Skip to content

[None][chore] Upgrade ray version to 2.55.1#15819

Merged
shuyixiong merged 2 commits into
NVIDIA:mainfrom
shuyixiong:user/shuyix/update_ray_version
Jul 3, 2026
Merged

[None][chore] Upgrade ray version to 2.55.1#15819
shuyixiong merged 2 commits into
NVIDIA:mainfrom
shuyixiong:user/shuyix/update_ray_version

Conversation

@shuyixiong

@shuyixiong shuyixiong commented Jul 1, 2026

Copy link
Copy Markdown
Collaborator

Summary by CodeRabbit

  • Bug Fixes
    • Updated Ray compatibility so environments and tests work with newer Ray releases while still supporting older versions.
    • Improved placement-group handling to avoid import issues across different Ray versions.
    • Bumped the Ray dependency used in setup and CI to the latest supported version.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

  • PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

  • PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

  • Test cases are provided for new code paths (see test instructions)

  • If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

  • Any new dependencies have been scanned for license and vulnerabilities

  • CODEOWNERS updated if ownership changes

  • Documentation updated as needed

  • Update tava architecture diagram if there is a significant design change in PR.

  • The reviewers assigned automatically/manually are appropriate for the PR.

  • Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

Signed-off-by: shuyixiong <219646547+shuyixiong@users.noreply.github.com>

@Superjomn Superjomn left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@coderabbitai

coderabbitai Bot commented Jul 1, 2026

Copy link
Copy Markdown
Contributor

Review Change Stack

📝 Walkthrough

Walkthrough

Ray dependency version is bumped from 2.54.1 to 2.55.1 across Jenkins CI configuration, a Slurm install script, and a Verl integration config file. Additionally, PlacementGroupSchedulingStrategy imports are updated with version-tolerant try/except logic across executor and test files to support Ray >= 2.55.0's new import location.

Changes

Ray 2.55.1 Upgrade

Layer / File(s) Summary
Ray version pin updates
jenkins/L0_Test.groovy, jenkins/scripts/slurm_install.sh, tests/integration/defs/verl/verl_config.yml
Ray package version pinned at 2.55.1 (previously 2.54.1) in CI stage install commands, Slurm install script, and Verl config install commands.
Version-tolerant PlacementGroupSchedulingStrategy imports
tensorrt_llm/executor/ray_executor.py, tests/integration/defs/ray_orchestrator/RL/run_rl_perf_reproduce.py, tests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.py, tests/unittest/_torch/ray_orchestrator/multi_gpu/test_multi_instance.py
Each file now attempts to import PlacementGroupSchedulingStrategy from ray.util.scheduling_strategies (Ray >= 2.55.0) with a fallback to ray.util.placement_group for older Ray versions; other placement_group imports remain unchanged.

Estimated code review effort: 1 (Trivial) | ~5 minutes

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (2 warnings)

Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description is only the template placeholder and lacks the required Description and Test Coverage details. Fill in the Description and Test Coverage sections with the issue, solution, and relevant tests; keep the checklist only if completed.
Docstring Coverage ⚠️ Warning Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Write docstrings for the functions missing them to satisfy the coverage threshold.
✅ Passed checks (3 passed)
Check name Status Explanation
Title check ✅ Passed The title is concise, correctly formatted, and matches the main change: upgrading Ray to 2.55.1.
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

Comment @coderabbitai help to get the list of available commands.

@coderabbitai coderabbitai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🧹 Nitpick comments (1)
tensorrt_llm/executor/ray_executor.py (1)

11-20: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Consider centralizing the Ray compat import.

This try/except for PlacementGroupSchedulingStrategy is duplicated verbatim in ray_executor.py, run_rl_perf_reproduce.py, test_executor.py, and test_multi_instance.py. A single shared compatibility shim (e.g. a small ray_compat helper) would avoid drift if the import path changes again in a future Ray release.

🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@tensorrt_llm/executor/ray_executor.py` around lines 11 - 20, The Ray
compatibility import for PlacementGroupSchedulingStrategy is duplicated across
ray_executor.py and the related scripts/tests, so centralize it into a shared
ray_compat helper and import from there instead. Move the try/except fallback
into one reusable module, then update ray_executor.py and the other callers to
use that shared symbol so future Ray import-path changes only need one fix.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@tensorrt_llm/executor/ray_executor.py`:
- Around line 11-20: The Ray compatibility import for
PlacementGroupSchedulingStrategy is duplicated across ray_executor.py and the
related scripts/tests, so centralize it into a shared ray_compat helper and
import from there instead. Move the try/except fallback into one reusable
module, then update ray_executor.py and the other callers to use that shared
symbol so future Ray import-path changes only need one fix.

ℹ️ Review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: b283933e-117b-407a-a166-00c428426ed0

📥 Commits

Reviewing files that changed from the base of the PR and between d41d091 and df9f829.

📒 Files selected for processing (7)
  • jenkins/L0_Test.groovy
  • jenkins/scripts/slurm_install.sh
  • tensorrt_llm/executor/ray_executor.py
  • tests/integration/defs/ray_orchestrator/RL/run_rl_perf_reproduce.py
  • tests/integration/defs/verl/verl_config.yml
  • tests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.py
  • tests/unittest/_torch/ray_orchestrator/multi_gpu/test_multi_instance.py

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #56864 [ run ] triggered by Bot. Commit: df9f829 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #56864 [ run ] completed with state SUCCESS. Commit: df9f829
/LLM/main/L0_MergeRequest_PR pipeline #45675 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #56928 [ run ] triggered by Bot. Commit: d336064 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #56928 [ run ] completed with state SUCCESS. Commit: d336064
/LLM/main/L0_MergeRequest_PR pipeline #45735 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@shuyixiong shuyixiong enabled auto-merge (squash) July 2, 2026 01:30
@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57068 [ run ] triggered by Bot. Commit: d336064 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57068 [ run ] completed with state FAILURE. Commit: d336064
/LLM/main/L0_MergeRequest_PR pipeline #45864 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

1 similar comment
@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57178 [ run ] triggered by Bot. Commit: d336064 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57178 [ run ] completed with state SUCCESS. Commit: d336064
/LLM/main/L0_MergeRequest_PR pipeline #45960 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57201 [ run ] triggered by Bot. Commit: d336064 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57201 [ run ] completed with state SUCCESS. Commit: d336064
/LLM/main/L0_MergeRequest_PR pipeline #45976 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

  • Please check the failed tests and fix your PR
  • If you cannot view the failures, ask the CI triggerer to share details
  • Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

1 similar comment
@shuyixiong

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57385 [ run ] triggered by Bot. Commit: d336064 Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57385 [ run ] completed with state SUCCESS. Commit: d336064
/LLM/main/L0_MergeRequest_PR pipeline #46135 completed with status: 'SUCCESS'

CI Report

Link to invocation

@shuyixiong shuyixiong merged commit bf2ef86 into NVIDIA:main Jul 3, 2026
8 checks passed
BrianLi23 pushed a commit to BrianLi23/TensorRT-LLM that referenced this pull request Jul 9, 2026
Signed-off-by: shuyixiong <219646547+shuyixiong@users.noreply.github.com>
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

Successfully merging this pull request may close these issues.

4 participants