Say why the stable route proof failed #6
Loading…
Reference in a new issue
No description provided.
Delete branch "routing-proof-reason"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Follow-up idunn-deploy-provenance:follow_up:chat-mind-20261113T162719900Z-t62 (live web deploy tx-216475c6 failed only with 'the Routing phase passed its deadline').
advance_routingjournals each failedprove_stable_routeonce per distinct reason (existingreport_once/ReportOnce, cleared on recovery): silent (nothing answered), refused (HTTP status line, etc.), presence not active, authority disagreement, challenge-binding failures. Texts name states and codes only.last_error("; its last attempt failed: ..."), so the abort record andidunn status --commandshow the reason instead of only the clock.statusalready renderedlast_erroras 'waiting on' while the phase is live.Tests: unit tests for each failure class text, report-once and clear, and the deadline abort carrying the proof failure. 44 route_proof-filtered tests fail identically on main and this branch under uid 985 (environment); not caused by this change.
🤖 Generated with Claude Code
Soul (two passes,
a484a7eand88e2f24): pass with findings, tests-only fix verified, mutations M1/M5/M7 now killed. Open, accepted: the continuity test pins text stability, not print count, so a mutant that prints every tick while keeping the text stable survives (M6c); dedup is the existing report_once owner, tested at its other call sites. Also noted: installs/driver errors in advance_routing are not journalled (status shows them), dedup is same-as-last not per-distinct-reason, and the reason text prints twice (ChallengeFailure Display plus source, predates this PR). Merging as steward; these go to a follow-up.