soak drill COMPLETE: 7 phases, 4 findings, the observer found the worst one (R-414)

Overnight soak 22:39->06:10 CEST. demo-hp the victim, demo-felhom the untouched observer.
No production code, no golden, no version bump. Report at
documentation/audits/DRILL-soak-2026-08-31/REPORT.md.

VERDICTS: 1 lock-collision FAIL, 2 guard-interactions PASS-with-one-defect, 3 R-357 PASS,
4 proof-edges PASS, 5 mutated-cycle PASS, 6 observer FAIL, 7 teardown PASS.

R-414 - THE MOST VALUABLE FINDING, AND ONLY AN UNTOUCHED BOX COULD HAVE FOUND IT. On
demo-felhom the nightly proof fired for the first time unattended at 05:30 and REFUSED:
"nowhere to restore to - nincs regisztralt adatmeghajto". Cause established, not inferred:
storage_paths is EMPTY, so there is no path to put a scratch on. It will fail this way every
night forever with only a WARN, and because the error path reaches no verdict,
last_proof_result stays ABSENT - which is also what a pre-0.231.0 controller sends. The hub
cannot tell "never ran" from "not deployed": the StatsKnown trap one level up. The box is
NOT unprotected; its off-site backup ran fine in 46.9s. It is the PROOF that cannot run.

R-411 - measured, not reasoned: restic stats TAKES A LOCK; a customer full-restore runs it
while holding no acquireRunning; the integrity check is therefore not blocked, meets that
lock and escalates to unlock --remove-all. The sampler caught "restore ..." and
"unlock --remove-all" in the SAME sample. Contained: the check was classified unreachable,
not damage, so no false alarm.

R-412 - CORRECTED from HIGH to LOW. I filed it on a mechanism I had not finished measuring.
The off-site run has its own pre-push dump leg, so a hollow unit is REPAIRED before it
ships - proven on two apps and confirmed by pulling the snapshot back out of the store.
What survives is a narrow race, plus a success line over a backup holding no data.

R-413 - the R-87 proof caught a product-produced hollow snapshot unattended, and the
nightly job fired on its own schedule at 05:30 for the first time (bentopdf PASSED on
9d002b38 in 2.315s). Both were listed "not yet live-validated" yesterday.

R-403 mirror guard PROVEN live, with a negative control: it fired when a unit was hollow
("The copy was PRESERVED rather than replaced with an empty one") and skipped 0 legs at
teardown when every unit was sound.

R-357 PASS at last, six days owed: a real full filesystem, refused BEFORE StopStack, app
never stopped, live data byte-identical, and it worked once the space came back.

Phase 4 built the false-alarm control the whole R-87 design rests on: bentopdf is the only
template of 53 with neither a database nor a named volume. It passes silently.

EIGHT of my own instrument errors are named in the report, each caught by its own control -
including a time guard that fired an injection four hours early, and filing R-412 at the
wrong severity.

Teardown clean on all three layers of both boxes; both healthy on 0.231.0.
OWED: a golden for 0.231.0, and a decision on keeping bentopdf.

