Issue connecting to Jupyter Kernel

MongoDB University M220P: MongoDB for Python Developers
#1

Hey all,
Im having issues with connecting to the Jupyter kernel. I can open the Jupyter fine through the shell, then when I open an individual notebook I cannot run anything because I cannot connect to the kernel. I have updated Jupyter already using:
pip install --upgrade jupyter-client jupyter-core

The error from the shell is:
AttributeError: ‘MappingKernelManager’ object has no attribute ‘connect_control’
Kind of stuck here, please advise!

EDIT: Upon further review this seems to be the error:

image
image1335×302 13.9 KB

#2

Hi @Robert_67632,

Please run the below commands and let me know if it helped you.

pip3 install --upgrade --user ipython

pip3 install --upgrade --force jupyter-console

Note: You have to run the commands in sudo if you are using a Linux based operating system.

#3

Hi @Robert_67632,

Thanks for the help @Harshavardhan_PK. :slight_smile: Let me know if it works @Robert_67632.

Kanika

#4

Thank you, I did get a warning about not being a specified path, so I added it.
It looks like it reinstalled a bunch of things and when I run \mflix-python\notebooks>jupyter notebook
I get a “not Implemented Error”

#5

can you send a screenshot of the error?

#6

image
image1287×515 31.5 KB

#7

Hi @Robert_67632,

After googling, one user mentioned the below solution. I am not sure if it will work for you.

upgrade jupyter-core to 4.6.1

Source: https://github.com/jupyter/notebook/issues/4613

Thanks for notifying this. Let me know if it works, then maybe we can fix the requirements file in the project itself.

Kanika

#8

Not able to update at this moment. It looks like an issue with python 3.8 (from the source you provided)
. Is that a requirement for the course? Will I be able to drop down to 3.7 without breaking anything else?

#9

Hi @Robert_67632,

I haven’t tested the project with python 3.8 version yet. Let me do it and get back to you.

Kanika

#10

Hi @Robert_67632,

You are right, the issue is with Python 3.8

This can be fixed by the following steps.

Add the below code at the start of the below-mentioned script in place of import asyncio

File Path: c:\users\robert.riley\appdata\local\continuum\anaconda3\envs\mflix\lib\site-packages\tornado\platform\asyncio.py

code to paste in place of: import asyncio

import sys

if sys.platform == 'win32':
    import asyncio
    asyncio.set_event_loop_policy(asyncio.WindowsSelectorEventLoopPolicy())

Try running jupyter notebook again

Note: If you are not able to find import asyncio in the script, add the mentioned code at Line - 1 of the script

1 Like
#11

Hey guys,
I depreciated python and I had to install a few others, asuncio and bflask are the two I remember. Everything seems to be running fine, for now. I am able to establish a kernel connection with juypter.

1 Like
#12