jupyter server : not started, no kernel in vs code

Question

i am trying to use jupyter notebooks from vs code and installed jupyter notebook extension and i am using (base)conda environment for execution. while this happened

Error: Jupyter cannot be started. Error attempting to locate jupyter:
at A.startServer (c:\Users\DELL\.vscode\extensions\ms-python.python-2020.2.63990\out\client\extension.js:1:784356)
at async A.ensureServerAndNotebookImpl (c:\Users\DELL\.vscode\extensions\ms-python.python-2020.2.63990\out\client\extension.js:1:783811)
at async A.ensureServerAndNotebook (c:\Users\DELL\.vscode\extensions\ms-python.python-2020.2.63990\out\client\extension.js:1:783612)
at async A.submitCode (c:\Users\DELL\.vscode\extensions\ms-python.python-2020.2.63990\out\client\extension.js:1:780564)
at async A.reexecuteCell (c:\Users\DELL\.vscode\extensions\ms-python.python-2020.2.63990\out\client\extension.js:75:879318)

how to resolve this issue ?

I restarted VSCode and somehow this fixed it. – Shayan Apr 02 '23 at 02:26 — Shayan, Apr 02 '23 at 02:26

score 76 · Accepted Answer · edited Mar 30 '20 at 06:54

76

I had exactly the same problem when I installed Visual Studio Code and tried to run some Python code from a jupyter notebook on my fresh Ubuntu 18.04.

How I solved it:

1) Press Command+Shift+P to open a new command pallete

2) Type >Python: Select Intepreter to start jupyter notebook server

3) Open the notebook again

And it worked fine. Hope it works for you.

edited Mar 30 '20 at 06:54

keddad

1,398
3
14
35

answered Feb 26 '20 at 14:45

ABarrier

888
1
6
8

2

now i have only below one `>Jupyter: Select Intepreter to start jupyter notebook server` – KeepLearning Jun 21 '21 at 03:01

Antonio Velazquez Bustamante · Answer 2 · 2020-03-14T20:55:19.147

I have several versions of Python installed. It happened the same thing to me and I have fixed it this way. Ctrl+Shift+p and select Python: Select Interpreter to start Jupyter server

Then, select the version under the Visual Studio Code

Nothing will happen and then press again Ctrl+Shift+p and select Python: Create new blank Jupyter Notebook. And it works

I have even set the Python version to 3.8 at the bottom and it worked too with the new features print(a:=4) despite the fact that the version I have chosen was 3.7.5. Nevertheless, I have to lunch VS Code from Anaconda Navigator.

score 12 · Answer 3 · edited May 20 '22 at 12:28

12

Press Command+Shift+P on mac, Ctrl+Shift+p on windows
Type Jupyter: Select Interpreter to start Jupyterserver
It would show you a dropdown of python versions installed.
I chose python 3.7.5 and it worked for me you can choose the python version installed on your machine.

edited May 20 '22 at 12:28

Largence

13
3

answered Feb 27 '20 at 18:44

Bodunrin David

169
1
9

How if at number three, it didn't show any, yet using `jupyter kernelspec list`, there are some kernel. Restarting VSCode does help, but it is a hassle. – Muhammad Yasirroni Nov 16 '22 at 02:33
`Type Jupyter: Select Interpreter to start Jupyterserver` this command is not found. It would help to provide more information on the devcontainer setup that resulted in this command being available – sharchaea Jun 04 '23 at 16:49

score 10 · Answer 4 · answered Jun 10 '20 at 13:00

10

I have seen all possible solutions but not work, finally I just upgrade jupyter,notebook,and jupyterlab,like pip3 install -U jupyterlab, and I can choose the kernel in VScode!

answered Jun 10 '20 at 13:00

John Jim

111
1
2

1

yes! Make sure all of your dependencies are installed first. `pip` installing the above fixed my errors – StonedTensor Oct 04 '21 at 16:57

score 4 · Answer 5 · edited Apr 21 '20 at 10:55

4

I tried the following:

Press Command + SHIFT + P
Type Python: Select Interpreter to start Jupyterserver

Hope this answer was helpful.

edited Apr 21 '20 at 10:55

Sunny Sharma

4,688
5
35
73

answered Apr 21 '20 at 07:10

Anjali Reddy

61
2
10

score 4 · Answer 6 · edited Dec 13 '20 at 18:47

4

Making sure that in VS Code settings.json

"python.condaPath": "C:\\Program Files\\miniconda3\\Scripts\\conda.exe"

is pointing to the correct directory. It solved it for me.

edited Dec 13 '20 at 18:47

fala

125
4

answered May 21 '20 at 14:43

Mazel

49
4

score 4 · Answer 7 · answered Nov 22 '22 at 17:31

I faced a similar issue quite often in VS Code, sometimes I can't get the kernel from my virtual enviroment (instead, VS Code only finds other venvs that are not related to my current project).

I tried reloading the window, selecting interpreter to start jupyter, reloading the VS Code itself, but nothing worked.

In case all the above and the other answers fail, try that, it worked for me:

Ctrl+Shift+P
Jupyter: Filter kernels
Select only the kernel you want (in my case, my venv)
Go on "Select Kernel" directly on the notebook UI;
Select your right kernel.

