An error of torchrun in the docker container

Hi, I can successfully run my code using distributed training mode on my Linux server, but an error happens when I run my code in a docker container with the same environment as my server. The error is as follows:

I don’t know what (exit code: -7) means, and I don’t find a similar error on the web. Please help me.

1 Like

Have you resolved it? I also encountered the same issue.

1 Like