-
Notifications
You must be signed in to change notification settings - Fork 5.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
"Replacing stale connection" -- Kernal cannot connect #4457
Comments
Thanks! Figured out my issue was that I didn't have notebook installed. |
Oops, false alarm wasn't notebook, still doesn't work
|
I got the same issue after downgrading jupyter notebook from 5.7.6 to 5.7.4 due to #4467. |
downgrading to tornado 5.1.1 worked for me. |
Simply downgrading tornado to 4.2.0 worked for me. |
Upgrading |
upgraded notebook to 6.0.3 ,issue resolved |
This also worked for me - I had the latest tornado version 6.0.4 and the downgrade fixed the pblm |
I had the same problem when opening bash notebooks (not with python notebooks). Downgrading to different versions of tornado didn't help. Could this problem be related to port access? |
Thanks everyone, I downgraded tornado==5.1.1 to replaced 6.0.3 version, then jupyter (jupyter-notebook==5.6.0, jupyter core==4.6.1) works properly. |
Likewise, I was unable to start a new Jupyterlab session because of the "replacing stale connection" error message, but then I fixed the problem by running "pip install tornado==5.1.1" to downgrade from the latest available version. This thread was helpful for solving that error, thank you. |
Thank you guys. It worked for me too: tornado==5.1.1 + jupyter notebook 5.6.0 ! |
Whoa, works for me, thanks! |
Hey @kevin-bates, thanks for the response. So it seems like the kernel starts, but the client can't connect to it. It always shows as
Any insights would be appreciated. Thanks... |
Hmm. I suspect the websocket connection is not getting established. The output implies the notebook server is being started via Hub. To reduce interference, are you able to run the image directly using something like the following:
then enter the displayed URL beginning with It might also be helpful to enable debug logging. One way to do that is to just give the complete command in the
When I run that, I see the following surrounding the 'kernel started' info message:
It might be helpful to compare this output to yours. |
Thank you, that was the tip I needed. I upgraded my nginx config to upgrade to use websockets and now it connects and runs as expected. I appreciate you taking the time to point me in the right direction. Sorry for polluting this thread. |
No worries - glad to hear you're moving forward! I think it might be best to close this issue anyway due to its staleness and lack of details. It's easy to lump a general error message into the same bucket when it can be any number of things. In addition, the advice to downgrade relative to one and two-year versions is not prudent today. |
Hi, does anyone have an input on my issue? I am running Jupyter from Anaconda and updated everything to the latest. When I enter the notebook on port 8888 it never connects to the kernel and I cannot run the cells.
I followed every single guide I could read and tried changing ports, deleting adblocks, and downgrading tornado to 5.1.1 but still no luck.
Thanks!
The text was updated successfully, but these errors were encountered: