feat(failure-inject): nccl-hang wraps fr_parser (M11 cf #1/#2)

[trilamsr]

Summary

• Replace the ErrPending stub at tools/failure-inject/ncclhang/ with a deterministic wrapper over module/pkg/nccl/fr_parser.Synthesize. Output is one of the canonical M11 hang fixtures (nccl-2.29.x-hang / nccl-2.30.x-hang), selected by --seed mod 2; bytes round-trip through frparser.Parse and a re-synthesize is byte-identical — closes M4b carry-forward ci(deps): bump the gh-actions group with 5 updates #1.
• Pin the new SHA in tools/failure-inject/testdata/golden.sha256 so chaos.yml's harness-determinism job (matrix linux/amd64 + linux/arm64) replays the same argv on both arches and enforces cross-arch SHA equality — closes M4b carry-forward Bump the gh-actions group across 1 directory with 4 updates #2.
• Flip ⧗ → ☑ on the two M4b functional rubrics (round-trip, safe-opcodes) and the M4b determinism non-functional rubric, plus the M11 synthetic-fixture-generator rubric. Remove the failure-inject nccl-hang follow-up from docs/followups/M4b.md and from M11's carry-forward list.

Root cause

M4b shipped at v0.1 with the nccl-hang subcommand stubbed (ErrPending, exit 70) because pkg/nccl/fr_parser/synthesize.go was still pending under M11. M11 landed the synthesizer plus the canonical hang fixtures (fixture229Hang, fixture230Hang) in module/pkg/nccl/fr_parser/. The CLI shim was carry-forward — this PR is the wiring.

What's in the diff

• tools/failure-inject/ncclhang/ncclhang.go — Options{Seed uint64}; Run selects a hang variant by Seed % len(hangVariants), calls FixtureSpec.Bytes() (which delegates to frparser.Synthesize), writes to w. ErrPending deleted; ctx.Err() honoured before any write.
• tools/failure-inject/main.go — pass Options{Seed: *c.flagSeed} through to ncclhang.Run; drop the errors.Is(err, ncclhang.ErrPending) → exit 70 branch.
• tools/failure-inject/ncclhang/ncclhang_test.go — RED → GREEN: TestRun_RoundTrip (synthesize → parse → re-synthesize byte-identical), TestRun_SeedDeterminism (same seed → same bytes, 4 seeds), TestRun_SafeOpcodesOnly (delegates to frparser.Parse as the safe-opcode oracle — a naive byte scan false-positives on opcode bytes inside SHORT_BINUNICODE string literals), TestRun_CtxCancelled.
• tools/failure-inject/main_test.go — replace TestRun_NCCLHangReturnsNotImplemented with TestRun_NCCLHangRoundTrip + TestRun_NCCLHangSeedDeterminism so the contract is pinned through the actual argv path too.
• tools/failure-inject/testdata/golden.sha256 — add failure-inject --seed=0 nccl-hang → e6f49920…. The existing TestRun_GoldenSHA loop in main_test.go and the Golden SHA pin step in chaos.yml pick it up automatically.
• docs/MILESTONES.md — flip §M4b rubrics ⧗ → ☑ (round-trip, safe-opcodes, cross-arch determinism) and §M11 synthetic-fixture rubric; trim carry-forward list.
• docs/followups/M4b.md — mark the nccl-hang entry closed with the wiring-PR pointer.
• tools/failure-inject/README.md — add a nccl-hang section; remove nccl-hang from carve-outs (now only pod-evict --allow-cluster-write carves).
• module/receiver/ncclfrreceiver/README.md — replace stale tracecore failure-inject invocation with the actual go run ./tools/failure-inject path.

Test plan

• go test -race -count=1 ./tools/failure-inject/... — green (4 packages).
• (cd module && go test -race -count=1 ./pkg/nccl/fr_parser/...) — green (no semantic change here, gate against accidental drift).
• go build ./... && (cd module && go build ./...) — clean.
• Pre-commit gates: golangci-lint, go vet, go mod verify, attribute-namespace-check — all 0 issues.
• End-to-end determinism: failure-inject --seed=0 nccl-hang | sha256sum reproduces the pinned SHA (e6f49920…) twice in a row.
• Seed variance: --seed=1 produces a distinct SHA (2788a726…); --seed=42 (42 mod 2 = 0) matches --seed=0 per the documented modulo mapping.
• failure-inject nccl-hang --help documents --seed and --out and the round-trip-through-fr_parser purpose.

Self-grade

A+: round-trip green, determinism golden-SHA pinned, safe-opcode set verified via parser oracle, cross-arch SHA equality wired into existing chaos.yml matrix, MILESTONES.md flipped on four ⧗ rubrics, M4b.md follow-up closed with a pointer, doc drift swept.

tools(failure-inject): `nccl-hang` subcommand now produces parseable byte-deterministic NCCL FlightRecorder bytes via `pkg/nccl/fr_parser` (was a stub returning `ErrPending`). `--seed` flag selects variant + deterministic synthesis; cross-arch SHA enforced in `chaos.yml` (linux/amd64 + linux/arm64). Closes M4b carry-forward #1 + #2.

[trilamsr]

Grade: A

PR #474 ships M4b carry-forward #1/#2 closure: CLI wraps frparser.synthesize, emits canonical hang-dump pickle fixtures, determinism pinned on both linux/amd64 + linux/arm64 via golden SHA matrix. Four rubrics flipped ⧗→☑ across M4b + M11; all closures are legitimate.

Findings

tools/failure-inject/main_test.go:58: 🔵 nit: --seed flag help text is minimal ("Deterministic PRNG seed.") and doesn't explain the hang-variant modulo mapping. The mapping is well-documented in README + code, so this is comment clarity only — not a functional defect. Consider: "Selects which NCCL hang fixture variant [0=2.29.x, 1=2.30.x]; seed % variant_count." or point to README.

Everything else checks: Seed modulo distribution is Fair (2 variants via idx = Seed % 2; tests cover seeds 0/1/42); parser-as-oracle for TestRun_SafeOpcodesOnly is smart (avoids false-positives in string literals); golden SHA e6f49920... pinned and replayed on both arches; round-trip + determinism + safe-opcodes + cancellation tests all land; zero semantic drift in pkg/nccl/fr_parser/; README sweep is load-bearing.

Simplification sweep

Doc comment at line 3 of ncclhang.go names the contract twice ("byte-deterministic for a fixed Options" at L6, then "two calls with equal Options" at L8). Rephrase L6-L9 to one stronger statement and lose the repetition. Similar: MILESTONES.md M4b status line uses both "wrapped over" and "per the carry-forward closure" — first is clearer. Cut the second.

VERDICT

A — Ship. Carry-forward closures legitimate, test coverage strong, no blockers. Fix the nit before or after merge (comment flavor only).