain stuck connecting/executing, while ipykernel 6.x works.

   👍React with 👍2Reacted by hahampis and Lucas Rencker
8. Abdullah-Masood-05

   added a commit that references this issue on Oct 10, 2026

   fix(setup): pin notebook ipykernel to >=6.31.0,<7

   ...

   bc1c891