[https://nvbugs/6190759][fix] Promote the multi-node guard to a module-level pytest.mark.skipif so skipping ha

[tensorrt-cicd]

Summary

• Root cause: disagg_server fixture asserts len(NODE_LIST) >= 2 before test body runs, so the in-test pytest.skip on single-node never fires.
• Fix: Promote the multi-node guard to a module-level pytest.mark.skipif so skipping happens at collection time before fixtures run; remove the redundant in-test skip; drop the stale waive entry left over from the engineer's UCX_TLS fix (90694fe).
• Automated fix generated by repair-bot

Test plan

• Verify fix on the same GPU type as the original failure
• Check for regressions in related tests

Links

• Bug: https://nvbugs/6190759

Summary by CodeRabbit

• Tests
  • Removed a waived test entry to enforce test execution.
  • Refactored test skip conditions to use module-level markers for improved test organization.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 71aa2660-dab5-489e-ab2f-31d3a7e8e861

📥 Commits

Reviewing files that changed from the base of the PR and between 4517988 and 8a95750.

📒 Files selected for processing (2)

• tests/integration/test_lists/waives.txt
• tests/unittest/llmapi/apps/_test_disagg_serving_multi_nodes.py

💤 Files with no reviewable changes (1)

• tests/integration/test_lists/waives.txt

---

📝 Walkthrough

Walkthrough

This PR migrates node-count validation from an inline runtime check to a module-level pytest marker in a multi-node disaggregated test, then removes the corresponding waived test entry from the integration waivers list. The test now skips execution at pytest collection time when exactly 2 nodes are not available.

Changes

Disaggregated Multi-Node Test Refactoring

Layer / File(s)	Summary
Node-count validation migration and waiver removal tests/unittest/llmapi/apps/_test_disagg_serving_multi_nodes.py, tests/integration/test_lists/waives.txt	Module-level pytestmark is expanded to include a skipif marker checking NODE_LIST length; the inline guard in test_completion that performed this check is removed; the integration waiver entry for test_openai_disagg_multi_nodes_completion[ctx_tp1pp2-gen_tp1pp2] is deleted.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#14461: Directly edits the same waives file by removing a different waived test entry.
• NVIDIA/TensorRT-LLM#14450: Modifies tests/integration/test_lists/waives.txt by adding waived entries in the same file.
• NVIDIA/TensorRT-LLM#14503: Updates the same integration waives file with different test entries.

Suggested reviewers

• jieli-matrix
• ZhanruiSunCh
• mzweilz

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (1 warning, 1 inconclusive)

Check name	Status	Explanation	Resolution
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.
Description check	❓ Inconclusive	The PR description explains the root cause, the fix, and references the bug, but lacks explicit test coverage details and PR checklist confirmation required by the template.	Add a more explicit Test Coverage section listing the specific tests affected, and confirm checklist items (API changes, dependencies, CODEOWNERS, documentation) for completeness.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title correctly identifies the main change—promoting multi-node guard to module-level pytest.mark.skipif—with proper NVBugs reference and [fix] type annotation.
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.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

Warning

Review ran into problems

🔥 Problems

Stopped waiting for pipeline failures after 30000ms. One of your pipelines takes longer than our 30000ms fetch window to run, so review may not consider pipeline-failure results for inline comments if any failures occurred after the fetch window. Increase the timeout if you want to wait longer or run a @coderabbit review after the pipeline has finished.

---

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

[schetlur-nv]

Closing based on bug status