Closed nev7n closed 10 months ago
I'm running into this same issue. The extension is installed and enabled but this warning message appears and Jupyter-Scheduler doesn't appear in the launcher (or in the notebook nav bar). I can also confirm Jupyter-Scheduler was installed "in the same environment in which you run JupyterLab."
I've enabled --debug
when launching, but I'm still not seeing anything obviously wrong. The only minor difference I've noticed is that most of my labextension apps live in /opt/conda/envs/default/share/jupyter/labextensions
. In contrast, Jupyter-Scheduler and one other extension live in /home/<user>/.local/share/jupyter/labextensions
.
jupyter-scheduler == 1.3.1
@nev7n @iameskild Thanks for reporting this issue. Are there any errors you see in the server logs; it would help us investigate this issue.
Hi @3coins, I was trying to run this on a much older version of JupyterHub and once I set JUPYTERHUB_SINGLEUSER_APP=jupyter_server.serverapp.ServerApp
, I was able to get around this issue. It seems that older versions of JupyterHub run JupyterLab as an extension itself and this might have been the reason Jupyter-Scheduler wasn't being properly picked up.
Thanks for following up!!
@iameskild We haven’t delved into JupyterHub support but glad that you are able to run it.
@iameskild thanks for your response - it turns out we were also running an old version of JupyterHub (1.4.2), and after updating to JupyterHub==4.0.0 it resolved the issue.
We had a separate issue with a Javascript exception (calling method split on undefined, seemed to be related to trying to fetch file name from widget) when trying to add a schedule to a notebook by using right-click menu but resolved that by downgrading jupyterlab from 4.0.0 -> 3.6.3.
Hi all! I keep getting the same issue. I pip installed jupyter_schedule in the base environment which is where I have everything else installed and I keep getting the same error. I noticed that even after pip installing this, for some reason it keeps saying this is not installed when looking at the server extension list. Also anytime I try to pip install jupyter_scheduler, it keeps telling me the requirement is satisfied even after I try pip uninstalling and then re-installing. Please help. Thank you.
Regards,
Eric
Same problem here shows enabled OK but getting the server extension not found popup
Config dir: /home/jovyan/.jupyter
jupyter_scheduler enabled
- Validating jupyter_scheduler...
Package jupyter_scheduler took 0.1675s to import
jupyter_scheduler 2.4.0 OK
Config dir: /opt/conda/etc/jupyter
...
jupyter_scheduler enabled
- Validating jupyter_scheduler...
Package jupyter_scheduler took 0.0000s to import
jupyter_scheduler 2.4.0 OK
...
Config dir: /usr/local/etc/jupyter
JupyterLab v4.0.8
/opt/conda/share/jupyter/labextensions
jupyterlab_pygments v0.2.2 enabled X (python, jupyterlab_pygments)
nbdime-jupyterlab v2.2.0 enabled X
jupyter-matplotlib v0.11.3 enabled OK
@jupyter-notebook/lab-extension v7.0.6 enabled OK
@jupyterlab/git v0.41.0 enabled X (python, jupyterlab-git)
@jupyterlab/scheduler v2.4.0 enabled OK
@jupyterhub/jupyter-server-proxy v4.1.0 enabled OK
@jupyter-widgets/jupyterlab-manager v5.0.9 enabled OK (python, jupyterlab_widgets)
The following extensions are outdated:
jupyterlab_pygments
nbdime-jupyterlab
@jupyterlab/git
Consider checking if an update is available for these packages.
Other labextensions (built into JupyterLab)
app dir: /opt/conda/share/jupyter/lab
Hey folks, this error popup appears whenever the Scheduler lab extension (running in the browser) is unable to receive a response from the Scheduler server extension when JupyterLab is opened in the browser. As this is a very generic issue with several users suggesting different causes, the correspondence here is unlikely to be of benefit to future users. Therefore I will close this issue.
For us to help, we need a reproducible example, i.e. some instructions on how to reach the broken state that you are encountering. For example, you can export your Conda environment and include it in your issue:
conda list --explicit > env.txt
# then upload it as a GitHub gist and link to it in an issue comment
I also recommend testing Scheduler in a fresh Conda environment. If Scheduler works in this environment but fails in the previous one, then it is most likely an issue with your environment, not Scheduler. This can happen if you pip
blindly installs incompatible packages after installing jupyter-scheduler
and resolves some shared dependency to a version incompatible to jupyter-scheduler
. Otherwise, please open an issue, as it most likely is a bug in Scheduler.
To test Scheduler in a fresh Conda environment:
conda create -yn sch-test python=3.11
conda activate sch-test
conda install jupyter_scheduler~=2 jupyterlab~=4
jupyter lab
# now test on localhost:8888
Finally, I also recommend listing the dependencies of jupyter-scheduler
via pipdeptree
, a helpful utility that I'm fond of:
pip install pipdeptree # don't use conda install since the feedstock is outdated (for now)
pipdeptree --packages jupyter-scheduler
Happy Thanksgiving! Feel free to keep the conversation going here until you are able to reproduce the bug and file a separate issue on GitHub; we will still reply to comments in closed issues.
Description
After installing jupyter-scheduler==1.3.1 with pip we get the following error on starting JupyterLab:
I've tried running the suggested commands and it reports the extensions are both enabled:
I'm guessing this is related to a backend request failing with a 404 to (URL domain sanitised):
Not sure why this URL is returning a 404, I would assume the enabled server extension should handle this endpoint?
Reproduce
Expected behavior
Jupyter scheduler extension available after install
Context
Operating System and version:
Browser and version: Google Chrome: Version 112.0.5615.140 (Official Build) (64-bit)
Jupyter Server version: 1.24.0
Troubleshoot Output
Command Line Output
Browser Output