Skip to content

Remove breaking torchrun config for single-node runs #292

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.

Sign up for GitHub

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

Merged
hemildesai merged 2 commits into from
Jul 25, 2025
Merged

Remove breaking torchrun config for single-node runs #292

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1f4dfb3
remove breaking torchrun config for single-node runs
ri-roee Jul 15, 2025
8d18fad
fix lint
ri-roee Jul 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 0 additions & 13 deletions nemo_run/core/execution/skypilot.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,6 @@
Executor,
ExecutorMacros,
)
from nemo_run.core.execution.launcher import FaultTolerance, Torchrun
from nemo_run.core.packaging.base import Packager
from nemo_run.core.packaging.git import GitArchivePackager

Expand Down Expand Up @@ -330,18 +329,6 @@ def macro_values(self) -> Optional[ExecutorMacros]:
het_group_host_var=self.HET_GROUP_HOST_VAR,
)

def _setup_launcher(self):
super()._setup_launcher()
launcher = self.launcher
# Dynamic rendezvous has an error in Skypilot Kubernetes currently
if (
launcher
and isinstance(launcher, (Torchrun, FaultTolerance))
and self.cloud == "kubernetes"
):
launcher.rdzv_backend = "static"
launcher.rdzv_port = 49500

def to_task(
self,
name: str,
Expand Down
Loading