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

NCCL tests hang when run in slurm interactive job #568

Open
abergeron opened this issue Mar 26, 2018 · 2 comments
Open

NCCL tests hang when run in slurm interactive job #568

abergeron opened this issue Mar 26, 2018 · 2 comments

Comments

@abergeron
Copy link
Member

This is due to those tests using mpiexec which attempts to start subprocesses with srun (if built with slurm support).

This hangs because srun says that job launching is "temporarily disabled" forever when in an interactive job. This could be fixed by forcing a different mpi launcher for the tests if we think it matters enough.

@nouiz
Copy link
Member

nouiz commented Apr 18, 2018

Do you know how we can specify a different launcher? Is it easy?

@abergeron
Copy link
Member Author

I remember that there is a simple parameter like --launcher but it's not that.

I don't remember what it was exactly and the mpi documentation/google/man pages are of course not at all helpful about this.

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

No branches or pull requests

2 participants