documentation/backlog/OPEN-ITEMS.md

@@ -597,6 +597,7 @@ class (an image `VOLUME` at an unmounted path) is still live — `immich-server`
 | **R-411** | **A background job DELETES the lock of a live customer restore, and logs it as a crash that did not happen.** MEASURED on demo-hp 2026-08-31 during the overnight soak, through the product's own endpoints — this is R-408's consequence, which until tonight had only been reasoned about. **The chain, every step observed:** (1) a customer full-restore runs `OffboxRestorePrepareFull` → `restic stats`, and **`restic stats` TAKES A REPOSITORY LOCK** (clean-room test: nothing else running, 4x stats, sampler reads `locks=1`); (2) the restore holds `opRunning` but **NOT `acquireRunning`** (R-408), so the integrity check is not blocked and runs concurrently; (3) the check meets that lock, and `resticStep` escalates to **`unlock --remove-all`** — caught by the argv sampler at **20:50:51 with `restore 3c11059b --target …` and `unlock --remove-all` in the SAME sample**; (4) the log says *"cleared a stale exclusive lock left by a previous crash (single-writer repo)"* — **there was no crash**, and `resticStep` cannot know there was, because it fires on ANY `repository is already locked`. **THE CUSTOMER-FACING CONSEQUENCE WAS CONTAINED, and that is R-359's guard working:** the check returned `ok:false` in 6.7 s and was classified **Unreachable, NOT damage** — *"the check could not run to a verdict (other) — NOT reported as damage"* — so no `backup_integrity_failed` and no customer mail. Due-ness was not advanced either, so it retries. **What is NOT contained:** a live operation's lock is deleted by a background job; the single-writer premise `resticStep`'s own comment rests on is false in this pairing; and that night's integrity check silently did not verify the store, with only a WARN. **The opposite direction is FENCED and was measured too:** five restores fired into a running check at 5/15/25/35/40 s offsets were ALL refused by `restoreOpBlocked` (`offbox_handlers.go:358`), zero restic invoked — so the hazard is reachable only restore-FIRST. | **OPEN — MEDIUM** | R-408, R-407, R-359 | Decide ONE way, and R-408 is the same decision: either `RestoreOffboxScratch` (and the full-restore preparation) takes `acquireRunning`, or `resticStep`'s escalation stops claiming a crash it cannot verify and refuses instead of removing. **Pin whichever is chosen with a test that reproduces this pairing** — a unit test on `resticStep` alone cannot see it. Evidence: `audits/DRILL-soak-2026-08-31/phase1-lock-collision/`. | CC |
 | **R-412** | **A recovery unit lost DURING an off-site run — after its own dump leg, before its push — is shipped hollow and the run reports success.** **CORRECTED 2026-09-01 04:22, and the first wording of this row OVERSTATED it.** As first filed it claimed the hollow unit sat in the store for a whole cycle because "the volume-dump leg runs on the backup schedule, not on capture". **That is wrong, and measuring it overnight is what showed it:** the off-site run has its OWN pre-push dump leg — *"Stopping calibre-web for safe volume dump"*, *"Volume dump: calibre-web/calibre-web_calibre_web_config -> 877.5 KB"* — so a unit that is hollow when a run starts is **REPAIRED before it is pushed**. Proven twice: `opengist` (2026-08-31 21:0x) and `calibre-web` (2026-09-01 04:15) both went in hollow and came out complete, and the snapshot pulled back from the store (`6fee3b5a`) holds the volume tar and all 17 userdata files. **WHAT REMAINS REAL, and it is narrower:** the one hollow snapshot that DID reach the store (`35ba9fe7`, opengist) was created when the unit was destroyed **inside** a run that had already completed opengist's dump leg — so the push shipped what the capture had just rebuilt empty, and logged *"backed up opengist (… 0 mandatory path(s))"*, **a success line over a backup holding none of the app's data**. That race is real, it was observed, and the success wording is wrong either way. **The R-403 mirror guard holds throughout** — proven live: *"unit leg SKIPPED … The copy was PRESERVED rather than replaced with an empty one"*, secondary byte-identical. | **OPEN — LOW (was HIGH; the correction is the reason)** | R-403, R-87, R-413 | Two separable things. (1) The success line: a per-app push that carried no dumps and no tars should not read as a plain success — that is a wording fix in the run's own reporting, not a new guard. (2) The race: decide whether the push should re-read the unit it is about to send, or whether the window is small enough to accept. **Do NOT guard the capture** (08 §8.2). Evidence: `audits/DRILL-soak-2026-08-31/phase2-guard-interactions/` and `phase5-mutated-cycle/09-what-reached-the-store.txt`. | CC |
 | **R-413** | **R-87's proof caught a naturally-produced hollow snapshot, end to end, unattended — the validation yesterday's session could only do with a declared hand-built fixture.** 2026-08-31 soak, demo-hp. After R-412's chain left `opengist`'s newest off-site snapshot hollow, the nightly proof rotated to it and returned **`verdict:"fail"`, `reason:"volumes_expected_none_captured"`, missing `opengist_data`**, logged *"READABLE AND EMPTY — the store is not damaged; the backup does not contain this app's data"*, and pushed **one** `offsite_proof_empty` at severity `error`. The four apps ahead of it in the rotation all passed, so the discrimination is real and not a constant fail. **This is recorded as a row rather than only as a report line because it upgrades a claim:** the capability map's R-87 row cites a CONSTRUCTED failing case; it can now cite a natural one. | **CLOSED 2026-08-31 — the claim it upgrades is recorded** | R-87, R-412 | Nothing to build. When the capability map is next touched, cite this instead of the constructed case. | CC |
+| **R-414** | **The nightly off-site PROOF is INERT on a box with no registered data drive, every night, and the only signal is a WARN in the log.** FOUND 2026-09-01 by the soak's UNTOUCHED observer, `demo-felhom`, on the first unattended run of the job — which is exactly what an untouched box was for. At 05:30 CEST the job fired, picked `opengist`, and refused: *"proof: opengist has nowhere to restore to: nincs regisztralt adatmeghajto, ezert nincs hova visszaallitani"* — `offboxRestoreScratchDir`'s R-252 refusal. **CAUSE ESTABLISHED, not inferred:** that box has **zero** registered storage paths (`storage_paths: []`), so there is no non-network schedulable path to put a scratch on. The same absence explains its `tier2-backup` completing in **3 ms** — a no-op with no second drive to mirror to. **The box is NOT unprotected:** its off-site backup ran normally in 46.9 s, because units live on the system data path, which needs no registration. **It is the PROOF that cannot run.** **WHY IT IS WORSE THAN A FAILED RUN:** the Err path reaches no verdict, so `RecordProofVerdict` is never called, so `last_proof_result` stays **ABSENT** — and absent is also what a controller older than v0.231.0 sends. **The hub therefore cannot tell "never ran" from "not deployed"**, which is the StatsKnown trap the field was explicitly designed to avoid, reappearing one level up. It will fail this way every night forever with nothing but a WARN. | **OPEN — MEDIUM** | R-87, R-402, R-252 | Decide what a box with no registered drive should do: fall back to the system data path for the scratch (it already holds the units), or record a distinct NOT-APPLICABLE verdict so the hub can tell it apart from never-ran. **Do not leave it as a WARN** — that is the shape R-397 and R-107 both cost a drill. Evidence: `audits/DRILL-soak-2026-08-31/phase6-observer/`. | CC |
 
 <!-- DUE-CHECKS-BEGIN — machine-readable. Parsed by scripts/due_checks_gate.py.
      One row per dated check. The R-number must have a row above. Dates are UTC.