[Bug]: when try to test the benchmark swe-bench_Lite, I'm getting a running error with docker.py but I don't know how to fix it.

[Succoney]

Is there an existing issue for the same bug? (If one exists, thumbs up or comment on the issue instead).

• I have checked the existing issues.

Describe the bug and reproduction steps

When I wanted to verify the test results of OpenHands on Swe-bench, I followed the instructions in the repository to install the environment and configure the llm parameters. But when I started reasoning, an error was unexpectedly reported. It reported an error when running docker.py, saying that apt-get update had no permissions. But when I changed the code here to sudo apt-get update, another error was reported. In short, the error should be reported when downloading docker.

OpenHands Installation

Docker command in README

OpenHands Version

No response

Operating System

Linux

Logs, Errors, Screenshots, and Additional Context

web：
https://github.com/All-Hands-AI/OpenHands/tree/main/evaluation/benchmarks/swe_bench

bash：
./evaluation/benchmarks/swe_bench/scripts/run_infer.sh llm.deepseek-chat HEAD CodeActAgent 10 10 1 princeton-nlp/SWE-bench_Lite test

error：

[Succoney]

[csmith49]

Can't say I've seen this exact behavior, but 9/10 times errors when building the Docker image are caused by running out of storage. Probably worth trying a docker system prune and seeing if that fixes anything.

[Succoney]

Thank you for answering my question. This problem has been solved. I think it should be that the terminal running permissions of Ubuntu are low. When I re-run the code on Mac OS, there is no problem at all.