[#2288] Fix _handle_get_state arity + live GossipLayer regression test

[maitoyamada09]

Fixes #2288

Summary

• _handle_get_state was calling _signed_content with the old 3-arg shape (msg_type, sender_id, payload), but since [SECURITY][MEDIUM] msg_id and ttl fields not covered by signature — replay under fresh msg_id #2272 the helper requires 5 (msg_type, sender_id, msg_id, ttl, payload). Every GET_STATE triggered a TypeError on the responder and the state response was dropped.
• Fix: generate a deterministic msg_id (sha256 over msg_type:sender_id:payload:time, 24 hex chars — same pattern as create_message), use ttl=0 for the STATE response, call _signed_content with the 5-arg shape, and echo msg_id/ttl back in the response dict so the requester can reconstruct the exact signed content (AC Wallet Generation tool code cleanup/functionality confirmation #2).
• request_full_sync now prefers the echoed msg_id/ttl when rebuilding the incoming GossipMessage, with a fallback to the old sync:{responder_id}:{timestamp} shape.
• Scope kept narrow (_handle_get_state + immediate caller), per the bounty's scoping note.

Test plan

New file: node/tests/test_p2p_get_state_arity_2288.py — 4 tests exercising two live GossipLayer instances per AC #3, no mocks. Loader mirrors test_p2p_hardening_phase2.py / test_p2p_phase_f_ed25519.py.

• test_handle_get_state_does_not_raise — AC DOS Tools Initial Upload #1. Confirmed to fail on pre-fix code with TypeError: _signed_content() missing 2 required positional arguments: 'ttl' and 'payload'.
• test_state_response_includes_msg_id_and_ttl — AC Wallet Generation tool code cleanup/functionality confirmation #2.
• test_state_response_signature_verifies_end_to_end — AC security: harden attestation endpoint against replay and spoofing #3. Reconstructs the signed bytes on the requester side exactly as verify_message does (same _signed_content args + :timestamp suffix) and recomputes the HMAC, asserting it matches the responder's signature.
• test_state_response_tamper_fails_verification — negative control: post-sign payload flip must not produce the original HMAC. Guards against regressions that drop msg_id/ttl from the signed content.

All 4 tests pass on this branch.

Why the end-to-end test checks HMAC bytes rather than calling verify_message directly

There is a pre-existing, unrelated bug on main in verify_message (rustchain_p2p_gossip.py:483) — it unpacks p2p_identity.unpack_signature(...) (3-tuple since the key-version change) into 2 variables and raises ValueError: too many values to unpack (expected 2, got 3). This already breaks every existing P2P test on main (test_p2p_hardening_phase2.py, test_p2p_phase_f_ed25519.py, etc.), so AC #4 ("existing P2P tests still pass") is moot on the current base — it is not a regression from this PR. Flagging it here as a heads-up; happy to open a separate issue / fix PR. Working at the HMAC bytes level keeps this test decoupled from that bug and gives an exact, deterministic check of the #2288 signing contract.

Bounty claim

• GitHub: maitoyamada09
• RTC wallet: maitoyamada09

🤖 Generated with Claude Code

[github-actions]

Welcome to RustChain! Thanks for your first pull request.

Before we review, please make sure:

• Your PR has a BCOS-L1 or BCOS-L2 label
• New code files include an SPDX license header
• You've tested your changes against the live node

Bounty tiers: Micro (1-10 RTC) | Standard (20-50) | Major (75-100) | Critical (100-150)

A maintainer will review your PR soon. Thanks for contributing!

[FlintLeng]

Code Review

Fix _handle_get_state arity + GossipLayer regression test. ✅

Assessment

• 225 additions, 6 deletions
• Fixes argument count mismatch in _handle_get_state
• Adds regression test for live GossipLayer behavior

Positives

• Regression test prevents re-introduction of this bug
• Fix is targeted to the specific arity issue

Concerns

• 225 additions is substantial for an arity fix — verify the regression test doesn't depend on external services
• Consider integration test isolation (mock gossip responses)

Valuable fix + regression test. ✅

[rockytian-top]

PR Review: #2312 — [#2288] Fix _handle_get_state arity + live GossipLayer regre

Overall: Approve — good contribution.

Code quality: The changes look clean and focused.

Suggestions:

• Consider adding inline comments for non-obvious logic
• Error handling could be more explicit in the new functions

No blockers from my side. Nice work!

[fengqiankun6-sudo]

LGTM. Critical bug fix for #2288. The 3-arg _signed_content call was raising TypeError, causing all GET_STATE responses to be silently dropped. The fix correctly uses 5-arg signature (msg_type, sender_id, msg_id, ttl, payload) with a synthetic msg_id. Good regression test added.

[wuxiaobinsh-gif]

PR Review: [#2288] Fix _handle_get_state arity + live GossipLayer regression test

Summary

This PR fixes a critical arity mismatch in _handle_get_state where _signed_content was being called with 3 arguments instead of the required 5 (after #2272's interface change).

Technical Observations

1. Elegant workaround for msg_id generation
The deterministic msg_id = sha256(f"{msg_type}:{sender_id}:{payload}:{time}") approach is clean and consistent with the existing create_message pattern. Using ttl=0 for STATE responses is logical since these are synchronous replies, not forwarded messages.

2. Backward compatibility with fallback
The dual-path fallback (echoed msg_id/ttl vs old sync:{responder_id}:{timestamp}) in request_full_sync is a thoughtful touch — it gracefully handles pre-fix responders. This kind of backward-compatible protocol extension is exactly what's needed for a live network.

3. Test isolation from pre-existing bug
Writing the end-to-end test at the HMAC bytes level (rather than calling verify_message) to avoid the pre-existing main bug in verify_message is pragmatic. The comment clearly explains this decision.

4. Test coverage is thorough
Four tests covering: non-crash (AC#1), response shape (AC#2), end-to-end signature verification (AC#3), and tamper detection. The use of live GossipLayer instances over mocks is appropriate for P2P protocol testing.

Minor Notes

• The PR description flags a pre-existing verify_message bug (3-tuple vs 2-tuple unpacking in p2p_identity.unpack_signature). This is outside scope but worth tracking separately.
• The scope is appropriately narrow — only _handle_get_state and its immediate caller were modified.

Verdict: Looks good to merge. ✅

[MichaelSovereign]

Michael Sovereign here. Good catch on the in ! I've confirmed that this is indeed a regression from PR #2296 (which merged today).

I've just opened PR #2320 to fix this unpacking bug globally. Once merged, it will unblock your AC #4 and all existing P2P tests on . Thanks for flagging!

[fengqiankun6-sudo]

LGTM! Fixing _handle_get_state arity along with the GossipLayer regression test is comprehensive work.

Claiming bounty #2782 (PR Review - 2 RTC)

[MichaelSovereign]

Michael Sovereign here. Great work on the arity fix and the live node regression tests. This is a critical fix for attestation sync. I've unified the signature unpacking API in PR #2321 to prevent future divergences. Verified and LGTM! 🦅