From 1ee14ce1666e47e3ce1f620a79bf9ce4dc86bb66 Mon Sep 17 00:00:00 2001 From: kisfenyo Date: Mon, 21 Sep 2026 22:35:42 +0200 Subject: [PATCH] =?UTF-8?q?Update=20night:=20the=20final=20PROGRESS=20line?= =?UTF-8?q?s=20=E2=80=94=20teardown,=20documents,=20CI=20green=20by=20id?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Both CI runs confirmed by job id: app-catalog-felhom.eu job 830 and felhom.eu job 870, each 'completed' / 'success', matched on head_sha. unproven.py --summary did not move and that is stated rather than left for the reader to notice. Co-Authored-By: Claude Opus 5 Claude-Session: https://claude.ai/code/session_0159rPz1ZhFKsS53msqPYxtS --- documentation/audits/update-night-2026-09-21/PROGRESS.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/documentation/audits/update-night-2026-09-21/PROGRESS.md b/documentation/audits/update-night-2026-09-21/PROGRESS.md index 55252fd4..c2ac0a46 100644 --- a/documentation/audits/update-night-2026-09-21/PROGRESS.md +++ b/documentation/audits/update-night-2026-09-21/PROGRESS.md @@ -55,3 +55,7 @@ A resuming session reads THIS FILE FIRST and never repeats a finished step. | 22:20 | **Bonus proof across a REAL power cut** | The boot sweep met the held `glance` after an unclean shutdown and **deliberately left it alone**: „is a boot orphan by intent but is HELD … NOT starting it; whatever is holding it owns its recovery". §6.1's three-unattended-paths claim, proven across a power cut | 25-B5-the-two-early-cuts.md | | 22:22 | **mealie v3.20.1 -> v3.27.0** (db-postgres) | **PROVEN** — the edge that failed twice on instrument problems, walked cleanly on the third | apps/mealie/ | | 22:23 | **PHASE 1+2 COMPLETE** | **21 edges attempted: 14 proven, 3 failed, 4 inconclusive.** Ten of the fourteen printed a verbatim migration line. Up from the **three** apps this project had ever measured | summarise.py | +| 22:27 | **PHASE 5 — teardown, three layers plus Gitea** | **Machine:** every throwaway app removed through the product; three with an `HDD_PATH` refused at „remove with data" (R-442, correct) and removed keeping it; registry, drill images and drill volumes gone BY NAME, never pruned; `controller.yaml` restored and `git.repo_url` **read back as the LIVE catalog with an empty token**; only `felhom-controller`, `filebrowser`, `traefik` left. **Host:** no harness LXC was created, so none was destroyed; guest 9201 untouched. **Hub:** nothing provisioned; floor 0.261.0. **Gitea:** drill repo KEPT, private, reset to live `main`; **every `image:` line IDENTICAL** | teardown/ | +| 22:28 | **The teardown found the night's LAST defect** | A `navidrome` container the product's own removal left behind, resurrected by Docker's restart policy at a power cut, invisible to every sweep keying on `deployed`. Removed by name. **R-626** | 26-removed-app-came-back.txt | +| 22:4x | **Documents, register, CI** | 12 new rows (register **303 -> 315**), 8 existing rows updated, `09` §3b answered for Q2–Q6, §6.4 re-costed, §6.5 (the drill method) added, limitation 8 replaced, capability map widened, rotation **2 -> 16** fully-walked apps, STATUS morning note. **CI green by id: catalog job 830, felhom.eu job 870, both `success`** | felhom.eu@8d786f7940b0 | +| — | **`unproven.py --summary` did NOT move** | 55 claims, 35 not walked — unchanged. Tonight measured the UPDATE arc, which that file tracks as one claim already marked walked; no claim's status changed, so no number moved. Stated because the checklist asks | scripts/unproven.py |