It is weird that the venv kernel appears on "Filter Kernels", but not always appears on the kernel list. But doing this might solve the issue.

score 2 · Answer 8 · answered Sep 25 '20 at 09:16

2

just fix this by add

"python.terminal.activateEnvironment": false,

to settings.json

hopes this help.

answered Sep 25 '20 at 09:16

Peter

99
5

score 2 · Answer 9 · answered Nov 06 '20 at 10:57

just had the same issue and it did not help to update the interpreter within vscode. What helped was: Check your dependencies within pip! It seems that new dependencies came up with the latest update of python, which are not installed. For me this was pygments:

jupyter-console 6.2.0 requires pygments, which is not installed.

Linux solution step-by-step:

xyz@xyz-pc-ubuntu:~$ pip3 check
qtconsole 4.7.6 requires pygments, which is not installed.
nbconvert 5.6.1 requires pygments, which is not installed.
jupyter-console 6.2.0 requires pygments, which is not installed.
ipython 7.18.1 requires pygments, which is not installed.

xyz@xyz-pc-ubuntu:~$ pip3 install pygments
Successfully installed pygments-2.7.2

Afterwards, jupyter found the updated python interpreter automatically.

score 2 · Answer 10 · answered Jun 23 '22 at 09:44

i've stumbled upon this post, since i had a similar issue. provided that my context was different, since i was working remotely on a linux server, even if i selected the right interpreter (via shift+ctrl+P "Select Interpreter to start Jupyter server") the kernel remained unactive. i've checked the installed dependencies inside the venv and tried to switch virtual environment to make it work.. kept on reloading the server, reloading the window.. no way.

eventually, a tiny fancy detail arouse my attention: the "Jupyter server : remote" label in the bottom right. and tadaa : that was my issue. I've selected "default", letting VSCode starting a server on the local (remote) host, and then the interpeter / kernel was enabled.

hope it can help anyone stuck on the same issue.

score 1 · Answer 11 · answered Aug 14 '20 at 18:27

In my case, I had the server working in 3.7.6, but i wanted to use >3.8.0 versions too. After multiple attempts, which failed, I decided to:

Uninstall 3.8.5, and delete the folder in the installation directory.
Uninstall VSCode too.
Restart the PC, and re-install Python and VSCode. As a result, the Jupyter server initiated based on the latest version of Python automatically.

I hope this helps too! Cheers!

score 1 · Answer 12 · edited Feb 09 '23 at 11:57

1

Install the "Python Extension Pack" in vsc.
Then select the correct conda environment.

edited Feb 09 '23 at 11:57

Dorian Turba

3,260
3
23
67

answered Feb 03 '23 at 12:39

zeppeh

13
4

score 1 · Answer 13 · answered Apr 03 '23 at 07:41

1

In my case, the extension had to be reloaded, and indeed it read Reload required.

answered Apr 03 '23 at 07:41

reim

492
5
8

score 0 · Answer 14 · answered Nov 16 '22 at 02:38

For me the problem is that VSCode can't find the kernel, even in using the select interpreter option.

The most reliable solution that I can find and currently used is:

Install without cache:

pip install jupyter notebook jupyterlab pyzmq --upgrade --no-cache-dir

Restart VSCode

Another extra safety step is to uninstall first and followed with pip cache purge

volkit · Answer 15 · 2023-05-23T13:35:54.577

I am using Manjaro Linux and what helped me was to install the code-features package from AUR-repository:

pamac search -a code-features
pamac build code-features

First step is not necessary, just to confirm package is found in aur repo. Maybe helps with other arch based distros, too.

After that i could select the interpreter with Ctrl+Shift+p and select "Python: Select Interpreter to start Jupyter server".

score -1 · Answer 16 · answered Oct 03 '22 at 09:03

-1

For me uninstalling the Jupyter extension, closing VS code and then reinstalling it worked. Not a really great solution, but the only one that worked for me. Hope that may help someone.

answered Oct 03 '22 at 09:03

Heimson

63
4

score -1 · Answer 17 · answered Dec 14 '22 at 09:43

I had a comparable problem when trying to run a jupyter notebook on a remote server. The problem was solved by installing the Python Extension on this particular server from the extensions menu in VSCode. After that, the correct kernel was automatically chosen in the jupyter notebook and I could run everything smoothly.

score -1 · Answer 18 · answered Dec 16 '22 at 23:31

Fix this by going to settings and search for Environment, Under Extensions you'll find Python , Under python you can see a option called Python>Terminal : Activate Environment (Uncheck this box)

Then you can change your python interpreter to whichever version you like

hope this helps

score -1 · Answer 19 · answered Feb 28 '23 at 09:52

-1

In my case the output of a cell was very big and got stuck on loading. Clearing the output of the cells worked for me.

answered Feb 28 '23 at 09:52

Nick_Z0

86
13

score -1 · Answer 20 · answered Apr 18 '23 at 17:30

-1

I had disabled the Jupyter extension, and enabling it solved the issue.

answered Apr 18 '23 at 17:30

Lupilum

363
2
11

score -1 · Answer 21 · answered Jun 11 '23 at 09:17

-1

I faced this issue with this v2023.5.1001582324 version of jupyter vs code extension, i revert it back to old version and it fixed it.

answered Jun 11 '23 at 09:17

Prathamesh Zade

9
3

jupyter server : not started, no kernel in vs code

21 Answers21

Linked