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

Windows CI still flaky (again) #252

Closed
dabrahams opened this issue Jan 23, 2023 · 27 comments
Closed

Windows CI still flaky (again) #252

dabrahams opened this issue Jan 23, 2023 · 27 comments
Assignees

Comments

@dabrahams
Copy link
Collaborator

https://github.com/val-lang/val/actions/runs/3990232735/attempts/1

@compnerd
Copy link
Collaborator

I was unable to reproduce this locally over 25 runs. I wonder if there is some sort of parallelism issue here.

@C-BJ
Copy link
Contributor

C-BJ commented Feb 19, 2023

I think this Issue may be closed

@C-BJ
Copy link
Contributor

C-BJ commented Feb 19, 2023

@kyouko-taiga

@kyouko-taiga
Copy link
Contributor

Fingers crossed

@kyouko-taiga
Copy link
Contributor

Seems Windows CI is still flaky :(

@kyouko-taiga kyouko-taiga reopened this Feb 22, 2023
@C-BJ
Copy link
Contributor

C-BJ commented Feb 24, 2023

Can you give me a link? @kyouko-taiga

@kyouko-taiga
Copy link
Contributor

Seems I can't get a link to the failed CI jobs I re-ran. I'll post a link here next time there's a Windows failure.

@C-BJ
Copy link
Contributor

C-BJ commented Feb 25, 2023

image
@kyouko-taiga

@kyouko-taiga
Copy link
Contributor

kyouko-taiga commented Feb 25, 2023

I've been looking at that history but couldn't find the failure. I don't remember exactly which branch it was, I commit a lot of stuff these days.

The failure was about 3 days ago, feel free to look for yourself.

@C-BJ
Copy link
Contributor

C-BJ commented Feb 25, 2023

https://github.com/val-lang/val/actions/runs/4242066814/jobs/7373093947
Got it @kyouko-taiga
But anyway, things are much better now than they were before, and within acceptable limits

@kyouko-taiga
Copy link
Contributor

@C-BJ
Copy link
Contributor

C-BJ commented Mar 3, 2023

@kyouko-taiga
Copy link
Contributor

Sadly, the flakiness is getting worse and starts seriously impeding development. So I decided to disable Windows CI until we can figure out a solution (see #495).

@C-BJ
Copy link
Contributor

C-BJ commented Mar 11, 2023

I noticed that the things happens when using Swift locally same as CI.
Swift also crashes locally.
@kyouko-taiga

@C-BJ
Copy link
Contributor

C-BJ commented Mar 12, 2023

I'm thinking this isn't a problem we can solve either, so what should we do?

@C-BJ
Copy link
Contributor

C-BJ commented Mar 12, 2023

I have an idea

@C-BJ
Copy link
Contributor

C-BJ commented Mar 12, 2023

Assign this issue to me?
@kyouko-taiga

@compnerd
Copy link
Collaborator

@C-BJ would you be able to get a stack trace or minidump from the failure? That would help understand what is going on (hitting an assert/invalid memory access).

@C-BJ
Copy link
Contributor

C-BJ commented Mar 13, 2023

@compnerd how? I just run a Swift command and it doesn't respond, and then it exits

@C-BJ
Copy link
Contributor

C-BJ commented Mar 13, 2023

#517 Need to be solved first

@compnerd
Copy link
Collaborator

@compnerd how? I just run a Swift command and it doesn't respond, and then it exits

If it crashes, you should be able to enable user mode dumps and collect a minidump. There are details on how to configure that at https://learn.microsoft.com/en-us/windows/win32/wer/collecting-user-mode-dumps.

@kyouko-taiga
Copy link
Contributor

Is this thread relevant to our issue: https://github.com/orgs/community/discussions/26933

@C-BJ
Copy link
Contributor

C-BJ commented Jul 29, 2023

At this stage we have temporarily solved this problem by retrying. #521

@dabrahams
Copy link
Collaborator Author

I've recently reworked the CI quite substantially and uncovered/suppressed one possible cause of this problem so I'm going to close this issue until such a time as we see problems again. At that point we should file a new issue.

@dabrahams
Copy link
Collaborator Author

Well no, that wasn't the problem. Reopening until I implement the "try again" workaround.
This isn't a Hylo problem.

@dabrahams dabrahams reopened this Mar 6, 2024
@dabrahams
Copy link
Collaborator Author

I implemented a lot of workarounds, and now we're all green again… for now. I only pray this is an SPM problem that will be fixed by using cmake.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants