-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
Python Extension API incorrectly triggers 'remove' event #20292
Python Extension API incorrectly triggers 'remove' event #20292
Comments
Job https://github.com/microsoft/vscode-jupyter/actions/runs/3579027060 I'm uploading the two logs here separately
|
Running into this very often,
|
Still running into this on CI
This is most likely the cause of the bug that users are also running into microsoft/vscode-jupyter#11747 @karrtikr do you have an update on this bug, we're running into this issue very often on CI and I believe our users too are running into this issue on the stable release and having issues running code in notebooks. |
I'll try to have a look today. Meanwhile you can work around this issue at Jupyter's end by selectively ignoring |
I have created a PR #20424 which should hopefully take care of this, leaving this open so you can verify. Happy holidays! |
For #20292 We currently remove environments from cache which aren't part of the latest discovery run. This is not correct as consumers can still select envs not reported by discovery run, which we shouldn't remove.
Can you ship a insider release now, so this can be tested by Jupyter users as well as SteVen. Also, whats the possibility of having a recovery release for this, we have a few users running into this. @brettcannon /cc |
I'm OOF starting now until 3rd, Brett can help with the timelines. I've mentioned a workaround here in case the timeline doesn't work for you guys: #20292 (comment)
Done, should be out soon. |
And I also kicked it off, so you're doubly covered. 😉
Before the New Year, there's unfortunately no chance as there's no one around to handle any issues that may come up. But we can definitely consider doing a bugfix release the week of Jan 3 if you can confirm that the fix does what you need. |
Re-opening this issue as we're still running into this with the latest bits. We've had a few other failures as well, but forgot to report this. |
@DonJayamanne This probably won't make it in the current iteration, as I'm OOF most of the week. Consider using the workaround I suggested above for this release. |
@karrtikr that work around also does not work as Python extension is caching the information.
Bug: I think there's another bug in the API |
Strange, 'remove' event is only fired after the environment is removed from cache, so 'resolveEnvironment' should ideally be returning 'undefined'. Maybe there's some other caching that's going on. |
@karrtikr just wanted to check if there was any progress on this. If this isn't resolved for this iteration or cannot be resolved, then we might want to add some work around in the Jupyter extension. |
We've other priority items that are being worked on right now, but we should be able to tackle this in this iteration as well. But given this has a simple workaround, please use that meanwhile.
It just means the next step is to investigate, we can resolve this immediately after in this iteration. |
FYI all of our "needs" labels are scoped to the issue, not to a specific iteration; it just records what is keeping the issue open. |
Thanks for the clarification |
I'm not able to reproduce this issue on Jupyter CI, confirmed after removing the workaround which was added. No tests fails, and I checked verbose logging to ensure that no |
please note it was flaky, it wasn’t something that happened 100% of the time |
Okay I'll try it a few more times, I have added additional logging in the Python extension to diagnose in case this happens again. |
Able to repro it finally after running 10 runs concurrently 😄 Reason The cause of this issue was different than what was originally reported in the issue. |
On CI we're using a Python global Python env,
In the middle of the tests, the Python extension API triggers and event indicating the python environment has been deleted
and this then causes a few other issues in Jupyter extension
Note: We do not delete the global python env hence I do not expect to receive such and event.
If we're using the event incorrectly please do let us know.
Here are the interesting logs
& here are the logs from Python extension at the same time
The text was updated successfully, but these errors were encountered: