REPORT: the night's agent fixes (R-921, R-1070, R-1010), not released

The alarm step built its link from GITHUB_SERVER_URL, which on this runner
is the in-cluster Gitea address, and from the run NUMBER, which Gitea 1.26
reads as a run id first (felhom.eu /actions/runs/1116 opened run id 1116 =
run number 613). Now https://gitea.dooplex.hu + GITHUB_RUN_ID, and the link
is printed into the job log as ALARM-LINK. Same change in all four repos.

.claude/rules/unprompted-work.md

@@ -44,8 +44,12 @@ it **first** in the morning note. A decision you cannot write in that shape is o
    Throwaway apps only; the standing apps and `bentopdf` stay.
 5. **Evidence off the machine at the end of each phase**, before any revert (R-320).
 6. **One release per repo per session**, with a CHANGELOG entry (controller: with its `MinAgent`
-   line), REPORT overwritten, floor raised to deliver it. **No golden unless a drill or fresh install
-   needs one** (the waiver, R-468). **No `--no-verify`.**
+   line) and REPORT overwritten. **A controller release needs no golden** (operator ruling 2026-10-15
+   11:26, `09` §3 decision 250; decision 7 stands): push the image only when it is ready for EVERY box (the
+   household's update button offers the highest tag); raise the floor, MinAgent declared, only after it ran
+   on Tester 1, demo-hp and demo-felhom through the normal update path and the box walk's checks are green
+   on all three — not at night unless the night brief says so. **Agent releases still stop for the
+   operator's vouch.** The golden is baked weekly and vouched by the operator; CC never vouches. **No `--no-verify`.**
 7. **An enumerated gap becomes a row in the same session — or, if it is small, is fixed in it** (the size rule).
    Prose is not a record.
 8. **Hungarian text is searched with ASCII fragments**, with a positive and a negative control.

.gitea/workflows/gates.yml

@@ -16,7 +16,16 @@
 # A failing run must reach a person — a detector nobody hears is the defect R-29 filed, rebuilt one
 # layer up. That is the last step, and it runs ONLY on failure.
 name: gates
-on: [push]
+# R-1129: `workflow_dispatch` is the manual PROOF that the failure mail can be sent. A push run is
+# unchanged by it: the proof step below is skipped on a push, and `force_alarm` is then empty.
+on:
+  push:
+  workflow_dispatch:
+    inputs:
+      force_alarm:
+        description: "Fail this run on purpose so the alarm step sends its real mail (a proof, not a gate result)"
+        type: boolean
+        default: false
 
 jobs:
   gates:
@@ -55,8 +64,27 @@ jobs:
         env:
           # In-cluster, so the check does not depend on public DNS or the ingress TLS chain.
           GITEA_BASE: http://gitea.gitea-system.svc.cluster.local:3000
+          # R-986: the security-scan gate fetches the PINNED gitleaks here (sha256-checked); no Go, so govulncheck says
+          # NOT CHECKED in CI, out loud. Warn-first either way: the scan lists, it never fails this job.
+          FELHOM_SCAN_FETCH: "1"
         run: cd ws/felhom-agent && python3 scripts/agent_gates.py
 
+      - name: Manual proof - alarm key present, optional forced failure (dispatch only, R-1129)
+        # Runs ONLY on a manual dispatch, never on a push. It prints the key's LENGTH, never the key,
+        # so "CI can see the secret" is an observable rather than an assumption. With
+        # force_alarm=true it then fails on purpose so the alarm step below sends its real mail;
+        # the commit itself stays green, because its push run skips this step.
+        if: github.event_name == 'workflow_dispatch'
+        env:
+          RESEND_API_KEY: ${{ secrets.RESEND_API_KEY }}
+          FORCE_ALARM: ${{ github.event.inputs.force_alarm }}
+        run: |
+          python3 -c 'import os;print("RESEND_API_KEY length: %d" % len(os.environ.get("RESEND_API_KEY","")))'
+          if [ "$FORCE_ALARM" = "true" ]; then
+            echo "force_alarm=true - failing on purpose so the alarm step runs (R-1129 proof)"
+            exit 1
+          fi
+
       - name: Alarm on failure
         # THE POINT OF THE WHOLE THING. Probe P5 measured that a failed run produces NO mail, NO
         # notification row and NO log line from Gitea itself — a red tick in a web UI nobody watches
@@ -70,6 +98,8 @@ jobs:
         if: failure()
         env:
           RESEND_API_KEY: ${{ secrets.RESEND_API_KEY }}
+          # Empty on a push, so a push run's mail is exactly what it was before R-1129.
+          FORCE_ALARM: ${{ github.event.inputs.force_alarm }}
         run: |
           python3 - <<'PY'
           import json, os, sys, urllib.request, urllib.error
@@ -81,13 +111,23 @@ jobs:
 
           repo = os.environ.get("GITHUB_REPOSITORY", "?")
           sha  = os.environ.get("GITHUB_SHA", "?")
-          run  = os.environ.get("GITHUB_RUN_NUMBER", "?")
-          srv  = os.environ.get("GITHUB_SERVER_URL", "https://gitea.dooplex.hu")
+          # R-1135: the link is for a PERSON, so it is the public address and the run ID.
+          # NOT GITHUB_SERVER_URL: the runner registers at the in-cluster address, so that variable
+          # reads http://gitea.gitea-system.svc.cluster.local:3000, which nobody outside the cluster
+          # can open (forced-test mail, job 1997). NOT GITHUB_RUN_NUMBER: Gitea 1.26 reads
+          # /actions/runs/<n> as a run id first, so a run number that is also an older run's id
+          # opens THAT run (measured 2026-10-18: felhom.eu /actions/runs/1116 opened the run whose
+          # id is 1116 = run number 613). The link is printed too, so the job log carries it.
+          run  = os.environ.get("GITHUB_RUN_ID", "?")
+          srv  = "https://gitea.dooplex.hu"
+          print("ALARM-LINK %s/%s/actions/runs/%s" % (srv, repo, run))
 
           body = json.dumps({
               "from": "Felhom CI <monitoring@felhom.eu>",
               "to": ["admin@felhom.eu"],
-              "subject": "[felhom CI] gates FAILED in %s" % repo,
+              "subject": ("[felhom CI] FORCED TEST, not a real failure - gates FAILED in %s"
+                          if os.environ.get("FORCE_ALARM") == "true"
+                          else "[felhom CI] gates FAILED in %s") % repo,
               "text": (
                   "The gate entry point exited non-zero.\n\n"
                   "Repository : %s\n"

.security-accepted

@@ -0,0 +1,18 @@
+# .security-accepted — hits security_scan_gate.py (felhom.eu/scripts, R-986) lists as ACCEPTED, not WARN.
+# One per line: <tool> <id> <where>  # reason. govulncheck: OSV id + module dir; gitleaks: rule@<sha256[:12] of the line> + file.
+# The first scan's verdicts (2026-10-19): felhom.eu documentation/audits/night-2026-10-19/P7/triage.md
+gitleaks generic-api-key@1ad7cc8eb57e internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@46ce5a3eccc4 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@46fe126846b0 internal/escrow/identity_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@56474f1754f5 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@631c53997f7e internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@6e790ef41ef3 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@78334b854242 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@7b88d323a44f internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@b0eeee93cf24 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@b1cb2659967a internal/localapi/escrow_stage_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@b66bec43cc8a internal/escrow/recover_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@e0525160fe03 internal/pbsdr/wrapper_read_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@e442e455c360 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@f23f93f461b2 internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret
+gitleaks generic-api-key@f471b78e0f6e internal/escrow/recover_retained_test.go  # false: a test constant or a decoy value, no live secret

CHANGELOG.md

@@ -1,3 +1,402 @@
+## Unreleased (2026-10-19) — three fixes for the next release (R-921, R-1070, R-1010); two texts made true after 0.160.0
+
+Not released (the night brief of 2026-10-19: agent code + tests + commit only; the release is the operator's word). No
+sudoers, wrapper or config-bundle change.
+
+- **R-921 — `GET /backup/status` says when the host is busy.** New field `host_busy` (`{busy, what}`): whether the
+  host-wide heavy-operation gate is held right now (another guest's backup, the restore test, a night OS step, the
+  stale-lock sweep, a power-off) — the state that makes a `POST /backup` answer 409. Served on every request; absent
+  when the gate is not wired. Controller ≥ 0.335.0 reads it before it stops the household's apps for a tier, so a
+  refused tier costs no stop (the measured case of 2026-10-08: the night OS step, all 15 apps stopped for a refusal).
+  A read, never a reservation. No new route. Test `TestBackupStatus_R921_HostBusy`, red-proved (assignment removed →
+  `free gate: host_busy = <nil>`).
+- **R-1070 — a box with no tunnel set up no longer fails its host OS update.** Docker 29 prints „error: no such
+  object" in lower case, so the tunnel probe's case-sensitive match fell through to `unknown` („could not ask the
+  guest"), and the host health rule failed every box without a cloudflared container (Tester4, 2026-10-15,
+  `os_update_health_failed`). Now the match ignores case and a missing container is a new state, **`absent`** („no
+  tunnel set up on this box"); the host rule reads it as not applicable only when the reading from before the update
+  (Docker answered) also had no cloudflared container. A tunnel that is down, gone or unreadable still fails. The hub
+  alarms only on `not_running`, so `absent` never mails. Tests `TestHostHealthVerdict_R1070_NoTunnelSetUp` (7 rows) and
+  three `TestClassifyTunnel` rows (Docker 29's wording measured on demo-hp), red-proved both ways.
+- **R-1010 — the stale-lock janitor clears a half-removed `vzdump` snapshot.** When the delete fails ONLY with
+  „Failed to find logical volume …_vzdump" (one disk's snapshot already gone — demo-hp 2026-10-14 after a backup ran
+  out of space), it retries once with Proxmox's own `force=1` on the same API call (`DeleteSnapshotForce`; the same
+  token privilege, no sudoers line, no root command — `pvesh usage … -command delete` lists `--force`). Any other error
+  is returned unchanged; a failed forced delete is returned, never a second force. Tests
+  `TestStaleLock_R1010_ForcesOnceOnlyWhenTheVolumeIsGone` (6 rows) and `TestDeleteSnapshotForce_R1010`, red-proved
+  (single plain delete → `calls [plain], want [plain force]`).
+
+- **The security-scan gate (R-986), tooling only:** `agent_gates.py` registers the SHARED
+  `felhom.eu/scripts/security_scan_gate.py` (FAST set: the hook and CI) — govulncheck (called vulnerabilities) and gitleaks
+  on the pushed commits, **warn-first**. `.security-accepted` holds the first scan's 15 accepted hits (escrow and
+  local-API test constants), each with a reason. CI fetches the pinned gitleaks (`FELHOM_SCAN_FETCH=1`). Decoys: the
+  shared 8 plus the gate on this repo, in `test_gate_decoys.py`.
+
+### Earlier on 2026-10-19 (texts only)
+
+- The ceremony-complete log line now says the package uploads at the claim **or at the typed-back confirm** (R-1151);
+  it read „package uploads at the claim" for both orders (seen in demo-hp's journal, 2026-10-19 01:47 UTC).
+- The R-973 comments in `cmd/felhom-agent/main.go` and `internal/localapi/backup_cap.go` named the hub event
+  `agent_capability_degraded`; since hub v0.159.0 (R-1150) a refused backup raises `whole_guest_backup_refused`.
+  Found by the hub helper of the day brief 2026-10-19.
+- **gofmt, R-1160:** `1edfd8e` formats the 14 Go files that were not gofmt-clean (formatting only, `go test ./...`
+  green). New gate `scripts/gofmt_gate.py` (the controller's R-454 shape) in `agent_gates.py`, fast set; its decoys in
+  `scripts/test_gate_decoys.py` plant an unformatted file in a SCRATCH copy (convicted by name), the formatted twin
+  (passes), and the no-Go modes (CI passes out loud, a dev machine is INCONCLUSIVE). Red-proved: with the conviction
+  switched off the FACT case reports a LIVE HOLE. The next release carries it; no release for it.
+
+## v0.160.0 — five fixes: an abandoned or un-typed-back escrow ceremony keeps the old code (R-978, R-1151); a whole-guest backup flood is capped (R-973) and the local tier keeps one point per day (R-1149); an empty partition is not „data" (R-1093) (2026-10-19)
+
+Released 2026-10-19 on the operator's yes (02:48 answers, D17 „one per day, 3 days"). sha256
+`7ebb547f2b1ab93544ba828d3040351b60c4126327a5b85bbcb33f8f2b3dfca1`, config bundle
+`6b06090190ff609af1fe9842447ebf385fd52da4be382bc7ca708b558c8e202f`. The vouch is the operator's.
+
+- **R-1093 — an EMPTY partition no longer asks the typed „data will be erased" confirmation.** `lsblk` prints the
+  parent disk's table type (PTTYPE) on every partition, so `SudoHostOps.InspectDevice` read a blank partition as
+  „device has a partition table". Now lsblk's PTTYPE is set aside only when `blkid -p` on the partition itself proved
+  it read the device and that it is a partition of a table of exactly that type (`PART_ENTRY_NUMBER` +
+  `PART_ENTRY_SCHEME` equal to lsblk's PTTYPE). The partition's own signatures still come from `blkid -p`: a
+  filesystem or a nested table inside it stays data-bearing; a blkid that printed nothing, a whole disk, or a scheme
+  that differs keeps the old, safe reading. **No new command, no sudoers or config-bundle change** (the two probes are
+  the same allow-listed argv). Test `TestInspect_R1093_PartitionVerdicts` (8 rows), red-proved: with the old
+  `if pttype != ""` both empty-partition rows fail `DataBearing=true reason="device has a partition table"`.
+  **Hardened after the same night's review (fail-closed on every path):** `PART_ENTRY_*` is read from the PARENT's
+  table, so lsblk's PTTYPE is set aside only when blkid also **exited 0** — exit 2, an I/O error or a missing binary
+  keep „device has a partition table"; and an `lsblk` that exits 0 with output that does not parse into a device (or
+  lists none) now leaves the device UNPROBED („could not be reliably inspected") instead of „blank" — that hole
+  predates R-1093. Six more rows in the same test, each red-proved (`blkidErr` check removed → „device is blank";
+  `&& parsed` removed → `Probed=true … „device is blank"`).
+- **R-978 — an abandoned second escrow ceremony no longer makes the customer's recovery code useless.** The root
+  ceremony (the FELHOM_ESCROW argv, unchanged: `--upload --output=json`) used to upload the new package before the new
+  code was claimed, and the hub overwrites the current package when the sealed secrets are unchanged — so a customer
+  who started the ceremony again and walked away (or an agent restart in the 10-minute window) was left with a
+  written-down code that opened nothing. Now json mode hands the package back (`CeremonyOutput.PendingUpload`, output
+  **version 2**; text mode still uploads at once) and the daemon uploads it at `POST /escrow/ceremony/claim`, **before**
+  the code is handed out (`EscrowCeremonyConfig.Upload` → `putEscrowBody`, the same PUT and per-host key). A failed
+  upload answers 502 and keeps the code claimable; the TTL and a new start wait while the claim's upload is in flight;
+  no uploader → 503; an output that uploaded by itself or carries no package fails the ceremony. Status `uploaded`
+  now reads false until the claim. The controller is unchanged (it never read `uploaded`; its claim already assumed
+  the package is on the hub when the claim returns). **No sudoers or config-bundle change.** Tests (localapi)
+  `TestEscrowCeremony_R978_*` (abandon by TTL / by a new start / by a restart → the hub's package still opens with
+  code 1; claim stores before handing out; upload failure; in-flight TTL; no uploader; no package) and (cmd)
+  `TestEscrowOptsFromCLI_JSONModeDefersUpload`, `TestDeliverEscrowPackage_JSONModeSendsNothing`,
+  `TestR978_CeremonyUploadIsWiredIntoTheDaemon`. Red-proved: upload-at-create restored → „code 1 no longer opens it
+  (uploads=2)"; the in-flight TTL guard removed → „the TTL voided the code while its package was uploading";
+  `deferUpload: false` → „json mode must defer the upload"; the `Upload:` line removed → „has no Upload". **Residual:**
+  the customer's typed-back check after the claim is browser-only, so „confirmed" here means „claimed" — a customer
+  who opens the code and closes the page without writing it down has replaced the package.
+- **R-1151 — the new recovery code replaces the hub's package only after the household typed it back.** R-978 moved
+  the upload to the claim; but the wizard's typed-back check was the browser's alone, so a household that opened the
+  new code and closed the page without writing it down had already replaced the package their old code opens. Now a
+  claim with `{"defer_upload": true}` (controller ≥ 0.332.0) hands out the code with two word positions
+  (`verify_positions`, drawn by crypto/rand) and keeps the package; the new `POST /escrow/ceremony/confirm
+  {"words": [w1, w2]}` checks the two words (constant time, case and spaces aside) and only then uploads it. Right → 200;
+  wrong → 422, still waiting; one word → 400; nothing waiting → 404; the 30-minute window ran out → 410 and the job
+  reads `unconfirmed_void` (the package dropped, the previous code still works); hub failure → 502, still waiting. Only
+  the two words are kept for the check, never R; a new start, the window and a restart drop them with the package.
+  Status gains `awaiting_confirm` and `confirm_expires_in_sec`. **An old controller (≤ 0.331.0) claims with `{}` and
+  never confirms: it keeps the R-978 order (upload at the claim)**, so its household's ceremony still completes; a new
+  controller on an old agent sees no `confirm_required` and needs no confirm. No sudoers or bundle change. Tests
+  `TestEscrowCeremony_R1151_*` (shown but not typed back — by the window, a new start, a restart — code 1 still opens
+  the hub's package; typing; the old controller's order; hub failure at the confirm) and `TestPickConfirmPositions`.
+  Red-proved: the defer flag ignored → „a code that was shown but never typed back replaced the package: code 1 no
+  longer opens it (uploads=2)" in all three; the word check skipped → „one wrong word: got 200 (uploads=1)".
+- **R-1149 — the local whole-guest tier keeps one restore point per day for 3 days** (operator decision D17,
+  2026-10-19). The primary (local) tier's per-run prune is now `--prune-backups keep-daily=N` (N =
+  `local_backup_retention`, default 3) instead of `keep-last=N` — the same at most 3 archives, but a run of extra
+  backups inside one day replaces only that day's point, so with the R-973 cap a flood cannot push out the earlier
+  days' copies. Additional tiers keep their explicit `keep_last` (the weekly off-site „last 2" is two weeks).
+  `config.BackupTier.PruneSpec()` is the one place the spec is built (runDaemon used to format keep-last itself); the
+  „backup tier armed" log line names it. Tests `TestBackupTierPruneSpec_R1149_LocalKeepsOnePerDay` and the
+  consequence `TestBackupCap_R1149_LocalFloodKeepsOlderDays` (the agent's real spec, a Proxmox-shaped prune model,
+  three nights then a day-long flood → the two earlier nights survive). Red-proved: keep-last for the primary →
+  „primary tier prune = "keep-last=3", want "keep-daily=3"" and „holds 0 copies made before the compromise". No
+  config or sudoers change.
+- **R-973 — the local-API whole-guest backup route has a cadence cap.** Every successful run prunes its tier, so a
+  compromised controller calling `POST /backup` in a loop pushed out every pre-compromise restore point (local
+  keep-last 3; off-site via ep0's keep-last 2). Now (`internal/localapi/backup_cap.go`): a tier whose cadence is ≤ 24 h
+  allows **3 successful runs per rolling 24 h** (the `07` §6 day: the night run, a „Mentés most" press, and the next
+  night, which since R-899 still runs and can land inside the same 24 h — so a press never costs the night); a longer
+  tier allows **1 per max(cadence − 24 h, cadence/2)** — 6 days for the 7-day off-site tier, which the controller only
+  runs when due and the button never targets. Only successes count (a failed run prunes nothing). A refusal answers
+  **429** with the next allowed time, is logged at WARN, is saved, and for 24 h raises the CRITICAL host-report status
+  `localapi:backup-cadence-cap` — the hub's existing operator-only `agent_capability_degraded` event, **no hub change**.
+  **The cap survives an agent restart:** `BackupSuccessState` (the R-894 file) now also keeps the newest 8 success
+  times and the last refusal per tier and guest (additive JSON fields; a pre-R-973 file counts its one known
+  success). Tests `TestBackupCap_*` (an off-site flood keeps a pre-compromise copy for 5+ days; a local flood runs 3
+  a day; the scheduled day is never refused; restart; failures; the operator signal) + `TestBackupCadenceCap`,
+  `TestBackupSuccessState_R973_*`, `TestBackupCapStatus_*`, `TestR973_CapRefusalIsWiredIntoTheReport`. Red-proved:
+  the cap disabled → „the off-site tier lost every pre-compromise copy: 10 runs"; the disk history ignored → „after a
+  restart the 4th run in a day was allowed"; the wiring line removed → „never binds backupCapRefusal". **Limits, stated:**
+  with keep-last 3 the local tier's three points can still be replaced within one day (3 per day is the schedule's own
+  pace plus one press); the hub's event text reads „agent privileged capability degraded" (the type is generic). No
+  sudoers or bundle change.
+- **CI workflow (R-1135, no code change).** `.gitea/workflows/gates.yml`, the alarm step: the mail's run link is now `https://gitea.dooplex.hu/<repo>/actions/runs/<run id>`.
+  It used `GITHUB_SERVER_URL`, which on this runner is the in-cluster address (`http://gitea.gitea-system.svc.cluster.local:3000`,
+  the address the runner registers at), so nobody outside the cluster could open it. It also used the run NUMBER, and
+  Gitea 1.26 reads `/actions/runs/<n>` as a run id first: a number that is also an older run's id opened that other run
+  (measured 2026-10-18: `felhom.eu/actions/runs/1116` opened run id 1116, run number 613). It now uses `GITHUB_RUN_ID` and
+  prints the link (`ALARM-LINK …`) into the job log. Same change in all four repos' workflows.
+
+## v0.159.2 — the orphan thin check runs only after a failed backup (R-1012 follow-up) (2026-10-18)
+
+**Released 2026-10-18 (committed 2026-10-17): sha256 `4d98868e922d6c10665a6c0da9323e8627ea9c589094e81d27d741e625d9d8f9`, bundle `477704ea2980948433ffc34bb59f008cc11410fd9683d386bfd3e2f085d55bc9`; the vouch is the operator's.** No `configs/` change since v0.159.1. The independent review of 2026-10-18 found nothing in it. v0.159.0/v0.159.1 ran the R-1012 orphan check
+before EVERY snapshot-mode whole-guest backup, and that check takes a short metadata snapshot of the live thin pool.
+Now `BackupRunner` runs it only when the snapshot step has failed (the vzdump start or its task) — the moment the
+answer is needed. The finding is still appended in plain words to the failed backup's error, exactly as before; a
+successful backup never runs the check and keeps a clean record; stop mode and an unwired runner are unchanged. The
+`thin-orphan-check` capability text says „after a failed whole-guest backup". Tests
+`TestBackup_SnapshotFailure_NamesTheOrphan` (now: exactly 1 call, after the failure) and
+`TestBackup_ThinCheck_SilentWhenNotApplicable` (now: 0 calls on a successful backup), red-proved against the v0.159.1
+`runner.go`: failure legs „thin check ran 2 times, want 1", success leg `calls=1`. No root file, sudoers or bundle
+change. (Installer side, felhom.eu, same night: `felhom-host-install.sh --rotate-recovery-only`, R-1141.)
+**Proven live 2026-10-18 (demo-hp):** a green snapshot backup with 0 `felhom-thin-check` calls (the 0.159.1 night backup
+had 1). The failed leg stays covered by the fixture test.
+
+## v0.159.1 — the signed decommission prunes the guest-bind record (found by v0.159.0's own live proof) (2026-10-17)
+
+**v0.159.0 is superseded — do not vouch it.** Everything of v0.159.0 below, plus: `signedjobs.DecommissionExecutor`
+drops the drive from every guest's bind record (`SetGuestBinds`, wired to the agent's `GuestBindStore`), as the
+self-serve decommission always did. Without it a drive decommissioned by an operator-signed job and then forgotten
+(R-1124) left a record the 20-second guest-bind re-assert warned about every tick („reconcile: skipping non-enrolled
+drive") — seen on demo-hp minutes after the R-1124 proof. Test `TestDecommissionExecutor_PrunesTheGuestBindRecord`,
+red-proved. Released: tag `v0.159.1`, sha256 `de8f6ad9…aff9`, bundle `2af3d27d…7cd6` (the same root files as v0.159.0).
+**Not vouched.** Delivery to a box on the v0.158.1 bundle: `agent_update 0.159.1` → `agent_config_update 0.159.0-step1`
+(sha `1d65edb3…d6cf`, published) → `agent_config_update 0.159.1` (the bundle adds `felhom-thin-check`, R16).
+
+## v0.159.0 — forget a decommissioned drive that is gone (R-1124); name an orphan thin device in a backup failure (R-1012); Go 1.27.2 (R-972) (2026-10-17)
+
+**Superseded by v0.159.1 — do not vouch.** Task: „agent release, the catalog allow-list, break-glass rotation, off-site
+history check" (felhom.eu `REPORT-agent-allowlist-rotation-2026-10-17.md`).
+
+- **Go 1.27.2 + x/crypto v0.58.0** (R-972, `37b99d7`): the first release that carries them (v0.158.1 was built with
+  go1.26.0).
+- **R-1124 — a decommissioned drive that is gone is forgotten.** `SudoHostOps.ForgetRetiredDrives` (agent start and the
+  20-second storage tick): for a Felhom drive unit whose intent is `decommissioned`, whose `/dev/disk/by-uuid` link is
+  gone and which is not mounted, `systemctl disable` + the new root verb `felhom-priv-apply unit-forget <unit>` +
+  `daemon-reload`. The unit is gone, so a later boot no longer waits for the drive and the agent stops listing it (no
+  „absent by UUID" line every 20 s). A decommissioned drive still plugged in KEEPS its unit and raw mount, so the
+  one-click re-enroll (H3) still works. `unit-forget` re-checks as root: a felhom by-UUID drive unit (not a network
+  unit), disabled, its `Where` not mounted, its by-UUID link absent — otherwise refused. Sudoers `FELHOM_MOUNT` gains the
+  one verb. Tests `TestForgetRetiredDrives_*` and `configs/test_felhom_priv_apply.py` `UnitForget`, red-proved.
+  **Live on demo-hp:** a loop-backed test drive enrolled through the local API, decommissioned by a signed job (kept
+  its unit while present), unplugged → 8 s later the unit and its wants link were removed and `GET /disks` dropped it.
+- **R-1012 — an orphan thin device is NAMED in the backup failure, never removed.** New READ-ONLY root helper
+  `felhom-thin-check <vg> <pool>` (config bundle): LVM's thin ids before and after, the pool's ids through a metadata
+  snapshot it reserves and releases (refuses — exit 3 — when someone else holds one; releases on every exit; sends no
+  other pool message). `internal/thincheck`: an orphan = in the pool and in neither LVM list, on two reads 3 s apart;
+  fails open. `BackupRunner.SetThinCheck`: before a snapshot-mode backup (a WARN when found) and again after a failure —
+  the finding is appended in plain words to the error the controller mails to the operator, pointing at the new
+  runbook `whole-guest-backup-orphan-thin.md`. Sudoers `FELHOM_THINCHECK`, capability `thin-orphan-check`. Tests
+  `internal/thincheck`, `TestBackup_SnapshotFailure_NamesTheOrphan`, `configs/test_felhom_thin_check.py`, red-proved.
+  **Live on demo-hp:** on a scratch thin pool (a loop file on `/mnt/hdd_1`, never `pve/data`) the helper listed a made
+  orphan (device 7) and refused a snapshot held by someone else; on `pve/data` it read 1, 2 = LVM's and released; a
+  normal whole-guest backup of 9201 ran the check first and finished green.
+- Released: tag `v0.159.0`, sha256 `3cfd0a25…edfed5`, bundle `40663c4f…6d681`; step bundle `0.159.0-step1`.
+
+## CI only — a manual proof that the failure mail can be sent (R-1129, 2026-10-17)
+
+No agent release; no binary changes.
+
+`.gitea/workflows/gates.yml` gains a manual `workflow_dispatch` trigger with one input, `force_alarm` (default false), and
+one step that runs ONLY on a dispatch: it prints the length of `RESEND_API_KEY` (never the key) and, with
+`force_alarm=true`, fails on purpose so the existing alarm step sends its real mail; the mail's subject then starts
+„FORCED TEST, not a real failure". A push run is unchanged: the new step is skipped and the subject is the old one. Same
+change in all four repos' workflows. Purpose: prove the failure mail now goes out after the operator set the key
+(every failed run until today ended „ALARM FAILED: RESEND_API_KEY is empty").
+
+## scripts only — the decoy suite is a gate of the full run (R-1116, 2026-10-17)
+
+No agent release; no binary changes. `scripts/test_gate_decoys.py` is now the `gate-decoys` row of `agent_gates.py`:
+the full run (which CI runs on every push), not `--fast` — it takes ~20 s in the workspace (measured: 30 cases; ~5 s in a
+CI-shaped shallow clone, also green). It was in no runner, the shape that let the catalog's R-1115 fail unseen for a
+night. It plants only in scratch clones and asks a fake Gitea on 127.0.0.1; it does not call the runner (no
+recursion). Declared in its own `COVERS`; `test_agent_gates.py` pins it (registered once, full run only, named by
+`--fast`'s SKIPPED line). Red-proved: a reuse-refs FACT case flipped to expect rc 0 turned the full run red
+(`CONVICTED: gate-decoys`); restored, green. CLAUDE.md's `--fast` sentence names it.
+
+## unreleased — Go 1.27.2 and golang.org/x/crypto v0.58.0 (R-972, 2026-10-16)
+
+**Not released, not deployed** — agent releases wait for the operator's vouch; the next vouched release carries this.
+`go.mod`: `go 1.27.2` (was 1.25.0; the shipped 0.158.1 binary was built with go1.26.0), x/crypto v0.58.0 (was v0.52.0),
+x/sys v0.49.0. `go build`, `go vet`, the full `go test ./...` and `agent_gates.py --fast` green on Go 1.27.2.
+`release-agent.sh` builds with the toolchain `go.mod` names (GOTOOLCHAIN=auto), so no script change.
+
+## scripts only — `release-agent.sh` refuses when the registry cannot say the version is new (R-1098, 2026-10-16)
+
+No agent release; no binary changes. `scripts/release-agent.sh` already refused a version whose tag exists or whose
+package answers 200. It let everything else through — a registry it could not reach (curl prints `000`) or a 500 read
+as „new". Now only a clean `404` proves the version is new; anything else refuses with the code it got. New
+`scripts/test_release_push_once.sh` (the real script, a fake `curl` and `go` on PATH; published / absent / unreachable /
+server error / tag exists), red-proved: with the 404 rule removed, the unreachable and server-error cases fail. Same rule
+as the controller `build.sh` and, since today, felhom.eu `scripts/build-hub.sh`.
+
+## v0.158.1 — security review of v0.158.0: the mkfs wrapper trusts no sibling claim; the internal-place helper fails closed (2026-10-16)
+
+**v0.158.0 is superseded — do not vouch it.** A security review of `912f7c1` found two defects; both fixed here, each test
+seen failing against the 0.158.0 script:
+- **`felhom-mkfs-guarded`**: 0.158.0 let a format through when a SIBLING part was mounted at /mnt/<name> AND under
+  /mnt/felhom-drives („Felhom's own drive"). That corroboration is a mount the agent can make itself, so a compromised agent
+  could unlock a host owner's disk — the exact case this root wrapper exists for. Removed: any mount of any node outside
+  /mnt/felhom-drives refuses again (now read from the kernel's mount table for every node; an unlistable disk refuses).
+  Consequence: the free part next to an in-use Felhom drive cannot be formatted — `CandidatePartition.Formattable` is false
+  for every part of a disk with a mounted node, and the controller does not offer it (a part with a filesystem can still be
+  USED). Pinned by configs/test_felhom_mkfs_guarded.py + partitions_test.go.
+- **`felhom-internal-place`**: a failed `lvs` listing read as „nothing promised" (a too-large place) and a `blkid` error read
+  as „no signature" (the only gate before mkfs). Now: a failed listing, a listing without the pool, or any blkid exit other
+  than 0/2 REFUSES. Three new tests, all failing against 0.158.0.
+- Released: tag `v0.158.1` (code `98f3cdd`), sha256 `20264df6…84b76a1`, bundle `c23c3d35…cb1c41b`. **Not vouched.**
+
+## v0.158.0 — the internal data place; partitions as places; add_internal_place (2026-10-16)
+
+Design: felhom.eu `architecture/16-storage-physical-identity.md` §5, §6, §7, §9 (operator rulings D3–D5; the D3 sizes
+ruled 2026-10-16 after the space check, felhom.eu `audits/storage-identity-buildB-2026-10-15/SPACE-CHECK.md`).
+
+- **The internal data place („Belső tárhely")** (`internal/localapi/internalplace.go`): when the record
+  `/var/lib/felhom-agent/internal-place.json` names a filesystem, the agent mounts it at `/mnt/belso` by fs-UUID like an
+  enrolled drive (startup + every 20 s tick) and records it `enrolled`; `GET /disks` reports it with kind `internal`.
+  It never takes `/mnt/belso` from a household drive, and eject/decommission/format of it are refused (device-mapper on
+  the system disk). Its identity is the system disk's, so build A never counts a copy to `sys_drive` as a second copy.
+- **Root helper `configs/felhom-internal-place`** (new in the bundle): `plan|create <planned GiB>` — the ruled size rule
+  (reserve max(16 GiB, 10 %) ≤ 48 GiB, made only when ≥ 20 GiB, nothing promised beyond pool − reserve), the only
+  lvcreate/mkfs of a system-disk volume; formats only a `felhom-belso` with no signature. Run by the installer (1.33.0)
+  and by the new signed op.
+- **Signed op `add_internal_place`** (old one-disk boxes, `16` §9): no parameters (the size is the rule's), destructive
+  class (operator key). Runs the helper with `create 0`, writes the record, mounts.
+- **Partitions as places** (`internal/storage/partitions.go`): `GET /disks/candidates` lists each partition of a free
+  disk as empty / filesystem / in_use. A data-bearing part's wipe id binds disk id + part number + fs-UUID
+  (`byid:<…-partN>+fsuuid:<uuid>`); both must name the same device at format time.
+- **`felhom-mkfs-guarded`** (REVERTED in v0.158.1): a SIBLING part that is Felhom's own drive no longer blocked formatting
+  the free part next to it. First tests of the wrapper: `configs/test_felhom_mkfs_guarded.py`.
+- `configs/` changed (sudoers `FELHOM_INTERNALPLACE`, the helper, the wrapper, the bundle list) — **a config-bundle job
+  is needed** with the agent update. **Not vouched** (the operator's act).
+- Released: tag `v0.158.0` (code `912f7c1`), sha256 `a3589d1d…2caba`, bundle `17066273…fb5c76`. **Superseded by v0.158.1 —
+  do not vouch** (its mkfs wrapper trusted a sibling claim the agent can fake; its helper failed open).
+
+## v0.157.0 — every storage place knows its physical disk(s); POST /disks/same-disk (2026-10-15)
+
+Design: felhom.eu `architecture/16-storage-physical-identity.md` §3, §4, §10 (operator rulings D1–D6, 2026-10-15). Spike:
+felhom.eu `audits/storage-identity-2026-10-15/`.
+
+- **Identity of a storage place** (`internal/storage/identity.go`): the physical disk(s) under it, from `/sys`, `/run/udev/data`
+  and `/proc/mounts` only — partition → disk, device-mapper / LVM thin → slaves, loop → backing file → its disk. Keys:
+  `eui:` (NVMe), `serial:<bus>:<model>:<serial>`, `wwn:` only without a serial and only with a vendor (OUI ≠ 000000).
+  `proven`, or `unknown` with one reason: `vm` (systemd-detect-virt), `network`, `usb_bridge` (a placeholder serial — the rule
+  is the design's, UNPROVEN on a real enclosure, R-1077), `weak_wwn`, `no_serial`. No block I/O.
+- **Reported:** `GET /disks` rows and the hub report's `storage_targets` gain `kind`, `physical_disks`, `identity`,
+  `identity_reason` (additive; the golden and the hub's copy change together). A dir storage is judged only on its OWN mount — an
+  absent drive is "not computed", never the root disk.
+- **`POST /disks/same-disk`** `{a, b}` (guest paths) → `{different, proven, reason, a, b}`. Maps `/mnt/felhom-drives/<name>` to
+  the host mount at exactly that path, and `/mnt/sys_drive`, `/var/lib/felhom`, `/var/lib/docker` to the guest's data volume's
+  storage. Anything else, or an absent drive → `unmapped`. A shared disk key is "same, proven" even when the disk is unknown.
+  Scoped like every `/disks` call.
+- Tests: fixture `/sys` + udev trees for every spike case (one disk LVM thin, two disks, two partitions, a loop on a file, NVMe eui,
+  vendor-less WWN, a VM, a USB placeholder serial, a network share) and the same-disk call (equal, different, VM, NAS, absent
+  drive, scope); red-proofed (the LVM "diamond" bug was seen failing first; the VM, WWN, own-mount and sys_drive rules broken on purpose).
+- `configs/` unchanged — no config-bundle job needed. **Not vouched** (the operator's act).
+
+## Unreleased (2026-10-15, afternoon) — the shared rule file; no code change
+
+- `.claude/rules/unprompted-work.md` §3 item 6: **a controller release needs no golden** — the floor carries it after the
+  three test boxes; agent releases still wait for the operator's vouch; the golden is weekly (operator ruling 2026-10-15
+  11:26, `09` §3 decision 250; the same morning's „bake first", decision 247, reversed). Same text in all five copies.
+
+## Unreleased (2026-10-15) — the shared rule file; no code change
+
+- `.claude/rules/unprompted-work.md` §3 item 6: **bake first** — a controller release is released only when its golden is
+  baked; bake → the operator vouches → the operator raises the floor; never ask for a floor for a version with no golden
+  (operator ruling 2026-10-15, `09` §3 decision 247, superseding decision 7). Same text in all five copies.
+
+## v0.156.1 — R-1008 by process liveness, not the clock; 0.156.0 superseded — do not vouch it (2026-10-14)
+
+Released by `scripts/release-agent.sh`: binary sha256 `83c7da27d457b63ba8a4cde5a8717b3e5f9db9f83a46e63f38902dc7471bffb9`, bundle
+`5a678def792c70ec1cbbd198c14542161fac91d11a69bd0ce23dd86469e96ec3` (tag `v0.156.1` = `7ae3081`). Step bundle `0.156.1-step1`
+sha256 `ac83ea41507616d4c1056cccfd35de1410ff01e0296f4e64ff925207a43a8490` (the 0.155.0 bundle with only `felhom-os-apply`
+replaced), published as package version `0.156.1-step1`. **Not vouched** (the operator's act). Delivery as 0.156.0: per box
+`agent_update 0.156.1` → `agent_config_update 0.156.1-step1` → `agent_config_update 0.156.1`.
+
+- **0.156.0 is superseded — never vouch it.** A background security review of the pushed commit flagged „fail-open state
+  drift": 0.156.0 called a listed vzdump task dead when its wall-clock start was before the kernel's `btime`. `btime` is
+  derived from the wall clock and moves when the clock is corrected after the boot (an NTP step after a power cut), so a
+  backup started between the boot and the correction could look „from before the boot" — and the stale-lock pass would
+  clear a LIVE backup's lock.
+- **Now:** `taskAlive` is Proxmox's own rule (`PVE::ProcFSTools::check_process_running`, read on demo-hp): the UPID's pid and
+  pstart against `/proc/<pid>/stat` — same start tick (field 22) and not a zombie. No clock is involved. Fails safe (running)
+  when the UPID does not parse, the stat cannot be read, or this agent cannot see root's processes at all (PID 1 not visible:
+  a `hidepid` /proc or `ProtectProc=` would hide every worker — demo-hp: `/proc` plain, `ProtectProc=default`).
+- Tests: worker gone, pid reused, zombie, live worker, a „clock step" (wall-clock start absurdly early, worker alive → still
+  running), doubt, hidden /proc — each red-proofed; 19 guards in all (`power-cut-fixes-2026-10-14/redproof.txt`).
+- R-1006 (the cleanup) is unchanged; its „before this boot" rule still reads `btime`, but a drifted `btime` cannot remove a
+  live partial: the Go side asks this liveness rule, and the root helper refuses while any vzdump runs or anything holds
+  the file open.
+
+## v0.156.0 — after a power cut: the dead half-written backup is removed (R-1006); a backup task from before the boot is not „in flight" (R-1008) (2026-10-14)
+
+Released by `scripts/release-agent.sh`: binary sha256 `8db2278838dcb733d5201fba84b75a829b9adebdf899c89dd7de0939ec10d51a`, bundle
+`a21605217966ebc953d55a35157a9eb0243ca542c50dbd94fc9fa67257f27369` (tag `v0.156.0` = `a2f3aee`). Step bundle `0.156.0-step1` sha256
+`8be2bb8f3d54bff8e7140cf99714664df31adecf13aa721597323fb2b9d43db5` (the 0.155.0 bundle `795f2ab4…` with only `felhom-os-apply`
+replaced; `scripts/build-step-bundle.py`), published as package version `0.156.0-step1`. **Not vouched** (the operator's act).
+
+**Delivery: the agent binary AND the config bundle, and the bundle ADDS a path** (`/usr/local/sbin/felhom-vzdump-partial-clean`)
+— so per box: signed `agent_update 0.156.0` → `agent_config_update 0.156.0-step1` (the step bundle) → `agent_config_update
+0.156.0`. Evidence: felhom.eu `documentation/audits/power-cut-fixes-2026-10-14/` (the spike C1–C4, the red-proofs).
+
+- **R-1008 — one in-flight answer.** `staleLockController.BackupRunning` (the ONE implementation all four callers reach:
+  the start-up stale-lock pass, the guest-power watchdog, the household power-off guard, the controller supervisor) now
+  answers through `vzdumpInFlight`: a vzdump task for the vmid that started at or after the host's boot. Proxmox keeps a
+  dead task in `tasks?source=active` for ~30–60 s after a power cut (measured, C3) — the agent starts inside that window
+  and called it „genuinely in-flight". A task from before the boot is dead; logged once (vmid, UPID, start, boot). The
+  start time comes from the list or, if absent, the UPID; an unreadable boot time or start time FAILS SAFE (running), as
+  before. The boot time is `poweroff.ProcBtime` (no third /proc/stat parser).
+- **R-1006 — the half-written backup is removed.** New `internal/vzpartial`: on a `dir` backup target it finds exactly
+  `vzdump-(lxc|qemu)-<vmid>-YYYY_MM_DD-HH_MM_SS.(tar.dat|vma.dat|tmp)` of THAT guest (never a finished archive, a log,
+  another guest, a symlink, a wrong kind), and — only when no vzdump of the guest is in flight (R-1008) and the entry is
+  dead (modified before this boot, or untouched for 120 s) — asks the root helper to remove it. Runs at agent start (after
+  the stale-lock pass, every owned guest × every backup tier) and before every backup's space preflight
+  (`BackupRunner.SetBeforeBackup`). Any doubt leaves the entry and logs why.
+- **The root helper `configs/felhom-vzdump-partial-clean`** (config bundle; sudoers `FELHOM_VZPARTIAL`, one exact line: a
+  storage id, a vmid, one half-written name) re-checks everything as root: a `dir` storage from `/etc/pve/storage.cfg`; the
+  name of that vmid; not a symlink; directly in `<path>/dump`; **no vzdump on the host at all** (a job for several guests
+  carries no vmid in its task name); **no process holds the entry open** (C2: after a killed worker, `tar` and `zstd` kept
+  writing into the `.tar.dat` for 4 min 40 s, none named „vzdump"); dead (before this boot, or quiet 120 s).
+- **Not changed, by evidence (C4):** the space preflight's sizing (R-1006 part 2) — on „No space left" vzdump removes its
+  own partial on both its paths (rsync and tar); only a power cut leaves one.
+- **Tests:** `stalelock_inflight_test.go` (the helper, the production controller, three callers end to end, a structural
+  pin that every `BackupRunning` call goes through `s.staleLock` and there is one implementation), `vzpartial_test.go`
+  (removes; keeps in flight / unknown / non-dir / other names / recent / wrong kind / symlink / a helper refusal — each keep
+  with its positive control), `r685_space_test.go` (the cleanup runs BEFORE the preflight), the helper's own 21 Python
+  tests (`configs/test_felhom_vzdump_partial_clean.py`, run by the Go suite), 8 sudoers injection cases. **17 guards
+  red-proofed** by mutation (`power-cut-fixes-2026-10-14/redproof.txt`).
+- **Fixed without a row:** `internal/localapi/guestbind.go` and `controllersupervisor_test.go` were not gofmt-clean on
+  `main` (an import order and a struct alignment) — formatted.
+
+## v0.155.0 — the household can turn the machine off: POST /host/poweroff (R-998); the escrow reply race fixed (2026-10-13)
+
+Released by `scripts/release-agent.sh`: binary sha256 `bb339e944eff4aa6d10ae09158e0943de2e5c3777f52bf993c796d57e46800c3`, bundle
+`795f2ab41661e993e3fc4ce65df4f11ce94280d94871ab2e5cf1993ae3fbff7d` (tag `v0.155.0`). **Not vouched** (the operator's act).
+**The bundle changes `felhom-agent.sudoers`** (no new path — no step bundle): delivery is a signed `agent_update` +
+`agent_config_update`. Design: felhom.eu `documentation/architecture/14-host-power-off.md`.
+
+- **`POST /host/poweroff`** (local API, `withGuest`-scoped like `/guest/reboot`) — the controller's „A gép kikapcsolása"
+  (controller v0.310.0). Refused (409, a stable `reason`) in the first 10 min after boot; a second time in one boot
+  (`internal/poweroff`, a file keyed by `/proc/stat` btime); while ANY other guest or VM runs on the host (read from
+  `/proc`: `lxc-start -n N`, `kvm -id N` — demo-hp's scratch guest and Tester 1 VM block it); while the heavy-operation
+  gate is held (backup, restore test, OS step, sweep, trim), a backup job, a controller swap, a drive format, a NAS verify
+  or an escrow ceremony runs, or Proxmox runs a vzdump of the guest. Every check fails closed.
+- On acceptance: the record first; the heavy gate TAKEN and kept; a host report carrying `deliberate_off.requested_at`
+  accepted by the hub (bounded 60 s, `Loop.SetReportedHook`); then before each step the jobs and other guests are
+  checked again — a job waits (≤ 15 min), another guest cancels at once, a job still running after the wait cancels
+  (gate released, box stays on); `pct shutdown <vmid> --timeout 120`; `systemctl poweroff`.
+- **Sudoers:** `FELHOM_POWEROFF` = `/usr/bin/systemctl poweroff`; `pct ^shutdown [0-9]+ --timeout [0-9]+$` joins
+  `FELHOM_GUESTHOOK`; nine injection cases (`--forceStop`, `poweroff --force`, `reboot`, `halt`, `kexec`…).
+  **New `TestEveryAliasIsGranted`:** an alias missing from the grant line passed the coverage check before.
+- **Host report:** `deliberate_off` (omitempty; the cross-repo golden unchanged). Hub v0.149.0 holds its silence alarms
+  for 14 days on it.
+- **Fixed without a row:** `go test -race` found a data race in the escrow ceremony start reply (it read `job.Phase`
+  while the job goroutine wrote it); the reply fields are copied first. Red before, green 3/3 after.
+- Security review (three passes the same evening): other guests, a gate not held, fail-open unwired checks, jobs and
+  guests starting after acceptance — each fixed and red-proofed (`hostpoweroff_test.go`, 11 tests).
+
 ## v0.154.0 — no OS leg after a household press; the after-boot kernel report carries the real ring (R-899); no „wrong code" when older packages were not checked (R-304); the last backup survives a restart in the host report; the Secure Boot meta-package leaves the Proxmox lane (2026-10-09)
 
 Released by `scripts/release-agent.sh`: binary sha256 `36a54ad20e896cd61d7b9944e2a521f4209c2b76940d8f1827af05b645463226`, bundle

CLAUDE.md

@@ -56,7 +56,8 @@ This is in the core because breaching it is how this component stops being audit
 gate in its `GATES` table (that table is the list); the shared ones — `reuse_refs_check`,
 `instructions_gate`, `observations_gate` — are the copies in `felhom.eu/scripts/`, never copied into
 this repo (a copy recreates the drift they detect; an absent sibling clone FAILS). `--fast` selects the
-gates touching no network and no container runtime, and skips `published` (network), naming it. **A missing gate is a FAILURE, never a skip.**
+gates touching no network and no container runtime, and skips `published` (network) and `gate-decoys` (the decoy
+suite, ~20 s; R-1116), naming them — CI runs the full set. **A missing gate is a FAILURE, never a skip.**
 
 **The pre-push hook** (`.githooks/pre-push`) runs it with `--fast` and refuses a failing push. It is
 **per-clone** — switch it on once with `git config core.hooksPath .githooks`, and a manual run WARNS

CONTEXT.md

@@ -1,5 +1,40 @@
 # CONTEXT — felhom-agent working state
 
+> **2026-10-19 — v0.160.0 RELEASED (operator yes), NOT VOUCHED; on demo-hp, Tester 1, demo-felhom.** R-1093, R-978,
+> R-973 (below) plus R-1149 (the local tier prunes `keep-daily=N`, `BackupTier.PruneSpec`) and R-1151 (a claim with
+> `{"defer_upload": true}` keeps the package until `POST /escrow/ceremony/confirm` brings the two words at the drawn
+> positions; only those two words are held, 30-min window; a claim without it keeps the R-978 order). Vouch: version
+> 0.160.0, sha256 `7ebb547f…dfca1`, bundle `6b060901…e202f`. After it, unreleased: `4545819` (two texts). felhom.eu
+> `REPORT-day-2026-10-19.md`.
+> **2026-10-18 (night) — UNRELEASED on main: R-1093, R-978, R-973** (release + vouch: the operator's). R-1093 an empty
+> partition is blank (lsblk's PTTYPE on a partition is the parent's; set aside only on a clean blkid exit 0 with a
+> matching `PART_ENTRY_SCHEME`; an unparsable lsblk is unprobed). R-978 the escrow package reaches the hub only at the
+> claim (ceremony output v2, `PendingUpload`; `EscrowCeremonyConfig.Upload`). R-973 the `POST /backup` cadence cap.
+> **Decision (decided by CC unattended — operator may reverse): the cap's numbers.** Q: how many whole-guest runs may
+> the guest ask for? Options: (a) 2 per 24 h on the daily tier — keeps one local point older than a day, but refuses
+> the NIGHT after a press (R-899's night can fall inside the same 24 h); (b) 3 per 24 h + 1 per 6 days on the weekly
+> tier — the schedule never refused, a second press in a day refused, the local tier still replaceable in a day;
+> (c) none — status quo. Chose (b): the off-site tier is the one `07` §8 row 10 relies on and it is fully protected
+> for ~6 days; the local half's real fix is a time-bucketed prune (keep-daily), which is a retention change for the
+> operator. No sudoers or bundle change in any of the three.
+
+> **2026-10-17 — v0.159.1 RELEASED (0.159.0 superseded, do not vouch).** Go 1.27.2; R-1124 a decommissioned drive that is
+> gone is forgotten (`ForgetRetiredDrives`, root verb `unit-forget`); R-1012 the read-only `felhom-thin-check` names an
+> orphan thin device in a failed backup's text; the signed decommission prunes the guest-bind record. Bundle adds a path →
+> step bundle `0.159.0-step1`. Proven on demo-hp; delivered by signed jobs on the operator's yes. Vouch: the operator's.
+
+> **2026-10-16 — v0.158.1 RELEASED and VOUCHED (storage identity build B; 0.158.0 superseded, do not vouch).** The internal
+> data place (root helper `felhom-internal-place`, the record in the state dir, mounted at /mnt/belso like a drive), the
+> signed `add_internal_place`, partitions in `/disks/candidates` (`formattable` only while nothing on the disk is mounted),
+> the partition wipe id `byid:<…-partN>+fsuuid:<uuid>`, the strict mkfs wrapper. Delivered through step bundle
+> `0.158.1-step1` (a bundle that adds a path, R16). Next: R-1093 (an empty part's probe reads the parent's partition table).
+
+> **2026-10-15 (night) — v0.157.0 RELEASED, NOT VOUCHED (storage identity build A, felhom.eu `16` §13).** Every storage
+> place carries `kind`, `physical_disks`, `identity` (+ reason) in `GET /disks` and the hub report; `POST /disks/same-disk`
+> answers the controller (v0.317.0, MinAgent 0.157.0). Computed on every read, never cached. Delivered by signed jobs to
+> Tester 1, demo-hp, N100 only. Next: the operator vouches 0.157.0 → the global floor 0.317.0 / MinAgent 0.157.0. Build B
+> (partition candidates, the internal place) waits for a task file (R-1079).
+
 > **2026-10-08 (day) — UNRELEASED on main, ships with tomorrow's release (decision 178).** R-899: `POST /backup?trigger=manual`
 > runs no OS leg (`localapi/server.go`); after-boot kernel reports carry the saved block's ring (`kernelReportRing`). R-304:
 > `escrow.ErrRetainedUnchecked` → HTTP 424 `older_unchecked` when not every earlier package was tried (withheld, caps,

REPORT.md

@@ -1,15 +1,20 @@
-# REPORT — v0.154.0 released and delivered (2026-10-09)
+# REPORT — agent, evening/night brief 2026-10-19 (code + tests + commit only, NOT released)
 
-Built from the 2026-10-09 kernel-night read-back: (1) the host report now adds the saved newest backup success per tier
-(R-894's file, one shared instance), so a restart right after a night backup no longer empties the hub's evidence — on
-demo-felhom it had caused a false „newest backup is 48h old" alarm; (2) `proxmox-secure-boot-support` joins the
-boot-chain lane, so it no longer pulls `shim-signed-common` into the ring-0 Proxmox plan (demo-hp's step was refused R6
-and its kernel step skipped). Red-proofs: `TestCollectBackups_SavedSuccessSurvivesARestart`,
-`TestR894_LastKnownBackupsIsWiredIntoTheDaemon`, `test_ring0_pending_pve_leaves_the_secure_boot_meta_out`.
+Commit `56aabc7` (CHANGELOG „Unreleased"), scanner wiring `84ef581`. No sudoers, wrapper or config-bundle change.
+The release is the operator's word (STATUS morning list: „Release agent 0.161.0?"), then the vouch.
 
-`scripts/release-agent.sh 0.154.0`: sha `36a54ad2…`, bundle `93487989…`, tag `v0.154.0` = `db25b46`, verified by
-download. CI 1576 (code) success; 1579 (CHANGELOG) failed in its fetch step (Gitea answered 500), re-run → success.
-Signed `agent_update` → demo-hp, demo-felhom, Tester 1 (07:16–07:18 local); signed `agent_config_update` → `BUNDLE DONE
-written=1 same=26 self-check=ok`, capability probe 68/68 on all three. NOT vouched (the vouch needs a golden at or above
-the fleet's controller; no golden today). Live: demo-felhom's first report after the update carries the 02:40Z backup.
-Evidence: `felhom.eu/documentation/audits/release-2026-10-09/delivery/`.
+- **R-921 — `host_busy` on `GET /backup/status`:** whether the host-wide heavy-operation gate is held (`{busy, what}`),
+  the state in which `POST /backup` answers 409. Controller 0.335.0 (the floor) reads it before any stop. Test
+  `TestBackupStatus_R921_HostBusy`, red-proved (assignment removed → `free gate: host_busy = <nil>`).
+- **R-1070 — tunnel `absent`:** Docker 29 prints „error: no such object" in lower case; the case-sensitive match read
+  a missing cloudflared container as `unknown`, which failed the host OS-update rule (Tester4, 2026-10-15). Now a
+  missing container is `absent`, and the host rule passes it only when the reading from before the step also had no
+  cloudflared container. Tests `TestHostHealthVerdict_R1070_NoTunnelSetUp`, `TestClassifyTunnel` (Docker 29's wording,
+  measured on demo-hp), red-proved both ways.
+- **R-1010 — the janitor's forced snapshot delete:** on „Failed to find logical volume …_vzdump" only, one retry with
+  the API's `force=1` (`DeleteSnapshotForce`; `pvesh usage … -command delete` lists `--force`). Tests
+  `TestStaleLock_R1010_ForcesOnceOnlyWhenTheVolumeIsGone` (6 rows), `TestDeleteSnapshotForce_R1010`, red-proved.
+- **R-986 — the security-scan gate** (tooling): the shared `felhom.eu/scripts/security_scan_gate.py` in the fast set;
+  `.security-accepted` holds 15 accepted test constants; decoys in `test_gate_decoys.py`.
+
+Gates: `go build/vet/test ./...` green; `agent_gates.py --fast` OK; decoys 36 cases OK. CI 2103, 2110 success.

REUSE.md

@@ -15,13 +15,14 @@
 | `SudoHostOps.run` | internal/storage/hostops.go | `run(ctx, name, args...) error` | allowlisted exec with stderr-wrapped error | Every arg pre-validated via validate.go before this is called |
 | `Prober.Probe` | internal/capability/probe.go | `Probe(ctx) []Status` | live sudo-policy capability check (`sudo -n -l --`) | Needs a DIRECT runner (never the sudo-prefixing one — double-sudo); never executes probed cmds. v0.86.0: config-gated caps (`Capability.GatedBy` + `Prober.GateActive`) report `inactive`/"disabled by configuration" ONLY when healthy — broken plumbing stays degraded; the pbsdr-* gate answers from `pbsdr.Manager.DRConfigured` (marker-backed across restarts) |
 | ~~`stageTemp`~~ (REMOVED v0.146.0, R-861) | — | — | — | Nothing the agent writes is `install`ed where root reads it any more: use `felhom-priv-apply` (below) or ship a fixed file in the bundle |
-| `felhom-priv-apply` (v0.146.0, R-861) | configs/felhom-priv-apply | `felhom-priv-apply unit <name> \| dnsmasq <tmp> <name> \| wg \| sshd-config \| sshd-key \| controller-image <vmid>` (the last reads the ref on stdin, R-861 (a) A1) | ANY agent-rendered file a root program reads (systemd unit, dnsmasq drop-in, wg-quick conf, OOB sshd) — fixed source + destination, CONTENT checked against the agent's own renderers | A new renderer needs a verb + a contract test (`internal/privapplytest.Check`) feeding its REAL output; never a new `install` sudoers line |
+| `felhom-priv-apply` (v0.146.0, R-861) | configs/felhom-priv-apply | `felhom-priv-apply unit <name> \| unit-forget <name> (v0.159.0, R-1124) \| dnsmasq <tmp> <name> \| wg \| sshd-config \| sshd-key \| controller-image <vmid>` (the last reads the ref on stdin, R-861 (a) A1) | ANY agent-rendered file a root program reads (systemd unit, dnsmasq drop-in, wg-quick conf, OOB sshd) — fixed source + destination, CONTENT checked against the agent's own renderers | A new renderer needs a verb + a contract test (`internal/privapplytest.Check`) feeding its REAL output; never a new `install` sudoers line |
 | `privapplytest.Check` | internal/privapplytest/check.go | `Check(t, verb, name, content) string` | the Go↔root-checker contract: a renderer's real output must read `OK` | Skips without python3; one call per rendered shape + one refused control |
 | `BUNDLE_FILES` + `Bundle` (mode `bundle`, `--install-bundle`; mode `agent_update` v0.146.0, R-861) | configs/felhom-os-apply | the ONE table of root-owned paths + the installer of them | ANY new root-owned file the installer writes (sudoers line, wrapper, unit) — add it to the table, never a new installer fetch (R-840) | The builder (`scripts/build-config-bundle.py`) and the installer read the same table; `test_every_root_file_the_installer_writes_is_in_the_bundle` fails on a path the bundle lacks. Trust files (`/etc/felhom/os-trust.json`, `operator-signers`) are NEVER bundle paths (R17) |
 | `osupdate.ConfigUpdateExecutor` | internal/osupdate/bundle.go | signed op `agent_config_update` {agent_version, bundle_sha256} | delivering the bundle to an installed box | a courier only: the root wrapper re-verifies signature, host, nonce and sha itself |
 | `osupdate.Leg.SendUnsent` / `lockPass` (v0.144.0, R-868) | internal/osupdate/unsent.go | `(ctx) int` | an OS-pass report the agent never sent (killed mid-pass): the wrapper keeps `report-<run>-<layer>-apply.json` beside the plan; the agent deletes it once the hub has it | any new caller that runs an apply pass must hold `lockPass` (flock, across processes) — the sender must never take a running pass's copy |
 | `osupdate.LoadSavedBlock` (v0.144.0, R-866) | internal/osupdate/leg.go | `(planDir) (block, savedAt, ok)` | the hub's newest os_update block as the daemon last received it (`os-update-block.json`) | the debug pass uses it ONLY when the hub cannot be reached, and says so in its header; no saved block → no pass |
 | `dpkg_state()` / `DPKG_STATE_SCRIPT` (v0.145.0, R-876) | configs/felhom-os-apply | `audit, journal = self.dpkg_state()` | dpkg's state in ONE call: `--audit` AND the update journal | never gate a repair on `--audit` alone — a crash leaves only the journal (measured); keep it one call (R-845) |
+| `thincheck.Checker.Check` + `Describe` (v0.159.0, R-1012) | internal/thincheck/thincheck.go | `Check(ctx, vmid) []Finding` | name an ORPHAN thin device (pool holds it, LVM does not) in a whole-guest backup failure — `BackupRunner.SetThinCheck` | READ-ONLY by design: the root helper `felhom-thin-check <vg> <pool>` sends only reserve/release_metadata_snap; an orphan needs two reads; fails open (never stops a backup). The repair stays human |
 | `guesthook.SnippetReady` / `Register` (v0.146.0, R-861) | internal/guesthook/install.go | `SnippetReady(path) error` | is the pre-start hook (a FIXED file from the bundle) in place — register only then | The agent never installs the hook (Proxmox runs it as root); a missing hookscript stops a guest start, so never `Register` without `SnippetReady` |
 
 ### Disk / format safety (role gates, durable IDs, format guards)
@@ -41,6 +42,11 @@
 | `storage.ResolveStorageDevice` | internal/storage/durable_device.go | `ResolveStorageDevice("uuid:<fs-uuid>") (dev, err)` | re-resolve enrolled STORAGE drives (remount) | `uuid:` scheme ONLY — distinct from the wipe schemes; never trust a remembered /dev node |
 | `deriveDurableID` | internal/storage/durableid.go | `deriveDurableID(typ, s, backingDevice, uuid) string` | storage-target durable id (DR re-attach key) | Deterministic per type; `uuid:` for usb/local-dir; PBS id includes `#<fingerprint>` |
 | `SystemDisks` / `isSystemBacked` / `RoleForStorage` / `RoleForRawDevice` | internal/storage/role.go | `RoleForRawDevice(device, sysDisks, sysKnown) DeviceRole` | protection-tier classification | Fails safe to `system` (most protected) on any ambiguity; role is AGENT-derived, never caller-supplied |
+| `storage.IdentityResolver` / `Relation` / `MountFor` | internal/storage/identity.go | `HostPath(path) PlaceIdentity`, `Device(dev) PlaceIdentity`, `Relation(a, b) (different, proven, reason)` | physical-disk identity of a storage place (felhom.eu `16-storage-physical-identity.md` §3, v0.157.0) | Reads /sys + /run/udev/data + /proc/mounts only (no block I/O). Judge a drive ONLY on its own mount (`localapi.ownMountIdentity`) — never the containing fs, or an absent drive reads as the root disk. A shared disk key is "same, proven" even when unknown. Pinned by identity_test.go (fixture trees) + samedisk_test.go |
+| `localapi.Server.EnsureInternalPlace` / `InternalPlaceRecord` | internal/localapi/internalplace.go | `EnsureInternalPlace(ctx)` (startup + 20 s tick); `ReadInternalPlace` / `WriteInternalPlace` | the one-disk box's internal data place (felhom.eu `16` §5, v0.158.0): mounts it at /mnt/belso by fs-UUID like an enrolled drive, records it `enrolled`; the controller binds + registers it | Never takes /mnt/belso from a household drive (a foreign mount, or a unit naming another UUID, waits). Never formats/ejects (dm on the system disk → role system). The ONLY maker is the root helper `configs/felhom-internal-place` (installer + signed `add_internal_place`). Pinned by internalplace_test.go |
+| `configs/felhom-internal-place` | configs/ (bundle → /usr/local/sbin) | `plan\|create <planned GiB>` | the ruled size rule (reserve max(16, 10 %) ≤ 48 GiB; ≥ 20 GiB or nothing) + the only lvcreate/mkfs of a system-disk volume | Formats only a felhom-belso with NO signature; refuses a foreign one; never resizes/removes. Sudoers: `^(plan\|create) 0$` only. Pinned by configs/test_felhom_internal_place.py |
+| `storage.candidatePartitions` / `CandidatePartition` | internal/storage/partitions.go | `CandidateDisk.Partitions` | partitions as places (`16` §6): empty / filesystem / in_use per part of a free disk; `Formattable` only while no node of the disk is mounted | A part with children or a member type is never listed; a raw /mnt mount counts as Felhom's only when corroborated (felhomOwnedMounts) — for LISTING only: the root mkfs wrapper trusts no such claim (v0.158.1). Pinned by partitions_test.go |
+| partition wipe id `byid:<…-partN>+fsuuid:<uuid>` | internal/storage/durable_device.go | `DeviceDurableID` / `ResolveDurableDevice` | a data-bearing PART binds disk id + part number + filesystem (`16` §6) | Both halves must resolve to the SAME device or the wipe refuses. Whole disks keep the plain id. Pinned by partition_durable_test.go |
 | `ValidateUUID/MountPath/BlockDevice/FSType/SMARTDevice/LVMName`, `UnitNameForMount` | internal/storage/validate.go | `Validate*(v) error` | EVERY arg that reaches a root shell-out | The security boundary; strict whitelists (no by-* symlinks, no dm, no traversal); `systemdEscapePath` computed in-process |
 | `ValidateNetworkMountSpec` | internal/storage/netmount.go | `ValidateNetworkMountSpec(spec) error` | NAS mount input boundary | Same discipline as validate.go; SMB requires a creds ref; mountpoint confined under `NetworkMountRoot` |
 
@@ -51,6 +57,7 @@
 | `SudoHostOps.EnsureMount` | internal/storage/hostops.go | `EnsureMount(ctx, MountSpec) error` | persistent by-UUID systemd .mount | Validate→render→stage→`install`→`enable --now`; idempotent |
 | `SudoHostOps.Unmount` | internal/storage/hostops.go | `Unmount(ctx, where) error` | detach a mount unit | DESTRUCTIVE — caller MUST have gated it; does not self-authorize |
 | `SudoHostOps.ReassertEnrolledMounts` | internal/storage/hostops.go | `ReassertEnrolledMounts(ctx)` | reboot remount (re-resolve by UUID) | Re-asserts unless mounted AND enabled (`shouldReassertMount`); skips absent UUIDs |
+| `SudoHostOps.ForgetRetiredDrives` (v0.159.0, R-1124) | internal/storage/forget_retired.go | `ForgetRetiredDrives(ctx, IntentReader) int` | drop the host mount unit of a DECOMMISSIONED drive that is GONE (start-up + the 20 s tick) | Only decommissioned + absent + unmounted; a present decommissioned drive keeps its unit (H3 re-enroll). The file goes only through `felhom-priv-apply unit-forget`, which re-checks all three as root |
 | `GuestBinder.AttachDrive` / `DetachDrive` | internal/localapi/intermediary.go | `AttachDrive(ctx, vmid, where) (guestPath, err)` | live drive hot-swap under `/mnt/felhom-drives` | Normalizes to EXACTLY ONE bind via `countHostMounts` (converges double-binds); force re-bind when guest can't see it |
 | `GuestBinder.EnsureSharedParent` | internal/localapi/intermediary.go | `EnsureSharedParent(ctx) error` | shared-parent bind + boot unit | make-private+make-shared ONLY on first bind — re-running orphans the guest's slave; F2-a: compares script AND unit for staleness |
 | `StablePathForRaw` / `DriveNameFromRaw` | internal/localapi/intermediary.go | `StablePathForRaw("/mnt/<n>") string` | raw host mount → stable guest path | Single source of truth both repos derive from |
@@ -107,8 +114,8 @@
 | `Engine.RunSignedJob` | internal/reconcile/job.go | `RunSignedJob(ctx, intent, signed, exec) JobResult` | executing a gated destructive job | Idempotency by nonce; journaled |
 | `escrow.Create` | internal/escrow/escrow.go | `Create(ctx, CreateOptions) (CreateResult, R, error)` | PBS-key escrow (zero-knowledge) | Recovery code returned SEPARATELY from the result (anti-log); self-verifies recoverability |
 | `escrow.GenerateRecoveryCode` / `joinSafe` / `RecoveryCodeSep` | internal/escrow/wordlist.go | `GenerateRecoveryCode() (string, error)` | minting the customer recovery code R | Draws from the EFF large list **filtered of every word containing `RecoveryCodeSep`** (4 entries: drop-down, felt-tip, t-shirt, yo-yo) so a code always segments back into exactly 10 words — a hyphenated word made codes ambiguous to transcribe AND flaked the test ~1/5 (v0.93.0). Generation-only: **already-issued codes stay valid**, R is verified as a whole passphrase and never re-split. Never count words by splitting the joined string — count what the generator drew |
-| `escrow.CeremonyBinary` / `CeremonyArgs()` / `CeremonyOutput` | internal/escrow/ceremony.go | the ONE fixed sudo self-invocation argv + the `--output=json` wire object (v1) | controller-driven ceremony (v0.88.0) | SINGLE SOURCE shared by the localapi exec, the capability manifest entry, and (byte-identically) the FELHOM_ESCROW sudoers line — `TestEscrowCeremonyArgvPinned` + `TestManifestCoveredBySudoers` lock all three. Never flag-helpers, never `--`→`-` (spike §2.2) |
-| localapi escrow ceremony job | internal/localapi/escrow_ceremony.go | `POST /escrow/ceremony` + status + ONE-SHOT claim + preflight | the wizard's agent half | R lives ONLY in `Server.escrowR` (NEVER the job struct — snapshots must be structurally R-free); zeroed on claim/supersede/10-min TTL (`unclaimed_void`); in-memory BY DESIGN (restart loses R safely; re-run supersedes); subprocess stdout is SECRET-BEARING → parsed then zeroed, never logged |
+| `escrow.CeremonyBinary` / `CeremonyArgs()` / `CeremonyOutput` | internal/escrow/ceremony.go | the ONE fixed sudo self-invocation argv + the `--output=json` wire object (v2, R-978: json mode returns the package in `PendingUpload` instead of uploading) | controller-driven ceremony (v0.88.0) | SINGLE SOURCE shared by the localapi exec, the capability manifest entry, and (byte-identically) the FELHOM_ESCROW sudoers line — `TestEscrowCeremonyArgvPinned` + `TestManifestCoveredBySudoers` lock all three. Never flag-helpers, never `--`→`-` (spike §2.2) |
+| localapi escrow ceremony job | internal/localapi/escrow_ceremony.go | `POST /escrow/ceremony` + status + ONE-SHOT claim + preflight | the wizard's agent half | R lives ONLY in `Server.escrowR` (NEVER the job struct — snapshots must be structurally R-free); zeroed on claim/supersede/10-min TTL (`unclaimed_void`); in-memory BY DESIGN (restart loses R safely; re-run supersedes); **R-978: the new package reaches the hub only at the claim, before R is handed out (`EscrowCeremonyConfig.Upload`); TTL and a new start wait while that upload is in flight — an abandoned ceremony leaves the previous package current**; **R-1151: a claim with `{"defer_upload":true}` hands out R + two word positions and keeps the package until `POST /escrow/ceremony/confirm` brings those two words back (30-min window, only the two words kept, never R); a claim without it keeps the R-978 order (old controllers)**; subprocess stdout is SECRET-BEARING → parsed then zeroed, never logged |
 | `poke.Listener` + `poke.Port` | internal/poke/poke.go | `NewListener(resolve, trigger, port, logger)`; `poke.Port = 51822` | agent-plane immediate-sync (Direction-2a, v0.89.0) | Binds a contentless UDP socket EXCLUSIVELY to the box's WG /32 (`wgtunnel.LoadAssignedAddr`), fires the hub-loop out-of-band trigger. **Port 51822 is a SHARED cross-repo contract** — the hub poke sender + the ep0 `felhom-poke` forced-command target the SAME number; change one → change all three. Contentless (payload ignored), leading-edge debounced (`DebounceWindow`), WG-confined (kernel EKEYREJECTED refuses non-peer /32s). Wired only when `wg_tunnel.enabled` |
 | `wgtunnel.LoadAssignedAddr` | internal/wgtunnel/manager.go | `LoadAssignedAddr(stateDir) (netip.Addr, bool)` | the box's own WG /32 without a Manager | Reads `registered.json`; ok=false until registered; strips the /32 → bare addr (the poke bind target) |
 | `fasttick.Loop` + `fasttick.SourceFunc` | internal/fasttick/fasttick.go | `New(out chan<- struct{}, interval, logger, sources...)`; `Source.Unconverged() (bool, reason)` | agent-plane immediacy SECONDARY (v0.90.0, R-28): pulse the SAME out-of-band trigger every 30 s while ANY source is unconverged, self-disarm on convergence | STATE-BASED (no timer, nothing to journal). Every source MUST be a CACHED read (no exec/network per tick) — `desiredProvider.Generation()`, `reconcile.Engine.LastResult()` (Planned−Pending>0), `pbsdrLoop.PBSDRStatus().State=="waiting_secret"` ONLY, `wgtunnel.Manager.TunnelConvergence()`. The LOUD pbsdr states + destructive `pending_signature` are DELIBERATELY EXCLUDED (a stuck-loud box must not hammer). Pulses the cap-1 channel non-blocking (coalesces with poke/watchdog); NEVER touch `MinPollSeconds`/`clampInterval`/the ticker |
@@ -158,6 +165,7 @@
 | `proxmox.Runner` | internal/proxmox/privileged.go | `*ExecRunner` (direct/sudo) | `mockRunner` internal/proxmox/mock_test.go; runner fakes in storage tests |
 | `storage.HostOps` | internal/storage/hostops.go | `*SudoHostOps` (prod), `NoopHostOps` (degraded) | fakes in internal/storage/observe_test.go, watchdog_test.go |
 | `storage.HostReader` | internal/storage/hostread.go | `*ProcHostReader` | `fakeHostReader` internal/localapi/disks_test.go; internal/storage/role_test.go. v0.87.0: `BlockSlaves(name)` lists `/sys/block/<name>/slaves` (root-free) — backs the `SystemDisks` dm/md walk (`physicalDisksOf`/`walkSlaves`, role.go); per-branch conservatism: an unresolvable slave fails the WHOLE walk → all-system fail-safe. NEVER weaken the signature test `TestSystemDisks_WalkTopologies` (root-backing disk always in the system set). |
+| `storage.IdentityHost` | internal/storage/identity.go | `*SysfsIdentityHost` | fixture /sys + udev trees in identity_test.go (`newFx`); `fakeIdentHost` in internal/localapi/samedisk_test.go. A SEPARATE interface from HostReader on purpose — adding methods to HostReader would break every existing fake. |
 | `localapi.DiskOps` / `StorageGate` / `GuestAttacher` / `GuestLister` | internal/localapi/disks.go | `*storage.SudoHostOps`; `storageGateAdapter` (cmd/felhom-agent/main.go); `*GuestBinder`; `*proxmox.Client` | `fakeDiskOps`/`fakeGate`/`fakeGuestAttacher`/`fakeGuestList` internal/localapi/disks_test.go |
 | `lanresolver.hostRoot` + `dnsmasqUnitPaths` (data seam, R-317) | internal/lanresolver/lanresolver.go | prod `hostRoot = "/"`; probe = the `dnsmasq` package's systemd UNIT, never `/usr/sbin/dnsmasq` (owned by `dnsmasq-base`) | internal/lanresolver/ensure_dnsmasq_test.go — fixture root tree + recording `proxmox.Runner`; the REAL `os.Stat` probe and `EnsureDnsmasq` run. `TestEnsureDnsmasq_ProductionProbeIsTheUnit` pins the production wiring |
 | `localapi.GuestAPI` / `BackupService` / `BackupStore` / `TokenAuthority` | internal/localapi/server.go | `*proxmox.Client`, `*backup.BackupRunner`, `*backup.Store`, `*TokenStore` | `fakeGuests`/`fakeBackups`/`fakeStore` internal/localapi/server_test.go |
@@ -166,6 +174,7 @@
 | `capability` store-grant probe (`storeGrantStatuses` / `storeGrantVerdict` / `Client.Permissions`) | cmd/felhom-agent/main.go, internal/proxmox/query.go | *"may the agent READ this backup tier?"*, one `capability.Status` per configured tier | R-185. **Never infer permission from an empty content listing** — `{"data":[]}` is what a FORBIDDEN tier and a NEWBORN tier both return, and that ambiguity hid an unreadable host tier on both demo boxes. Ask `/access/permissions` **as the agent's own token** (root always says yes). **The ungranted answer is not empty and not a 403** — it carries the privileges inherited from the box-wide `/` grant, so test for **`Datastore.AllocateSpace`** specifically; path-presence or `Datastore.Audit` reports a blinded storage healthy. Probed set comes from `BackupTiers()`, never a fixed list. Critical except the `local` fallback. Composes AROUND the sudo prober (the `poolReadStatus` precedent); `Status`'s wire shape is untouched so the hub alert is free. Unreachable PVE ⇒ degraded, never ok. |
 | `backup.RestoreTestState` | internal/backup/restoretest_state.go | `RecordSuccess(target,archive,tier,verified,t)` / `ProvenArchive(target)` / `ProvenRestoreTests(ctx)` / `LastSuccess(target)` / `OldestFirst(targets)` | Per-tier restore-test PROOF state, persisted (atomic tmp+rename) — **which archive** was proven, and when (R-86) | **Credit ONLY on success** — a permanently failing tier must keep sorting first, or it looks freshly proven and stops being retried. Ties break on target id: without it, two tiers proven in the same second rotate by Go's randomised map order. **This one NEEDS persistence unlike R-84** — R-84 had ground truth to consult (the archive is still on the storage); a restore-test destroys its scratch and leaves no artifact. **R-86: the ARCHIVE is the state, the time is metadata** — a time alone cannot answer "have we proven THIS archive", which is the due-check's whole question. A pre-R-86 file (bare RFC3339 per target) keeps its time and yields NO proven archive, so each tier is due once after the upgrade; reading a legacy time as proof of the current archive would invent a guarantee. **R-189: it is also the REPORTABLE half of the restore-test signal.** The in-memory `backup.Store` holds only this process's latest run, and under per-archive due-ness the agent will not re-test a proven archive — so a proof lost to a restart is not repeated for a whole archive generation (observed live: a passing 14.5 GB offsite restore reached no host-report). `ProvenRestoreTests` renders the stored proofs as `hub.RestoreTest` entries and the collector merges them; a record missing the archive or the tier is NOT emitted, because an unproven tier reading as proven is worse than the defect. **Only successes are stored, deliberately:** a success suppresses future work, a failure causes it. |
 | `backup.BackupSuccessState` | internal/backup/backup_state.go | `RecordBackupSuccess(target, b)` / `LastKnownSuccess(target, vmid)` | Newest SUCCESSFUL backup per tier+guest, persisted (atomic tmp+rename) — the due-check's fallback when the tier's storage cannot be read after a restart (R-894) | **Read ONLY when the storage cannot be read** — a storage that answers is the ground truth (R-84), and an archive absent there must make the tier due even when this file remembers one. A saved copy older than the cadence still reads due. Only successes are written. |
+| localapi backup cadence cap (R-973) | internal/localapi/backup_cap.go | `backupCadenceCap(cadence)`, `Server.capRefuses`, `Server.BackupCapRefusal()` | POST /backup refuses (429) a run the `07` §6 schedule would never ask for: ≤24 h tier → 3 successes per rolling 24 h; longer tier → 1 per max(cadence−24 h, cadence/2) | Counts SUCCESSES only, from this process AND `BackupSuccessState.RecentBackupSuccesses` (so a restart does not reset it); refusals saved (`RecordBackupRefusal`) and raised for 24 h as the CRITICAL host-report status `localapi:backup-cadence-cap` (`backupCapStatus` in main) → the hub's existing `agent_capability_degraded`. Tests `TestBackupCap_*`. |
 | `hub.ProvenRestoreTestReporter` + `Collector.SetProvenRestoreTests` | internal/hub/collect.go | the DURABLE restore-test source, merged with the in-memory one | R-189. Merge rule: **one entry per tier, newest by `TestedAt` wins** — a fresh failure beats a stored success (the failure is the news, and it lives nowhere else), a stored success beats a stale in-memory entry after a restart, and a tier never appears twice (the hub would read two tests). An unparseable timestamp counts as OLDER, so a malformed entry cannot displace a good one. **The wiring is pinned by an AST test** — the method this replaced (`RestoreTestState.Snapshot`) carried a doc comment naming a host-report gauge and had no caller for weeks. |
 | `backup.SpecBuilder` / `backup.TierPicker` / `(*BackupRunner).PickSettledRestoreCandidateOn` | internal/backup/schedule.go, runner.go | `func(ctx,archive) RestoreTestSpec`; `func(ctx,target,notAfter) (archive,landed,error)` | The per-run restore-test spec + per-tier **settled** candidate lookup (R-85, widened by R-86) | The spec is built **PER RUN**, never frozen at construction — the pre-R-85 immediately-invoked value made the offsite tier unschedulable AND went stale on any config change. `SourceTier` comes from **the archive**, never the configured target (the v0.100.0 rule). A tier with no archive returns `("", zero, nil)` — **`""` is NOT an error**, or every fresh box looks broken for its first week. **R-86: `notAfter` is the settle cutoff** (zero = no cutoff, which is what keeps `PickRestoreCandidateOn` a one-line call into it), and the picker now skips entries failing `archivePlausiblyComplete` — under per-archive due-ness an incomplete phantom would be picked forever, fail forever, never earn proof, and make the tier due at EVERY evaluation. |
 | `localapi.BackupTier` + `normalizeBackupTiers` / `config.BackupConfig.BackupTiers` | internal/localapi/backup_tiers.go, internal/config/config.go | `normalizeBackupTiers(tiers, legacy, cadence) []BackupTier`; `BackupTiers() ([]BackupTier, []string)` | THE R-82 multi-tier resolution — one runner per tier, primary first | **The untargeted local-API contract is FROZEN**: no `?target=` ⇒ primary tier ⇒ pre-R-82 response BYTES (Target is `omitempty` and stays empty). Never default a missing cadence — reject it and log the warning at ERROR. Never share one retention knob between tiers. Jobs are keyed by (vmid,target). |

cmd/felhom-agent/backup_cap_r973_test.go

@@ -0,0 +1,67 @@
+package main
+
+import (
+	"go/ast"
+	"strings"
+	"testing"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/capability"
+)
+
+// R-973: a cadence-cap refusal becomes a CRITICAL degraded status — what the hub's existing
+// agent_capability_degraded event (operator-only) keys on — and clears when the refusal ages out.
+func TestBackupCapStatus_RefusalIsCriticalDegraded(t *testing.T) {
+	at := time.Date(2026, 10, 18, 3, 0, 0, 0, time.UTC)
+	s := backupCapStatus(func() (string, int, time.Time, bool) { return "felhom-pbs", 9201, at, true })
+	if !s.Critical || s.Status != capability.StatusDegraded || !strings.Contains(s.Reason, "9201") || !strings.Contains(s.Reason, "felhom-pbs") {
+		t.Fatalf("a refusal must raise a critical degraded status naming the guest and the tier: %+v", s)
+	}
+	if s := backupCapStatus(func() (string, int, time.Time, bool) { return "", 0, time.Time{}, false }); s.Status != capability.StatusOK || !s.Critical {
+		t.Fatalf("no refusal → ok (and still critical, so a later refusal alerts): %+v", s)
+	}
+	if s := backupCapStatus(nil); s.Status != capability.StatusOK {
+		t.Fatalf("no local API → ok: %+v", s)
+	}
+}
+
+// Wiring (the built-but-never-wired class): runDaemon's capability probe appends backupCapStatus, and the
+// late-bound source is set from the local API server.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): delete `backupCapRefusal = localSrv.BackupCapRefusal` →
+// "runDaemon never binds backupCapRefusal to the local API server". Restored.
+func TestR973_CapRefusalIsWiredIntoTheReport(t *testing.T) {
+	_, f := parseMain(t)
+	var appended, bound bool
+	for _, d := range f.Decls {
+		fd, ok := d.(*ast.FuncDecl)
+		if !ok || fd.Name == nil || fd.Name.Name != "runDaemon" || fd.Body == nil {
+			continue
+		}
+		ast.Inspect(fd.Body, func(n ast.Node) bool {
+			switch x := n.(type) {
+			case *ast.CallExpr:
+				if id, ok := x.Fun.(*ast.Ident); ok && id.Name == "backupCapStatus" && len(x.Args) == 1 {
+					if a, ok := x.Args[0].(*ast.Ident); ok && a.Name == "backupCapRefusal" {
+						appended = true
+					}
+				}
+			case *ast.AssignStmt:
+				if len(x.Lhs) == 1 && len(x.Rhs) == 1 {
+					if l, ok := x.Lhs[0].(*ast.Ident); ok && l.Name == "backupCapRefusal" {
+						if sel, ok := x.Rhs[0].(*ast.SelectorExpr); ok && sel.Sel.Name == "BackupCapRefusal" {
+							bound = true
+						}
+					}
+				}
+			}
+			return true
+		})
+	}
+	if !appended {
+		t.Fatal("runDaemon's capability probe does not call backupCapStatus(backupCapRefusal)")
+	}
+	if !bound {
+		t.Fatal("runDaemon never binds backupCapRefusal to the local API server")
+	}
+}

