Modify launch script so that it has a configurable timeout #557
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
When multiple process groups exist (e.g., for vLLM weight sync in RLHF), initializing torch.distributed with device_id can cause NCCL hangs. This allows callers to pre-initialize torch.distributed without device_id before calling init_distributed(), which will then skip the initialization but still set up environment variables and CUDA device. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
When follow=False, make the launch command return immediately after submitting the job to Beaker, instead of waiting for the job to start. Changes: - In beaker.py: Resolve follow from self.follow before using it, and pass timeout=0 to gantry when not following to skip waiting - In experiment.py: Don't pass follow=True to launch(), and only pass step_soft_timeout when follow=True Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.