fix(onboard): preserve served model route by laitingsheng · Pull Request #10882 · NVIDIA/NemoClaw

laitingsheng · 2026-09-02T13:47:18Z

Outcome

Managed vLLM onboarding now keeps the catalog's served model identity when a resumable install checkpoint records a short model alias. Fresh installs no longer reject their own endpoint because the checkpoint replaced the canonical route with the NEMOCLAW_VLLM_MODEL selector.

Reason

The resume checkpoint records install intent, while route validation compares the endpoint identity. Treating the selector as the route identity made a valid catalog alias conflict with the model that the managed endpoint actually serves.

Related issues

Fixes #10876

Changes

Normalise a checkpointed catalog selector through the existing managed-vLLM alias resolver before seeding route state.
Preserve the original checkpoint value for resume compatibility and retain exact unknown model identifiers.
Keep strict managed-endpoint mismatch validation unchanged; weakening that guard would admit genuine model conflicts.
Add a regression test that reproduces the canonical install seed followed by the alias checkpoint overwrite.

Verification

npx vitest run src/lib/onboard/setup-nim-flow-vllm-resume.test.ts src/lib/onboard/setup-nim-vllm.test.ts src/lib/inference/vllm-fixed-catalog-install.test.ts — 52 tests passed
npm run build:cli — passed
npm run typecheck:cli — passed locally and in the pre-push hook
npm run format:check — passed
npx oxlint src/lib/onboard/setup-nim-flow.ts src/lib/onboard/setup-nim-flow-vllm-resume.test.ts — passed with no lint errors
npm run source-shape:check — passed
npm run checks:repository — passed
git diff --check — passed
Secrets review — the diff contains no secrets, API keys, or credentials

Review notes

Sensitive-path context: the change only adapts catalog-backed checkpoint input before route-state validation. Strict endpoint identity validation and checkpoint persistence remain unchanged.
Documentation review: one read-only review completed with no missing, stale, unsafe, or inaccurate documentation findings. Existing alias documentation remains accurate.
Exploratory adjacent run: 81 tests passed; five existing Spark-selection cases in setup-nim-flow.test.ts timed out locally without assertion failures. Those paths do not execute the changed checkpoint adapter; the directly affected suites pass 52/52.

Signed-off-by: Tinson Lai tinsonl@nvidia.com

Summary by CodeRabbit

Bug Fixes
- Improved vLLM installation recovery when configured models use aliases.
- Ensured aliases resolve to the correct served model before resuming installation.
- Improved checkpoint sequencing during interrupted installations.
- Added stronger sandbox identity and model-availability validation during recovery.
- Improved recovery handling for failed installation attempts, including non-interactive setups.

Signed-off-by: Tinson Lai <tinsonl@nvidia.com>

coderabbitai · 2026-09-02T13:48:24Z

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: fdac37ed-6a2b-4bb2-bfc5-6f4ccb345a88

📥 Commits

Reviewing files that changed from the base of the PR and between 102918e and 07ce27a.

📒 Files selected for processing (1)

src/lib/onboard/setup-nim-flow.ts

Included review availability: Your plan provides up to 12 included reviews per hour; 4 remain after this review.

📝 Walkthrough

Walkthrough

The vLLM recovery flow resolves checkpointed catalog aliases to served model IDs before install-route validation. Resume tests verify identity revalidation ordering, managed-install intent, checkpoint sequencing, and repeated alias resolution.

Changes

vLLM Alias Recovery

Layer / File(s)	Summary
Normalize the recovery route `src/lib/onboard/setup-nim-flow.ts`	Recovery resolves checkpointed model IDs with `selectVllmModelFromEnv` before seeding the install route. The setup flow passes this resolver to the recovery policy.
Validate alias resume behavior `src/lib/onboard/setup-nim-flow-vllm-resume.test.ts`	Tests use catalog aliases and verify served-model selection, identity revalidation ordering, managed-install intent, checkpoint persistence, and retry behavior.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to 07ce2

Managed vLLM resume onboarding now accepts supported catalog aliases by validating against the served model identity, while preserving strict mismatch handling and checkpoint compatibility. No merge-blocking risk remains.

Suggested reviewers: ericksoa, prekshivyas, apurvvkumaria

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 75.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 4 functions across 2 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: preserving the served model route during onboarding.
Linked Issues check	✅ Passed	The changes address issue [`#10876`] by resolving catalog aliases to served model IDs before route state is seeded. The tests verify alias handling, served-model selection, checkpoint sequencing, and st…
Out of Scope Changes check	✅ Passed	The changes are limited to vLLM onboarding recovery logic and related regression tests. They directly support the linked issue and contain no unrelated code changes.

Full details: Linked Issues check

Explanation

The changes address issue [#10876] by resolving catalog aliases to served model IDs before route state is seeded. The tests verify alias handling, served-model selection, checkpoint sequencing, and strict mismatch validation.

Fix all pre-merge checks with AI

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

Create stacked PR
Commit on current branch

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch fix/vllm-checkpoint-route-identity

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

github-code-quality · 2026-09-02T13:48:32Z

Code Coverage Overview

Languages: TypeScript

TypeScript / code-coverage/plugin

The overall line coverage in commit 07ce27a in the fix/vllm-checkpoint-... branch remains at 96%, unchanged from commit 16142eb in the main branch.

_{Updated September 04, 2026 00:15 UTC}

github-actions · 2026-09-02T13:51:55Z

PR Review Advisor finished for commit d1c6769. Include the Advisor findings in the complete PR feedback collection. Verify and group valid findings before repair.

