Skip to content

Conversation

@kaiyux
Copy link
Member

@kaiyux kaiyux commented Sep 29, 2025

Summary by CodeRabbit

  • Documentation
    • Expanded Troubleshooting with guidance on a known hang related to UnpicklingError in MPI contexts, including current status, suggested reporting channels, and next steps.
    • Clarified that bypassing MPI with custom InfiniBand communicators may not resolve the issue; the team will monitor dependency upgrades for potential fixes.
    • Added contextual notes alongside manual cleanup guidance to help users recover from stalled runs.

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)

  • Any new dependencies have been scanned for license and vulnerabilities

  • CODEOWNERS updated if ownership changes

  • Documentation updated as needed

  • 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

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

@kaiyux kaiyux requested a review from a team as a code owner September 29, 2025 02:20
@coderabbitai
Copy link
Contributor

coderabbitai bot commented Sep 29, 2025

📝 Walkthrough

Walkthrough

Documentation update to examples/wide_ep/README.md adding troubleshooting notes about an UnpicklingError related to MPI, including context, current stance, and guidance; note appears twice adjacent to manual cleanup guidance.

Changes

Cohort / File(s) Summary
Docs: Troubleshooting update
examples/wide_ep/README.md
Added troubleshooting section describing an MPI-related UnpicklingError, attempted workaround via custom InfiniBand communicator, maintainability concerns, plan to monitor dependency fixes, and guidance for reporting blockers; content duplicated near manual cleanup notes.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)
Check name Status Explanation Resolution
Description Check ⚠️ Warning The pull request description consists solely of the unmodified template and a placeholder line without any actual title, description of the hang issue, explanation of the solution, or test coverage information. None of the required sections—such as a filled title with ticket and type, the Description section explaining what and why, or the Test Coverage section listing relevant tests—are populated. As a result, the PR description does not meet the repository’s template requirements. Please complete the PR description by providing a proper title following the “[ticket][type] Summary” format, filling in the Description section with details on the hang issue and its resolution, listing relevant tests in the Test Coverage section, and ensuring that all PR checklist items are accurately addressed.
✅ Passed checks (2 passed)
Check name Status Explanation
Title Check ✅ Passed The pull request title clearly follows the repository’s title convention with a ticket indicator of None, type [doc], and a concise summary stating that a hang issue caused by UnpicklingError is documented; it accurately reflects the primary change without extraneous details.
Docstring Coverage ✅ Passed No functions found in the changes. Docstring coverage check skipped.
✨ Finishing touches
  • 📝 Generate Docstrings
🧪 Generate unit tests
  • Create PR with unit tests
  • Post copyable unit tests in a comment

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

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

Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

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

Actionable comments posted: 0

🧹 Nitpick comments (2)
examples/wide_ep/README.md (2)

128-135: Tighten language, clarify scope, and add actionable guidance for users hitting UnpicklingError

Current text is speculative (“seems to be existing in MPI”), grammatically rough, and lacks concrete next steps. Recommend a concise “Known issue” block with scope, detection, temporary workarounds, and status, plus neutral wording. Example patch:

-### Hang issue caused by `UnpicklingError`
-
-It's possible to see hang issue that is caused by an `UnpicklingError`, we've noticed that and recorded it as a known issue. The issue seems to be existing in MPI, because we are not reproducing again after by-passing the MPI route by implementing customized InfiniBand communicator and replacing MPI API calls with that. We did not proceed because:
-1. The implementation only works on InfiniBand, hence not general enough.
-2. The implementation largely duplicated with InfiniBand communicator implementation in NCCL, which is hard to maintain.
-
-That being said, we are aware of the `UnpicklingError`, but instead of pushing further, we decided to keep observing for a while to see if it would be gone with further 3rd-party dependency upgrade. Please let us know if it's a blocker in your workload, and we will do necessary adjustment based on the feedback.
+### Known issue: potential hang with `UnpicklingError`
+
+You may encounter a hang preceded by an `UnpicklingError`. We have tracked this as a known issue.
+
+- Suspected scope: observed when using an MPI-based path. We were unable to reproduce the hang after bypassing MPI via a custom InfiniBand communicator that replaced MPI API calls.
+- Why we did not adopt that workaround: (1) it only works on InfiniBand and is not general; (2) it duplicates functionality already implemented by NCCL, making it hard to maintain.
+
+What you can do:
+- If this blocks your workload, please open a GitHub issue with logs (including the full `UnpicklingError` message and stack), MPI flavor/version, UCX/transport settings, and cluster/network details.
+- If possible, try an environment without MPI for this path, or adjust to a configuration that avoids the MPI route.
+
+Status:
+- We are monitoring upstream/dependency updates and will revisit once a robust, maintainable mitigation is available.

Also consider linking to an issue ID tracking this known problem to centralize updates.

Please confirm MPI flavors/versions affected and add a link to the tracking issue so users can subscribe.


76-76: Spelling nit: use “Troubleshooting” in the header

Change “## Trouble shooting” to “## Troubleshooting” for consistency with common docs style.

-## Trouble shooting
+## Troubleshooting
📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 28b9a81 and fa9369e.

📒 Files selected for processing (1)
  • examples/wide_ep/README.md (1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
  • GitHub Check: Pre-commit Check

@kaiyux kaiyux requested a review from dongxuy04 September 29, 2025 03:01
@kaiyux kaiyux enabled auto-merge (squash) September 29, 2025 03:14
@kaiyux
Copy link
Member Author

kaiyux commented Sep 29, 2025

/bot skip --comment "doc changes"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20204 [ skip ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20204 [ skip ] completed with state SUCCESS
Skipping testing for commit fa9369e

@kaiyux kaiyux merged commit 030254f into NVIDIA:main Sep 29, 2025
7 of 9 checks passed
@kaiyux kaiyux deleted the user/kaiyu/wideep_known_issue branch September 29, 2025 10:26
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Nov 1, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Nov 3, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Nov 3, 2025
dominicshanshan pushed a commit to dominicshanshan/TensorRT-LLM that referenced this pull request Nov 3, 2025
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