ci: bump async-gym megatron test wall-clock budget to 120 min#3258
Merged
Conversation
The llm_grpo_nanov3_30BA3B_2n8g_megatron_generation_async_gym job completes all 8 GRPO steps but is SIGKILLed at the 90-minute Slurm limit during teardown before metrics.json is written. nemo-ci mislabels the exit-137 as Signal/OOM-Kill; there is no CUDA OOM in the logs. Signed-off-by: shanmugamr1992 <shanmugamr1992@gmail.com>
kajalj22
approved these changes
Jul 16, 2026
yuki-97
added
CI:Lfast
yuki-97
approved these changes
Jul 17, 2026
Contributor
|
/ok to test b6357cd |
yuki-97
enabled auto-merge (squash)
July 17, 2026 10:02
Contributor
|
/ok to test d0af02d |
arnavk-nvidia
pushed a commit
that referenced
this pull request
Jul 17, 2026
Signed-off-by: shanmugamr1992 <shanmugamr1992@gmail.com> Signed-off-by: Arnav Kundu <arnavk@nvidia.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
NUM_MINUTESfrom 90 to 120 ingrpo-nanov3-30BA3B-2n8g-megatron_generation-async-gym.sh.Async GRPO training complete!) but is SIGKILLed at the 90-minute Slurm wall-clock limit during teardown, beforejson_dump_tb_logs.py/check_metrics.pyrun. nemo-ci classifies exit 137 asSignal/OOM-Kill; there is no CUDA OOM in the driver log.Test plan
/ok to test <commit-sha>on the MRllm_grpo_nanov3_30BA3B_2n8g_megatron_generation_async_gymfinishes withmetrics.jsonwritten on cw_dfw