Skip to content
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

Test: Kernel Picker "Kernel Source" experiment #11765

Closed
2 tasks done
IanMatthewHuff opened this issue Oct 24, 2022 · 4 comments
Closed
2 tasks done

Test: Kernel Picker "Kernel Source" experiment #11765

IanMatthewHuff opened this issue Oct 24, 2022 · 4 comments

Comments

@IanMatthewHuff
Copy link
Member

IanMatthewHuff commented Oct 24, 2022

Refs: #11576

Authors: @IanMatthewHuff

Complexity: 4

Create Issue


@roblourens and @amunger. Felt like Don, Peng and I had seen this the most already so assigned the two of you to get new eyes on this.

To get the settings for this get the following in your settings:
"jupyter.showOnlyOneTypeOfKernel": false, "jupyter.experimental.kernelPickerType": "Insiders",

Note, this is all behind an experiment for now. Testing can be pretty broad here, but I've tried to list out some basic steps and a few known issues. Comments on the basic flow are better than testing every nook and cranny at this point.

Steps:

  1. Start without any Jupyter servers attached.
  2. Open up a notebook document
  3. You should have a Select Kernel Source command (also if you had a previously selected kernel it should be available) in the kernel picker

image

  1. Picking that should list Local Kernels
  2. After selecting "Local Kernels" you should see the normal kernel selection choices
  3. Pick the Jupyter Server: Local item and add a remote server

image

9. When a server is added you should also get the option to name it, try giving a custom name 10. After server(s) are added, they should show up as options in the "Select Kernel Source" command 12.

image

13. When picking a remote kernel source you should only see kernels from that source as options 14.

image

15. Open a second notebook file. 16. Connect it to a different kernel source 17. You should be able to select different sources for different open documents, and picking a different source in on should not kill other documents with other kernels sources (like the previous experiment used to do) 18.

image

19. In above example one document is connected to Server B kernel source and the other is connected to Local kernel sources
@ghost ghost assigned roblourens and amunger Oct 25, 2022
@roblourens
Copy link
Member

Pick the Jupyter Server: Local item and add a remote server

Why is there no option to add a remote server behind the "Select Kernel Source" option?

@amunger
Copy link
Contributor

amunger commented Oct 25, 2022

There is still a big disconnect between conntecting to a jupyter server and then selecting the kernel. I would really expect those two steps to be tied together instead of having to then find the same server in the kernel picker.
I was confused at where my remote kernels were after connecting and only found them after reading through the next steps listed in this issue.

@amunger
Copy link
Contributor

amunger commented Oct 25, 2022

There was some flakiness around re-selecting the same kernel when re-opening notebooks after having multiple open with different sources, but I'm struggling to get a consistent repro. I'm also confused about what is supposed to show up in the initial list along with 'select kernel source', most of the time it was only the currently selected kernel, which seemed kind of strange.

@amunger amunger removed their assignment Oct 25, 2022
@roblourens roblourens removed their assignment Oct 25, 2022
@github-actions github-actions bot locked as resolved and limited conversation to collaborators Sep 25, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

4 participants