[https://nvbugs/6245317][test] set Harmony tiktoken env for GPT-OSS disagg

[dongfengy]

Without this, tiktoken will try to download online and sometimes fail with
[05/26/2026-11:55:37] [TRT-LLM] [E] [serve] Error in harmony chat completion: %s error downloading or loading vocab file: failed to download or load vocab file

Summary by CodeRabbit

• Tests
  • Improved test environment configuration for disaggregated model testing to ensure proper resource handling and isolation.

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.

[dongfengy]

/bot run --disable-fail-fast

[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: 0a64bea4-3e77-4b4d-8a7a-6de4fe0ca6a3

📥 Commits

Reviewing files that changed from the base of the PR and between 846d0f4 and eff1338.

📒 Files selected for processing (1)

• tests/integration/defs/disaggregated/test_disaggregated.py

---

📝 Walkthrough

Walkthrough

A test in the disaggregated integration test suite now sets up tiktoken cache environment variables before running the test. The environment is copied from llm_venv._new_env and modified with TIKTOKEN_RS_CACHE_DIR and TIKTOKEN_ENCODINGS_BASE pointing to a tiktoken_vocab directory under llm_models_root()/datasets, then passed to run_disaggregated_test.

Changes

Tiktoken Environment Variables for Disaggregated Test

Layer / File(s)	Summary
Tiktoken cache environment configuration tests/integration/defs/disaggregated/test_disaggregated.py	The test prepares a modified environment by copying llm_venv._new_env, computing a tiktoken_vocab path under llm_models_root()/datasets, exporting that path via TIKTOKEN_RS_CACHE_DIR and TIKTOKEN_ENCODINGS_BASE, then passing the modified env to run_disaggregated_test.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

🚥 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 contains only the problem statement about tiktoken errors but lacks required template sections like Description, Test Coverage, and detailed explanation of the solution.	Add a clear Description section explaining the issue and solution, and list the relevant test(s) in Test Coverage section as required by the template.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main change: setting Harmony tiktoken environment variables for a GPT-OSS disaggregated test to prevent online vocab file downloads.
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

---

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

[tensorrt-cicd]

PR_Github #51970 [ run ] triggered by Bot. Commit: eff1338 Link to invocation

[tensorrt-cicd]

PR_Github #51970 [ run ] completed with state FAILURE. Commit: eff1338
/LLM/main/L0_MergeRequest_PR pipeline #41319 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

[dongfengy]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #52122 [ run ] triggered by Bot. Commit: e787532 Link to invocation

[tensorrt-cicd]

PR_Github #52122 [ run ] completed with state FAILURE. Commit: e787532
/LLM/main/L0_MergeRequest_PR pipeline #41447 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

[dongfengy]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #52136 [ run ] triggered by Bot. Commit: e787532 Link to invocation

[tensorrt-cicd]

PR_Github #52136 [ run ] completed with state SUCCESS. Commit: e787532
/LLM/main/L0_MergeRequest_PR pipeline #41460 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

[dongfengy]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #52383 [ run ] triggered by Bot. Commit: cc549e8 Link to invocation

[tensorrt-cicd]

PR_Github #52383 [ run ] completed with state SUCCESS. Commit: cc549e8
/LLM/main/L0_MergeRequest_PR pipeline #41678 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

[dongfengy]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #52443 [ run ] triggered by Bot. Commit: cc549e8 Link to invocation

[tensorrt-cicd]

PR_Github #52443 [ run ] completed with state SUCCESS. Commit: cc549e8
/LLM/main/L0_MergeRequest_PR pipeline #41736 completed with status: 'SUCCESS'

CI Report

Link to invocation