From 24e4039eb6af506623e5c50cae81eb12b9738471 Mon Sep 17 00:00:00 2001 From: Levi Neuwirth Date: Thu, 20 Aug 2026 17:31:13 +0200 Subject: [PATCH] docs(lane): record the one rd_precondition failure, without a cause The row rd_precondition_validates_the_whole_conformance_set failed once in a sweep-crdt run on 2026-08-20 and passed on the two sweeps after it. The message was not captured, so nothing here explains it --- the occurrence is recorded and the diagnosis is not. Also withdraws a mechanism I offered for it. I described the test as spawning 46 concurrent stubs under load; it runs 45 stubs SEQUENTIALLY plus one intentional nonexistent-path spawn probe, so there is no concurrency to be pressured and 46 was a miscount. Thirty consecutive user-run repetitions at load ~10.5 --- 1,350 stub executions --- did not reproduce it. Records the standing instruction that a recurrence must capture the exact case and error before anyone theorises again. Co-Authored-By: Claude Opus 5 (1M context) Claude-Session: https://claude.ai/code/session_016bqGA6s9tTUFzYpbeW3tai --- docs/active-work.md | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/docs/active-work.md b/docs/active-work.md index f7cf26c..6460507 100644 --- a/docs/active-work.md +++ b/docs/active-work.md @@ -284,6 +284,23 @@ waits for a signal that is not coming. test is crdt-gated and the macOS jobs build without `crdt`, while `Test (crdt)` is `runs-on: ubuntu-latest`. Recorded as an open gap, not closed. +- **ONE UNEXPLAINED FAILURE, 2026-08-20, AND NO MECHANISM IS RECORDED + FOR IT.** The row was + `rd_precondition_validates_the_whole_conformance_set`. + The row failed once in a `sweep-crdt` run in the `pmacs-mapping-gen` + worktree and passed on the two sweeps after it. **The failure message + was not captured**, so there is nothing to reason from — which is + exactly why no cause is asserted here. + - **NOT REPRODUCED**: 30 consecutive runs at load ~10.5, 1,350 + successful stub executions, no recurrence (user-run, 2026-08-20). + - **A "concurrent spawn pressure" explanation was offered and is + WITHDRAWN as false on its facts.** The test runs its **45 stubs + SEQUENTIALLY**, plus one intentional nonexistent-path spawn probe. + There is no concurrency to be pressured, and 46 was a miscount of + 45-plus-a-probe. + - **ON RECURRENCE, CAPTURE THE EXACT CASE AND ERROR** before + theorising. A single uncaptured failure supports no diagnosis, and + the sequential design means the failing case is identifiable. - **What shipped:** `scripts/check-sigint-deliverable` (validated `(status, token)` pair, 0 safe / 1 ignored / 2 error), a gate guard that refuses before any stage with no override, a target test that