[cpp-RCP-N2-06] HARA hazard H-001 rated ASIL-B but its S3/E4/C2 rating maps to ASIL-C #69

Location: HARA.md — hazard table row H-001 and its ASIL decomposition rationale.

Finding: Row H-001 ("loss of request delivery to a safety-critical endpoint") is rated Severity 3 / Exposure 4 / Controllability 2, and assigned ASIL-B. Per the standard ISO 26262-3 severity/exposure/controllability-to-ASIL determination table, an S3/E4/C2 combination maps to ASIL-C, not ASIL-B — the document's own rationale text even states plainly that no ASIL decomposition into lower-integrity elements is being claimed to justify the reduction. Since H-001 is the hazard the document's own "Target ASIL: ASIL-B (derived)" line is built on top of, and since every downstream safety goal in the request-delivery/watchdog/liveness chain (SG-001/SG-003/SG-007) is scoped to whatever integrity level H-001 carries, an understated H-001 propagates an understated rigor target through the rest of the safety case.

Recommendation: Re-classify H-001 as ASIL-C, or add and justify a proper ISO 26262-9 ASIL decomposition (with genuinely independent redundant elements) that legitimately allocates ASIL-B to the implementing mechanisms while the hazard itself remains ASIL-C. Propagate the corrected integrity level to SG-001/SG-003/SG-007 and to the document's stated target ASIL. While doing this pass, it's worth double-checking the other rows for the same kind of table-lookup error in either direction (a couple of the other rows appear to be conservatively over- rather than under-rated, which is lower-risk but should be confirmed as intentional).

Verification note: I cross-checked the S3/E4/C2 → ASIL-C mapping against the standard ISO 26262-3 severity/exposure/controllability table (this is an ISO 26262 methodology question, not a claim about the OPEN Alliance TC18 spec). I'm reasonably confident in this reading, but ASIL determination tables have known edge-case interpretation disputes in some organizations' internal guidance, so a second reviewer with the actual ISO 26262-3:2018 table in hand should confirm before treating this as a hard blocker.

Filed from the 2026-07-29 ecosystem audit register; independently re-verified against current HEAD before filing.