[https://nvbugs/5540979][fix] Potential fix for 5540979

[arekay-nv]

Summary by CodeRabbit

• Bug Fixes
  • Enhanced reliability of streaming generation responses with automatic retry logic for transient connection failures.
  • Improved error messages to include token count information for better debugging.
  • Refined error handling to gracefully manage connection interruptions during token streaming.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

Description

In disaggregated serving, the generation server may close the connection during token generation which was not being handled correctly. The current PR makes the error more verbose. It also adds a retry mechanism but the retry is only triggered if no tokens have been yielded. Retrying after tokens have been yielded will cause the generation phase to restart returning duplicate tokens. A better approach would be to issue a retry but skip the first n-tokens which were yielded in the failed phase, however that will have performance impact and may mask performance issues for debugging/performance tuning.

Test Coverage

PR Checklist

• [ ] 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.

[coderabbitai[bot]]

📝 Walkthrough

Walkthrough

Adds retry logic to the streaming generation path in merge_streaming_responses with error handling for client connectivity issues. Retries occur only when no tokens have been yielded and retry limit remains; failures raise HTTPException with status 500 including token count details. Improves error message formatting with actual type names.

Changes

Cohort / File(s)	Summary
Streaming retry and error handling tensorrt_llm/serve/openai_disagg_server.py	Wraps generation streaming path with retry loop (max_retries + 1 attempts). Retries on aiohttp.ClientError or OSError only when no tokens yielded and retry limit not reached. Raises HTTPException(500) on failure with token count in message. Improves TypeError formatting with actual gen_req type name.

Sequence Diagram

sequenceDiagram
    participant Client
    participant Server
    participant Backend
    
    rect rgb(200, 220, 255)
    Note over Server: Retry Loop (max_retries + 1)
    end
    
    loop Each Attempt
        Server->>+Backend: Stream generation request
        
        alt Success Path
            Backend-->>Server: Token chunks
            rect rgb(200, 255, 200)
            Note over Server: Stream chunks to client<br/>Update metrics
            end
            Server->>Client: Yield tokens
        else Client Connection Error
            Backend--X Server: aiohttp.ClientError / OSError
            
            alt Tokens Already Yielded
                rect rgb(255, 220, 200)
                Note over Server: Raise HTTPException(500)<br/>Include token count
                end
                Server->>Client: HTTPException
            else No Tokens Yielded & Retries Left
                rect rgb(255, 250, 200)
                Note over Server: Retry attempt
                end
            else No Tokens Yielded & Retries Exhausted
                rect rgb(255, 220, 200)
                Note over Server: Raise HTTPException(500)<br/>Include token count
                end
                Server->>Client: HTTPException
            end
        end
    end

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

• Core change is a localized retry wrapper around existing streaming logic with clear error boundaries
• Error handling follows standard patterns (retry-on-transient, fail-on-permanent)
• Single file modification with no API signature changes

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	You can run @coderabbitai generate docstrings to improve docstring coverage.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	Title follows the required format [NVBugs ID][type] and clearly references the bug being fixed, making the primary change evident.
Description check	✅ Passed	Description explains the issue and solution, though the Test Coverage section is empty and the PR checklist items remain unchecked.

✨ Finishing touches

• [ ] 📝 Generate docstrings

🧪 Generate unit tests (beta)

• [ ] Create PR with unit tests
• [ ] Post copyable unit tests in a comment

[!TIP]

📝 Customizable high-level summaries are now available in beta!

You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.

• Provide your own instructions using the high_level_summary_instructions setting.
• Format the summary however you like (bullet lists, tables, multi-section layouts, contributor stats, etc.).
• Use high_level_summary_in_walkthrough to move the summary from the description to the walkthrough section.

Example instruction:

"Divide the high-level summary into five sections:

1. 📝 Description — Summarize the main change in 50–60 words, explaining what was done.
2. 📓 References — List relevant issues, discussions, documentation, or related PRs.
3. 📦 Dependencies & Requirements — Mention any new/updated dependencies, environment variable changes, or configuration updates.
4. 📊 Contributor Summary — Include a Markdown table showing contributions: | Contributor | Lines Added | Lines Removed | Files Changed |
5. ✔️ Additional Notes — Add any extra reviewer context. Keep each section concise (under 200 words) and use bullet or numbered lists for clarity."

Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later.

---

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[pcastonguay]

/bot run --disable-fail-fast

[pcastonguay]

@arekay your branch has conflicts with main. Can you rebase and resolve conflicts so we can merge? Thanks.

[pcastonguay]

Closing this PR. Branch will be kept in case it's needed.