fix(installer): accept OTA-upgraded DGX OS on Station GB300 without image marker

[hunglp6d]

Summary

The Station express installer classifies a GB300 DGX OS host from its /etc/dgx-release metadata, but it only recognized DGX OS when the full-image identity field DGX_OTA_PRETTY_NAME="DGX OS" was present. A Station GB300 that was first installed from an older base image (for example 7.4.1-GB300ws) and later OTA-upgraded to 7.5.0 accrues DGX_OTA_VERSION/DGX_OTA_DATE but never gains that field, so it fell through to unsupported-dgx-os and hit the express boundary error even though it is a genuine, healthy DGX OS 7.5.0 GB300 host. This change classifies a host that carries an OTA history by its latest applied OTA version, so such a host is now recognized as supported-dgx-os and takes the existing stock-DGX-OS path that reuses the factory driver and container runtime without mutating host packages.

Related Issue

Fixes #7224

Changes

• scripts/prepare-dgx-station-host.sh: in dgx_station_release_profile, gate the DGX OS branch on OTA-history presence and the latest DGX_OTA_VERSION (7.2.0/7.4.0/7.5.0) rather than on the DGX_OTA_PRETTY_NAME image marker. When that marker is present it must still read exactly DGX OS; when it is absent, the host must identify as the GB300 workstation via DGX_PRETTY_NAME="NVIDIA DGX GB300WS" so that other release lineages which also emit DGX_OTA_* fields stay fail-closed. The now-shadowed DGX_OTA_VERSION guard in the no-OTA branch is removed; the DGX_OTA_DATE guard is kept as defense-in-depth.
• No new classification string or downstream wiring: the host resolves to the existing supported-dgx-os → stock-dgx-os profile, which already reuses the factory driver and container runtime with no package, service, or runtime mutation.
• test/install-station-dgx-os.test.ts: add a positive case for a marker-less OTA-upgraded GB300 workstation (mirrors the reported /etc/dgx-release, classified as supported-dgx-os), plus fail-closed cases for a non-workstation DGX Server identity and for an unreviewed latest OTA version.

Type of Change

• Code change (feature, bug fix, or refactor)
• Code change with doc updates
• Doc only (prose changes, no code sample modifications)
• Doc only (includes code sample changes)

Quality Gates

• Tests added or updated for changed behavior
• Existing tests cover changed behavior — justification:
• Tests not applicable — justification:
• Docs updated for user-facing behavior changes
• Docs not applicable — justification: no user-facing docs change; the DGX Station support matrix and boundary wording are unchanged (still Deferred; still DGX OS 7.2.0/7.4.0/7.5.0). Only the release-metadata classifier is broadened to recognize an OTA-upgraded host.
• Sensitive paths changed (security, policy, credentials, preflight, onboarding, inference, runner, sandbox, or messaging)
• Sensitive-path review completed or maintainer-approved waiver recorded — reviewer/approval link/justification:
• Non-success, skipped, or missing CI check accepted by maintainer — check name, approval link, and follow-up issue:

Verification

• PR description includes a Signed-off-by: line and every commit appears as Verified in GitHub
• Normal pre-commit, commit-msg, and pre-push hooks passed, or npm run check:diff passed when hooks were skipped or unavailable
• Targeted behavior tests pass for the current change set, or tests are marked not applicable above — command/result or justification:
• Applicable broad gate passed — npm test for broad runtime/test-harness changes; npm run check for repo-wide validation/coverage changes — command/result:
• Quality Gates section completed with required justifications or waivers
• No secrets, API keys, or credentials committed
• npm run docs builds without warnings (doc changes only)
• Doc pages follow the style guide (doc changes only)
• New doc pages include SPDX header and frontmatter (new pages only)

---

Signed-off-by: Hung Le hple@nvidia.com

Summary by CodeRabbit

• Bug Fixes

  • Improved DGX OS release detection for OTA-upgraded workstations.
  • Added validation for supported OTA versions and release identities.
  • Prevented unsupported DGX Server identities and unreviewed OTA versions from being accepted.
  • Correctly recognizes eligible OTA-upgraded GB300 workstations without a fresh-install marker.

• Tests

  • Added coverage for OTA-upgraded workstation release detection and validation scenarios.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[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: 572ff1fb-ac9c-4fbf-a1ad-47674da11bf9

📥 Commits

Reviewing files that changed from the base of the PR and between 3c4ea5a and 17dda63.

📒 Files selected for processing (2)

• scripts/prepare-dgx-station-host.sh
• test/install-station-dgx-os.test.ts

---

📝 Walkthrough

Walkthrough

The DGX Station release classifier now keys OTA detection on DGX_OTA_VERSION, supports GB300 workstations missing DGX_OTA_PRETTY_NAME, and retains supported-version and factory-image validation. Tests add OTA-upgraded fixtures plus accepted and fail-closed classification cases.

Changes

DGX OTA classification

Layer / File(s)	Summary
OTA-aware release classification scripts/prepare-dgx-station-host.sh	dgx_station_release_profile recognizes OTA history through DGX_OTA_VERSION, validates DGX identity and supported OTA versions, and removes the redundant no-OTA OTA-version rejection.
OTA release classification tests test/install-station-dgx-os.test.ts	Adds an OTA-upgraded GB300 release fixture and tests for supported classification, invalid workstation identity, and an unsupported OTA version.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• NVIDIA/NemoClaw#7108: Modifies the same DGX Station release classification flow and related marker tests.
• NVIDIA/NemoClaw#7126: Updates the same release identity validation and test coverage.
• NVIDIA/NemoClaw#7132: Adds a force-install override interacting with this release qualification boundary.

Suggested labels: area: install, bug-fix, platform: dgx-station

Suggested reviewers: senthilr-nv, ericksoa

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 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: accepting OTA-upgraded DGX OS GB300 hosts without the image marker.
Linked Issues check	✅ Passed	The code and tests implement #7224 by recognizing marker-less OTA-upgraded GB300 DGX OS hosts while fail-closing unsupported lineages and versions.
Out of Scope Changes check	✅ Passed	The changes stay focused on release classification and test coverage for the reported OTA-upgrade case.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/station-express-ota-upgraded-marker

---

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

[github-code-quality]

Code Coverage Overview

Languages: TypeScript

TypeScript / code-coverage/plugin

The overall coverage remains at 96%, unchanged from the main branch.

TypeScript / code-coverage/cli

The overall coverage in the fix/station-express-... branch remains at 80%, unchanged from the main branch.

Show a code coverage summary of the most impacted files.

File	main 3c4ea5a	fix/station-express-... 17dda63	+/-
src/lib/sandbox...rce-identity.ts	91%	91%	0%
src/lib/state/m...ock-identity.ts	95%	95%	0%
src/lib/state/m...-acquisition.ts	84%	89%	+5%
src/lib/domain/.../connect-env.ts	89%	97%	+8%

---

_{Updated July 20, 2026 07:03 UTC
Code Coverage is in Public Preview. Learn more and provide us with your feedback.}

[github-actions]

PR Review Advisor — Informational

Advisor assessment: Informational / high confidence
Next action: Review the warnings below.
Findings: 0 blockers · 1 warning · 0 suggestions
Status: Canonical ledger: 0 blocker(s), 1 warning(s), 0 suggestion(s).

Model lanes

• GPT-5.6 Terra (primary): Completed · high confidence · 0 blockers · 1 warning · 0 suggestions
• Nemotron 3 Ultra (second opinion): Completed · high confidence · 0 blockers · 0 warnings · 0 suggestions
• Model comparison: normalized findings differ; normalized E2E selections match; Nemotron reported the same number of blockers, 1 fewer warning, the same number of suggestions.

Nemotron output stays in workflow artifacts and does not change the assessment above.

E2E guidance

Advisory only. E2E / PR Gate selects and runs jobs independently.

Recommended E2E: None

1 warning · 0 suggestions

Warnings

Warnings do not block.

PRA-1 Warning — Document the marker-less OTA compatibility fallback's source and retirement condition

• Location: scripts/prepare-dgx-station-host.sh:202
• Category: docs
• Problem: The new fallback accepts a marker-less OTA release format based on the adjacent assertion that OTA upgrades never add DGX_OTA_PRETTY_NAME, but the repository does not identify an authoritative vendor format source or when this compatibility path can be removed.
• Impact: Future changes to the external DGX release-file format may leave an obsolete or overbroad eligibility exception in the installer allowlist without a documented basis for maintaining or retiring it.
• Recommendation: Add a concise reference to the authoritative vendor release-format/qualification evidence and a removal or revalidation condition next to this fallback.
• Verification: Read the comment and fallback at scripts/prepare-dgx-station-host.sh:202-210 and search repository documentation for a cited vendor contract or lifecycle condition for marker-less DGX_OTA history.
• Test coverage: The checked-in fixture test at test/install-station-dgx-os.test.ts:382-409 covers the current marker-less format; retain it and add any contract fixture needed when the cited vendor format is revised.
• Evidence: scripts/prepare-dgx-station-host.sh:202-210 introduces the compatibility fallback and states that OTA upgrades never add the identity field. test/install-station-dgx-os.test.ts:382-409 supplies a current consumer-facing regression fixture, but does not establish the external metadata contract or retirement condition. The linked issue contains no maintainer decision or external contract citation; its root-cause claim is untrusted context.

Workflow run details

This automated review informs maintainers. Warnings and suggestions do not require a response. A maintainer decides whether to merge.