Skip to content

fix: fix rollout metric name in RolloutManager unit test - #2778

Merged
yuki-97 merged 1 commit into
mainfrom
yukih/fix-rollout-metrics
Jun 11, 2026
Merged

fix: fix rollout metric name in RolloutManager unit test#2778
yuki-97 merged 1 commit into
mainfrom
yukih/fix-rollout-metrics

Conversation

@yuki-97

@yuki-97 yuki-97 commented Jun 11, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

fix test_async_rollout_manager_matches_original which is skipped by CI:Lfast in #2567.
https://github.com/NVIDIA-NeMo/RL/actions/runs/27322340912/job/80728668060

@yuki-97
yuki-97 requested a review from a team as a code owner June 11, 2026 07:26
@copy-pr-bot

copy-pr-bot Bot commented Jun 11, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@yuki-97 yuki-97 added the CI:L0 Run doctests and unit tests label Jun 11, 2026
@yuki-97

yuki-97 commented Jun 11, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test bfe854b

@yuki-97
yuki-97 requested review from RayenTian and terrykong June 11, 2026 07:27
@yuki-97

yuki-97 commented Jun 11, 2026

Copy link
Copy Markdown
Contributor Author

this PR only fix test_async_rollout_manager_matches_original which cause L0_Unit_Tests_Vllm_1 failed, and this test is passed in https://github.com/NVIDIA-NeMo/RL/actions/runs/27330998639/job/80757534319?pr=2778, so changed to CI:Lfast to merge.

the other two fails:

  1. L0_Unit_Tests_Mcore: related to Megatron offload_after_refit leaves params on GPU (re-onload via offload_before_refit) #2775, @bg51717 is fixing.
  2. L0_Unit_Tests_Other: fixed in fix: increase nightly test time 1820 -> 1890 #2777.

@yuki-97 yuki-97 added CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version) and removed CI:L0 Run doctests and unit tests labels Jun 11, 2026
@yuki-97
fix key name
34fcf2b 
Signed-off-by: Yuki Huang <yukih@nvidia.com>
@yuki-97
yuki-97 force-pushed the yukih/fix-rollout-metrics branch from bfe854b to 34fcf2b Compare June 11, 2026 10:43
@yuki-97

yuki-97 commented Jun 11, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test 34fcf2b

@yuki-97
yuki-97 enabled auto-merge (squash) June 11, 2026 10:44
@yuki-97
yuki-97 merged commit 0c0988d into main Jun 11, 2026
76 checks passed
@yuki-97
yuki-97 deleted the yukih/fix-rollout-metrics branch June 11, 2026 11:24
pengdurice pushed a commit to pengdurice/RL that referenced this pull request Jun 12, 2026
@yuki-97 @pengdurice
fix: fix rollout metric name in RolloutManager unit test (NVIDIA-NeMo…
31401a1 
…#2778)

Signed-off-by: Yuki Huang <yukih@nvidia.com>
ashors1 pushed a commit that referenced this pull request Jun 27, 2026
@yuki-97 @ashors1
fix: fix rollout metric name in RolloutManager unit test (#2778)
a2a9a77 
Signed-off-by: Yuki Huang <yukih@nvidia.com>
Signed-off-by: Anna Shors <ashors@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@RayenTian RayenTian RayenTian approved these changes
@terrykong terrykong Awaiting requested review from terrykong

Assignees

No one assigned

Labels

CI:Lfast Runs a fast test suite and re-use nightly `main` container (but sync dependencies to PRs version)

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@yuki-97 @RayenTian