Alan-Jowett / sonde / 23509802521

82%

push

github

fix(gateway): send PEER_ACK for duplicate PEER_REQUEST (#466)

* fix(gateway): send PEER_ACK for duplicate PEER_REQUEST with matching PSK

Spec (GW-1218):
- AC4: duplicate PEER_REQUEST with matching PSK still gets PEER_ACK
- AC5: duplicate with different PSK silently discarded

Design (§17.5):
- Step 6 updated: duplicate with matching PSK proceeds to PEER_ACK
  instead of discarding

Validation:
- T-1218a: duplicate with matching PSK receives valid PEER_ACK
- T-1218b: duplicate with different PSK is discarded

Code (engine.rs):
- On duplicate, look up existing node, compare PSK
- Matching PSK: log \duplicate_ack_resent\, fall through to PEER_ACK
- Mismatched PSK: log \duplicate_psk_mismatch\, return None

Fixes #465.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* fix: update tests for duplicate PEER_REQUEST with matching PSK

Existing tests expected silent discard on duplicate node_id, but
GW-1218 AC4 now requires PEER_ACK for duplicates with matching PSK.
Updated assertions in peer_request_duplicate_node_id and
t_1216_duplicate_node_id_rejected.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

* docs: align GW-1216 and T-1216 with duplicate PEER_REQUEST behavior

GW-1216: renamed to 'Node ID duplicate handling', updated to allow
PEER_ACK on duplicate with matching PSK, discard on PSK mismatch.

Design §17.5 step 6: renamed from 'Node ID uniqueness' to
'Node ID duplicate handling'.

T-1216: updated to test both matching-PSK (gets PEER_ACK) and
different-PSK (silently discarded) cases.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

---------

Co-authored-by: Alan Jowett <alan.jowett@microsoft.com>
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

6 of 6 new or added lines in 1 file covered. (100.0%)

21541 of 25177 relevant lines covered (85.56%)

163.98 hits per line