-
Notifications
You must be signed in to change notification settings - Fork 1.8k
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
[turborepo] turbo run
exits prematurely with code 2
#5260
Comments
Hey @be9, is this 100% reproducible? Or is this an intermittent issue? (looking for a hint to get a reproduction setup for debugging) |
It's failing in 100% of cases in Docker on our company's CI instance. Same commands work on macOS. |
@be9 do you happen to know if your docker images have |
@gsoltis sure they do. I can also add other stuff there, if needed. If you give me a build of |
@be9 Got it. To be clear, I will keep digging and see what I can come up with. |
@be9 Can you give it a try with We made some changes that should address it, and if not, should hopefully give us more information. |
Still works on macOS, but fails in Docker. |
+1 Experiencing the same exact problem. Possible cause? What exactly am I doing?
What have I found out?
The only difference that I can see between my local output and the CI output (beyond the exit code) is the addition of the Upon further investigation...
|
@be9 Odd that there's a platform difference. Are you sure both are using the same version of Btw, my theory is that you have a symlink that we are attempting and failing to hash. Any chance you know if that's correct? |
@jaa134 I believe this is the hashing issue. I suspect you have a symlink that we are attempting to hash. The |
I ran a CI pipeline for every version mentioned here from
Also, a different Turbo pipeline is now resulting in the Since, 2 different pipelines produced 2 different errors, I assumed this might have been a result of different issues in Turbo. Are you suspecting that's not the case and the 2 errors are actually related? Later tonight I will report back with results for |
We just released a new canary that should address the symlink issue, if
that is in fact the underlying cause. The quick version of that is we used
to accidentally ignore them, some new code stopped ignoring them, and now
we are explicitly ignoring them until we can fix how we handle them.
Can you check `turbo@1.10.4-canary.4`?
…On Tue, Jun 20, 2023, 5:39 PM Jacob Alspaw ***@***.***> wrote:
@gsoltis <https://github.com/gsoltis>
Are you sure both are using the same version of turbo?
I ran a CI pipeline for every version mentioned here
<https://github.com/vercel/turbo/releases?page=1> from 1.10.1 to
1.10.4-canary.3. I ended up trying ~15 different versions of Turbo in
total. It took me awhile, but I can assure you that:
- every version beyond 1.10.1 is broken
- the exit code 2 error message starts with version 1.10.2-canary.0
- the io error: Is a directory (os error 21) error message starts with
version 1.10.4-canary.0
Also, a different Turbo pipeline is now resulting in the io error: Is a
directory (os error 21) error message. The new error occurs while trying
to execute a build pipeline. The previous exit code 2 error occurred
while trying to execute a unit test pipeline (when the build pipeline
already succeeded). So upgrading to version 1.10.4-canary.0 has actually
made my CI process fail earlier on than it did in version 1.10.2-canary.0.
Since, 2 different pipelines produced 2 different errors, I assumed this
might have been a result of different issues in Turbo. Are you suspecting
that's not the case and the 2 errors are actually related?
Later tonight I will report back with results for --sumarize with version
1.10.4-canary.3 and 1.10.2-canary.0
—
Reply to this email directly, view it on GitHub
<#5260 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AADGD7IE3OSWFGELEPRLTF3XMI7FJANCNFSM6AAAAAAZAP6W6A>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
@gsoltis it was a symlink indeed! 🙏 |
@gsoltis Its working for me now too 🙏 . Thank you very much for the support! |
Super weird, I believe I'm seeing something along the same lines after the 1.10.4 canary push? I'm also unable to run a Errors:
1.10.2 - Works Perhaps it's not the same bug, happy to open a new ticket Edit, after more investigation, I don't think this is the same issue as it has nothing to do with symlinks, I've opened a new ticket. |
What version of Turborepo are you using?
1.10.2 (same problem with 1.10.3-canary.0)
What package manager are you using / does the bug impact?
pnpm
What operating system are you using?
Linux
Describe the Bug
I have a monorepo setup with 2 commands:
lint
andclean
.In a Docker image (Ubuntu 22.04-based), we run:
While it works 100% correctly with
turbo@1.10.1
, after upgrading to1.10.2
, turbo started to exit prematurely with code 2, not running anything. The log looks like this:Expected Behavior
What worked in 1.10.1, should continue working in 1.10.2.
To Reproduce
Unfortunately, I cannot provide sources 🤷
Reproduction Repo
No response
The text was updated successfully, but these errors were encountered: