[yifjiang]

Summary

Reject wrong rank, geometry, symlink, UUID/incarnation and stale descriptors before importing handles; no partial group acceptance.

Review slice D22 of #16748; part of #16747 and #16746. This draft targets codex/kv-reuse-d21-complete-rank-groups for a complete incremental review diff of 364 additions plus deletions, including tests, documentation and extraction glue.

Extraction details:

• The complete rank import implementation replaces the functional TP1-only path; no rank group was served before this gate.

Reuse remains experimental and default off. Retained inventory is advisory, complete compatible rank parts must precede readiness, and native operations retain ownership until quiescence. Supported GPU activation also requires the companion KVCR/native slices and the declared deployment profile.

Validation

All configured pre-commit hooks passed for this exact slice on allocated SLURM. Stack-level CPU regression coverage is summarized on D31; each draft still requires its own applicable CI in the Checks tab. Historical combined-head and hardware results retain their original source scope.

[devin-ai-integration]

⏹️ Dynamo PR CI cancelled — run 41941692280 (attempt 1) on cf67181d3b

Gate checks: ❌ backend-status-check · ✅ deploy-status-check · ✅ dynamo-status-check

Framework	Build
vLLM	⏹️ 1 ✅ 5
SGLang	⏹️ 1

Failure details

No job failed; 2 multi-arch image builds were cancelled mid-build after ~59 minutes each, and backend-status-check fails only because it mirrors them.

• ⏹️ sglang-runtime / Build multi-arch cuda13.0: cancelled in Build and Push Image while still in the linux/arm64 stage (##[error]The operation was canceled. after 3561 s). No earlier failures and the rest of the run continued, so this is not fail-fast; the ~1 h duration suggests a job time limit rather than a manual/concurrency cancel.
• ⏹️ vllm-runtime-efa / Build multi-arch cuda13.0: same pattern, cancelled in Build and Push Image during the linux/arm64 pre_runtime stage after 3549 s.

For agents

{"pr": 18040, "run_id": 41941692280, "run_attempt": 1, "head_sha": "cf67181d3bad248854c458cc5c9fd1b11e762dcd",
 "failures": [
  {"job": "sglang-runtime / Build multi-arch cuda13.0", "job_id": 126964028609, "failed_step": "Build and Push Image", "signature": "cancelled after ~59 min during linux/arm64 build", "tests": [], "log_cmd": "gh run view --job 126964028609 -R ai-dynamo/dynamo --log-failed"},
  {"job": "vllm-runtime-efa / Build multi-arch cuda13.0", "job_id": 126964028828, "failed_step": "Build and Push Image", "signature": "cancelled after ~59 min during linux/arm64 build", "tests": [], "log_cmd": "gh run view --job 126964028828 -R ai-dynamo/dynamo --log-failed"}
 ]}

_{Posted automatically by Devin for run 41941692280. Updated on every full-CI run of this PR.}

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[yifjiang]

/ok to test 436022d

[yifjiang]

/ok to test cf67181