All previous runs

Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com>

prekshivyas

Reviewed exact head 07ce27a. Focused resume/identity tests and CLI build pass; all successful PR Review Advisor specialists are clean. Remaining cancelled title-lint executions stopped in dependency installation before linting and are being rerun as infrastructure flakes.

## Outcome Adds the canonical dated documentation entry for v0.0.120 and records the release's material user-facing changes before tag planning. The Hermes rebuild guide now also documents the fail-closed immutable-base requirement for legacy sandboxes without an image hint. ## Reason Release planning requires a merged `docs/changelog/2026-09-04.mdx` containing exactly one `## v0.0.120` heading. The existing automation draft does not contain that required changelog and does not cover the full release scope, so this PR provides a fresh, independently reviewed release-docs update. ### Related issues Relates to #10919 ## Changes - Add three release-note lead paragraphs and detailed, user-facing v0.0.120 changes with canonical documentation routes. - Cover configuration export and doctor (#11015, #11012); Hermes runtime, recovery, and Discord policy (#10595, #11071, #11024, #10927, #10983, #10988, #10999, #11019, #10682); Shields retirement (#10722, #10996); OpenShell forwarding and runtime authority (#10695, #10814, #10815, #10810); onboarding and recovery (#10690, #10900, #11046, #10882, #10864); inference behavior (#10956, #10910, #11070); Deep Agents MCP projection safety (#10911, #10909); and provider-profile validation (#10884, #10895). - Scope the legacy Hermes immutable-base rebuild guidance to the Hermes-rendered recovery page. ## Verification - `npx vitest run --project integration test/generation/check-docs-links.test.ts test/generation/check-docs-published-routes.test.ts test/generation/post-merge-docs.test.ts` — 3 files and 125 tests passed. - `npm run docs` — passed with 0 errors and 5 existing Fern warnings. - Independent documentation audit — reconciled all 71 commits in `v0.0.119..origin/main`, validated all 29 PR links and published routes, and found no unsupported product claims or remaining corrections. - Normal `pre-commit`, `commit-msg`, and `pre-push` hooks — passed. - `git diff --check` — passed. - GitHub commit verification — `a22fe0989fd72c7daaa9b2e7a4734a3edc069aba` is Verified with reason `valid`. - Secret review — the diff contains no secrets, API keys, or credentials. ## Review notes The existing automation draft #10919 is intentionally left untouched. This PR supersedes its release-docs content with the complete canonical changelog and a variant-correct Hermes recovery update. --- Signed-off-by: Charan Jagwani <cjagwani@nvidia.com>  ## Summary by CodeRabbit - **Documentation** - Added release notes covering verified configuration export, host and gateway diagnostics, service forwarding, sandbox recovery, onboarding safeguards, inference retries, MCP projection safety, provider setup, and Discord runtime policy. - Clarified sandbox rebuild behavior, including use of the release-pinned immutable base image when required. - Documented that rebuilds stop before modifying sandbox data when the required image cannot be resolved or validated.

fix(onboard): preserve served model route

d1c6769

Signed-off-by: Tinson Lai <tinsonl@nvidia.com>

laitingsheng added provider: vllm vLLM local or hosted provider behavior area: onboarding Onboarding FSM, provider setup, sandbox launch, or first-run flow bug-fix PR fixes a bug or regression v0.0.119 labels Sep 2, 2026

github-actions Bot added v0.0.120 and removed v0.0.119 labels Sep 2, 2026

prekshivyas added 3 commits September 3, 2026 16:04

test(vllm): cover interrupted alias resume

97c985f

Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com>

test(vllm): prove alias resume authority guard

102918e

Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com>

refactor(onboard): unify managed vLLM route resolution

07ce27a

Signed-off-by: Prekshi Vyas <prekshiv@nvidia.com>

prekshivyas approved these changes Sep 4, 2026

View reviewed changes

prekshivyas enabled auto-merge (squash) September 4, 2026 00:24

prekshivyas merged commit f7d2b3a into main Sep 4, 2026
77 of 93 checks passed

prekshivyas deleted the fix/vllm-checkpoint-route-identity branch September 4, 2026 00:27

hulynn mentioned this pull request Sep 4, 2026

[DGX Station][Onboard] Cannot Express install with Nemotron Ultra 550B — vLLM ready but model does not match its own route #11045

Open

cjagwani mentioned this pull request Sep 5, 2026

docs: prepare v0.0.120 release #11100

Merged

Navigation Menu

Uh oh!

fix(onboard): preserve served model route - #10882

fix(onboard): preserve served model route#10882
prekshivyas merged 4 commits into
mainfrom
fix/vllm-checkpoint-route-identity

laitingsheng commented Sep 2, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

coderabbitai Bot commented Sep 2, 2026 •

edited

Loading

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

github-code-quality Bot commented Sep 2, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented Sep 2, 2026

Uh oh!

prekshivyas left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

laitingsheng commented Sep 2, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Outcome

Reason

Related issues

Changes

Verification

Review notes

Summary by CodeRabbit

Uh oh!

coderabbitai Bot commented Sep 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

github-code-quality Bot commented Sep 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Code Coverage Overview

TypeScript / code-coverage/plugin

Uh oh!

github-actions Bot commented Sep 2, 2026

Uh oh!

prekshivyas left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

laitingsheng commented Sep 2, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Sep 2, 2026 •

edited

Loading

github-code-quality Bot commented Sep 2, 2026 •

edited

Loading