JupyterLab stopped working for Julia 1.9 (404 GETs) #1078
Comments
|
Same here. Maybe specific to Arch Linux. |
|
|
|
Jupyter notebook stopped working for me recently. The error on start-up shows I tried re-installing and re-building the |
|
After removing the kernel with the |
|
Same issue here, after last update of my Arch Linux distribution. Could not get it work by downgrading julia to 1.8, neither managed to roll back to other versions of jupyter. So it is very problematic. Here is the log with IJulia in verbose mode, and trying to install a new kernel in the global Julia environment (nothing changed). |
|
Finally managed to find a solution. All of this is probably due to the archlinux julia package. As noted in the documentation of the package, it is unsupported : https://wiki.archlinux.org/title/Julia The official way, through the package juliaup in AUR, worked for me. I was able to install the LTS and make IJulia work. Then I also did |
|
It was working before on all 1.85, 1.9 & 1.11 but does not work now. Perhaps due to updating a bunch of packages related to jupyter & ipython??? 🤔 Tried the latest nightly also julia 1.11, same errors. Windows 10 64-bitversioninfo()Tried from
|
|
It finally worked 😎 but the steps I did were perhaps totally random. UPDATE: FOUND WHY IT WORKS or DOESN"T in my machine:
Checked if any updates to nbconvert ✔️Tried to check if Julia kernel can be run through jupyter ⛔Tried to figure out if any new IJulia updates have occured or not. ⛔Tried to run julia kernel through jupyter again. ✔️ |
|
It seems like people are reporting multiple unrelated problems in this issue, which isn't very productive. e.g. "Package IJulia not found in current path." probably means that IJulia wasn't installed into your default environment, the one that is used when launching Whereas "kernel not found" means that somehow you installed a kernel that refers to the wrong path for Julia. In both cases you could look into the |
Sorry. I wasn't aware of the nuanced rules about this issue. I felt this was an already open issue & would get more eyeballs in case others needed another perspective on the problem. Also I'm a complete newbie to Julia & coding. As mentioned above, the whole problem goes away when I start in jupyter in the root dir of Julia. If I start jupyter anywhere else, it gives those errors. This is my kernel.json file.{
"display_name": "Julia 1.11.0-DEV",
"argv": [
"C:\\julia\\bin\\julia.exe",
"-i",
"--color=yes",
"--project=q",
"C:\\julia\\.julia\\packages\\IJulia\\Vo51o\\src\\kernel.jl",
"{connection_file}"
],
"language": "julia",
"env": {},
"interrupt_mode": "message"
}
|
|
Do you have multiple Julia environments set up? Is IJulia installed in the global environment (see here on how to switch to the global environment)? |
|
Hi Steve, Thanks for the follow up. Appreciate it.🙏🏻 tldr: start I checked whatever you said. Yes, IJulia is not installed in the global environment of Julia. I wanted to keep the So I tested for global and Yes, if I install IJulia in the global enviroment, jupyter starts with Julia kernel in any directory but if I install IJulia in a project I'm not sure why it should affect jupyter. Although, I feel somewhere something project path related to tldr: end Also I have not used julia to install either conda or jupyter notebook but installed them separately in another directory. ================ Case 1: IJulia installed in global environment & therefore Julia kernel works in any directory.start jupyter in any directoryCase 2: IJulia is installed in
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
versioninfo()Installation
I used
pacman. Also removed any dotfiles/directories for jupyter and julia. The issue still occurs. Everything is up-to-date.Julia 1.8 would run without issues.
I try to create a Julia 1.9 notebook and try to run a simple code block, like:
This is the output:
The text was updated successfully, but these errors were encountered: