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

[Bug]: Sandbox image build failed on eval #4157

Closed
2 tasks done
neubig opened this issue Oct 2, 2024 · 9 comments
Closed
2 tasks done

[Bug]: Sandbox image build failed on eval #4157

neubig opened this issue Oct 2, 2024 · 9 comments
Labels
bug Something isn't working fix-me Attempt to fix this issue with OpenHands Stale Inactive for 30 days

Comments

@neubig
Copy link
Contributor

neubig commented Oct 2, 2024

Is there an existing issue for the same bug?

Describe the bug

I ran the following on the eval/24-sep branch:

$ ./evaluation/swe_bench/scripts/run_infer.sh llm.mini HEAD CodeActAgent 2 30 1 princeton-nlp/SWE-bench_Lite test &> ~/sand/eval_log.log

But I'm getting issues pulling the docker image:

�[92m20:44:21 - openhands:ERROR�[0m: runtime_build.py:386 - Sandbox image build failed: 404 Client Error for http+docker://localhost/v1.46/images/ghcr.io/all-hands-ai/runtime:d123cc51fdfb0f69ea7a3c8bdda940d7/json: Not Found ("No such image: ghcr.io/all-hands-ai/runtime:d123cc51fdfb0f69ea7a3c8bdda940d7")

Apparently this is caused by the fact that I ran this on an arm64 processor, which is not supported in SWE-bench evaluation.
To make this more clear for others who do the same thing, we should add a check in the main function of evaluation/swe_bench/run_infer.py and evaluation/swe_bench/eval_infer.py, and throwing an error if the processor type is arm64.

Current OpenHands version

eval/24-sep

Installation and Configuration

git clone

Model and Agent

No response

Operating System

No response

Reproduction Steps

No response

Logs, Errors, Screenshots, and Additional Context

No response

@neubig neubig added the bug Something isn't working label Oct 2, 2024
@xingyaoww
Copy link
Collaborator

Should be fixed by 9af6399 -- Feel free to reopen if it didn't fix it

@neubig neubig reopened this Oct 2, 2024
@neubig
Copy link
Contributor Author

neubig commented Oct 2, 2024

Thanks a bunch @xingyaoww ! But it seems like maybe it's still broken? Here's a log. I also deleted all my docker images to be sure it wasn't something residual.

eval_log.log

@xingyaoww
Copy link
Collaborator

�[92m23:05:08 - openhands:INFO�[0m: docker.py:31 - ---> [Warning] The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested

ahh i think it is probably because SWE-Bench doesn't officially support arm64 evaluation i believe? So maybe it is related to trying to build amd64 image on arm64?

@neubig
Copy link
Contributor Author

neubig commented Oct 2, 2024

Ah, I see... Maybe I'd have to do it on a VM.

@neubig neubig added the fix-me Attempt to fix this issue with OpenHands label Oct 2, 2024
Copy link
Contributor

github-actions bot commented Oct 2, 2024

OpenHands started fixing the issue! You can monitor the progress here.

Copy link
Contributor

github-actions bot commented Oct 2, 2024

An attempt was made to automatically fix this issue, but it was unsuccessful. A branch named 'openhands-fix-issue-4157' has been created with the attempted changes. You can view the branch here. Manual intervention may be required.

@neubig
Copy link
Contributor Author

neubig commented Oct 2, 2024

Hmm @xingyaoww , it seems that it's also failing on a computer where I have x86_64

$ arch
x86_64

Here is the log:
eval_log.log

Copy link
Contributor

github-actions bot commented Nov 2, 2024

This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

Copy link
Contributor

This issue was closed because it has been stalled for over 30 days with no activity.

@github-actions github-actions bot closed this as not planned Won't fix, can't repro, duplicate, stale Nov 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working fix-me Attempt to fix this issue with OpenHands Stale Inactive for 30 days
Projects
None yet
Development

No branches or pull requests

2 participants