cmd/felhom-agent/escrow_r978_test.go

@@ -0,0 +1,112 @@
+package main
+
+import (
+	"context"
+	"encoding/json"
+	"go/ast"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/config"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/escrow"
+)
+
+// R-978, the root half: the controller-driven ceremony (json mode, the FELHOM_ESCROW argv with
+// `--upload`) must hand its package back instead of sending it — the daemon sends it at the claim.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): set `deferUpload: false` in escrowOptsFromCLI → this fails
+// with "json mode must defer the upload (R-978): got deferUpload=false", and the delivery test's json leg
+// then sends the package at create time ("put called 1 times"). Restored.
+func TestEscrowOptsFromCLI_JSONModeDefersUpload(t *testing.T) {
+	// The FELHOM_ESCROW argv: --selftest=escrow-create --upload --output=json.
+	if o := escrowOptsFromCLI("", false, false, true, "", true); !o.deferUpload || !o.upload {
+		t.Fatalf("json mode must defer the upload (R-978): got deferUpload=%v upload=%v", o.deferUpload, o.upload)
+	}
+	// Text mode (an operator at a terminal) keeps the immediate upload; no --upload means no upload at all.
+	if o := escrowOptsFromCLI("", false, false, true, "", false); o.deferUpload {
+		t.Fatal("text mode must upload at once")
+	}
+	if o := escrowOptsFromCLI("", false, false, false, "", true); o.deferUpload {
+		t.Fatal("without --upload there is nothing to defer")
+	}
+}
+
+func TestDeliverEscrowPackage_JSONModeSendsNothing(t *testing.T) {
+	cfg := config.Config{}
+	cfg.Hub.URL, cfg.Hub.HostID, cfg.Hub.APIKey = "https://hub.invalid", "host-1", "k"
+	res := escrow.CreateResult{Blob: []byte("ciphertext"), KeyFingerprint: "f2:87", Posture: escrow.PostureZeroKnowledge, IdentityBlob: []byte("id")}
+
+	for _, jsonMode := range []bool{true, false} {
+		calls := 0
+		var sent []byte
+		put := func(_ context.Context, _ config.Config, body []byte) error { calls++; sent = body; return nil }
+		o := escrowOptsFromCLI("", false, false, true, "", jsonMode)
+		pending, uploaded, err := deliverEscrowPackage(context.Background(), cfg, res, "abc", o.deferUpload, put)
+		if err != nil {
+			t.Fatal(err)
+		}
+		if jsonMode {
+			if calls != 0 || uploaded {
+				t.Fatalf("json mode: put called %d times (uploaded=%v), want 0 — the package must wait for the claim", calls, uploaded)
+			}
+			var m map[string]any
+			if json.Unmarshal(pending, &m) != nil || m["blob_b64"] == "" || m["identity_blob_b64"] == "" || m["restic_pw_sha256"] != "abc" {
+				t.Fatalf("json mode: the pending package is not the hub upload body: %s", pending)
+			}
+		} else if calls != 1 || !uploaded || pending != nil || len(sent) == 0 {
+			t.Fatalf("text mode: put called %d times, uploaded=%v, pending=%q — want one immediate upload", calls, uploaded, pending)
+		}
+	}
+
+	// A hub that is not configured fails in both modes (as the pre-R-978 upload did).
+	if _, _, err := deliverEscrowPackage(context.Background(), config.Config{}, res, "", true, nil); err == nil {
+		t.Fatal("an unconfigured hub must fail the ceremony even when the upload is deferred")
+	}
+}
+
+// R-978 wiring (the built-but-never-wired class): runDaemon's localapi.EscrowCeremonyConfig carries an
+// Upload that calls putEscrowBody. Without it every claim of a new package answers 503 and no customer
+// can finish a ceremony.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): delete the `Upload:` line in runDaemon → this fails with
+// "runDaemon's EscrowCeremonyConfig has no Upload calling putEscrowBody". Restored.
+func TestR978_CeremonyUploadIsWiredIntoTheDaemon(t *testing.T) {
+	_, f := parseMain(t)
+	if !callsWithin(f, "main")["runDaemon"] {
+		t.Fatal("main → runDaemon is broken — the path this test asserts is not the live one")
+	}
+	wired := false
+	for _, d := range f.Decls {
+		fd, ok := d.(*ast.FuncDecl)
+		if !ok || fd.Name == nil || fd.Name.Name != "runDaemon" || fd.Body == nil {
+			continue
+		}
+		ast.Inspect(fd.Body, func(n ast.Node) bool {
+			cl, ok := n.(*ast.CompositeLit)
+			if !ok {
+				return true
+			}
+			sel, ok := cl.Type.(*ast.SelectorExpr)
+			if !ok || sel.Sel.Name != "EscrowCeremonyConfig" {
+				return true
+			}
+			for _, el := range cl.Elts {
+				if kv, ok := el.(*ast.KeyValueExpr); ok {
+					if k, ok := kv.Key.(*ast.Ident); ok && k.Name == "Upload" {
+						ast.Inspect(kv.Value, func(n ast.Node) bool {
+							if ce, ok := n.(*ast.CallExpr); ok {
+								if id, ok := ce.Fun.(*ast.Ident); ok && id.Name == "putEscrowBody" {
+									wired = true
+								}
+							}
+							return true
+						})
+					}
+				}
+			}
+			return true
+		})
+	}
+	if !wired {
+		t.Fatal("runDaemon's EscrowCeremonyConfig has no Upload calling putEscrowBody")
+	}
+}

cmd/felhom-agent/main.go

@@ -59,6 +59,8 @@ import (
 	"gitea.dooplex.hu/admin/felhom-agent/internal/selfupdate"
 	"gitea.dooplex.hu/admin/felhom-agent/internal/signedjobs"
 	"gitea.dooplex.hu/admin/felhom-agent/internal/storage"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/thincheck"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/vzpartial"
 	"gitea.dooplex.hu/admin/felhom-agent/internal/wgtunnel"
 )
 
@@ -192,7 +194,7 @@ func main() {
 	flag.StringVar(&pbsStorage, "storage", "", "for --selftest=escrow-create: the pbs storage whose key to escrow (default: escrow.pbs_storage_id)")
 	flag.BoolVar(&paperkey, "paperkey", false, "for --selftest=escrow-create: ALSO emit the raw-key paperkey (opt-in (a); single-factor, unrevocable)")
 	flag.BoolVar(&offline, "offline", false, "for --selftest=escrow-create: ALSO emit the R-wrapped offline copy to print (opt-in (b))")
-	flag.BoolVar(&upload, "upload", false, "for --selftest=escrow-create: upload the opaque blob to the hub")
+	flag.BoolVar(&upload, "upload", false, "for --selftest=escrow-create: upload the opaque blob to the hub (with --output=json: hand the package back in pending_upload — the agent daemon uploads it when the code is claimed, R-978)")
 	flag.StringVar(&blobPath, "blob", "", "for --selftest=escrow-consume: path to the R-wrapped escrow blob file")
 	flag.StringVar(&expectedFP, "fingerprint", "", "for --selftest=escrow-consume: the EXPECTED key fingerprint (the gate target)")
 	flag.StringVar(&keyDest, "keydest", "", "for --selftest=escrow-consume: where to install the recovered key (0600)")
@@ -402,6 +404,31 @@ func (r *gateRemounter) Remount(ctx context.Context, t storage.KnownTarget) {
 // poolReadStatus probes the felhom-pool membership read (GET /pools/felhom) the stale-lock reaper's
 // ownership scoping depends on (v0.62.0, audit A1). Needs `Pool.Audit` at `/pool/felhom` — granted
 // by host-install v1.9.0; on an older ACL this reports degraded and the reaper fail-safes (skips).
+// backupCapStatusName is the host-report status the R-973 cadence cap raises.
+const backupCapStatusName = "localapi:backup-cadence-cap"
+
+// backupCapStatus turns the local-API backup cadence cap's newest refusal (R-973) into a CRITICAL
+// capability status: degraded for 24 h after a refusal, so the hub raises its operator-only
+// whole_guest_backup_refused event (hub v0.159.0, R-1150; agent_capability_degraded before). The Reason names the guest, the tier and the time, and
+// says what to do. nil source (local API off) → ok.
+func backupCapStatus(refusal func() (string, int, time.Time, bool)) capability.Status {
+	s := capability.Status{
+		Name:     backupCapStatusName,
+		Feature:  "whole-guest restore points: a guest asked for more whole-guest backups than its schedule allows, and the extra ones were refused (R-973)",
+		Critical: true,
+		Status:   capability.StatusOK,
+	}
+	if refusal == nil {
+		return s
+	}
+	if target, vmid, at, ok := refusal(); ok {
+		s.Status = capability.StatusDegraded
+		s.Reason = fmt.Sprintf("refused a whole-guest backup of guest %d on target %q at %s — more than the schedule allows; the guest's controller may be misbehaving or compromised: check the guest",
+			vmid, target, at.UTC().Format(time.RFC3339))
+	}
+	return s
+}
+
 func poolReadStatus(ctx context.Context, px *proxmox.Client) capability.Status {
 	s := capability.Status{
 		Name:     "pve:pool-read",
@@ -776,6 +803,7 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 	hostReader := storage.NewProcHostReader()
 	hostOps := newHostOps(cfg, logger)
 	observer := storage.NewObserver(px, hostReader, hostOps, logger)
+	observer.SetIdentity(storage.NewIdentityResolver()) // `16` §3 physical-disk identity
 	// Backup + restore-test state store (slice 6): holds the latest backup-per-target +
 	// latest restore-test result; the collector reads it via the BackupReporter /
 	// RestoreTestReporter seams; the cadence scheduler writes it.
@@ -827,8 +855,13 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 		run: (&proxmox.ExecRunner{Mode: proxmox.RunnerMode(cfg.Privileged.Mode)}).Run,
 		log: logger,
 	}
+	// R-973: the local-API backup cadence cap's refusals ride the report as a CRITICAL status, so the hub's
+	// operator-only whole_guest_backup_refused event (hub v0.159.0, R-1150) tells the operator. Late-bound like
+	// drConfigured: the local API server is built further down, before the loop starts.
+	var backupCapRefusal func() (string, int, time.Time, bool)
 	probeAll := func(ctx context.Context) []capability.Status {
 		out := append(capProber.Probe(ctx), poolReadStatus(ctx, px))
+		out = append(out, backupCapStatus(backupCapRefusal))
 		return append(out, storeGrantStatuses(ctx, px, cfg, grantRepairer)...)
 	}
 	// (The startup self-check log runs AFTER the pbsdr manager is wired below, so its snapshot
@@ -957,10 +990,7 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 	// state. Gates the watchdog's self-heal re-mount to ENROLLED drives, and the local API records
 	// enroll/eject into it. Open failure → ungated legacy remount (logged) so the daemon still runs.
 	var intentReader storage.IntentReader
-	intentStateDir := "/var/lib/felhom-agent"
-	if cfg.Authz.NonceStorePath != "" {
-		intentStateDir = filepath.Dir(cfg.Authz.NonceStorePath) // share the agent's durable state dir
-	}
+	intentStateDir := agentStateDir(cfg)
 	intentStore, ierr := storage.OpenIntentStore(filepath.Join(intentStateDir, "drive-intents.json"))
 	if ierr != nil {
 		logger.Warn("storage: intent store unavailable — self-heal runs UNGATED (legacy remount-any)", "err", ierr)
@@ -1092,6 +1122,9 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 		decommIntent = intentStore
 	}
 	decommExec := signedjobs.NewDecommissionExecutor(decommIntent, logger)
+	if guestBindStore != nil { // typed-nil guard, as above
+		decommExec.SetGuestBinds(guestBindStore) // v0.159.1: prune the bind record like the self-serve decommission
+	}
 
 	// Agent self-update (TASK D1): the agent_update executor downloads the operator-signed binary,
 	// verifies it against the SIGNED sha256, and hands it to the root guarded wrapper (A/B flip +
@@ -1166,7 +1199,11 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 			}
 			logger.Warn("osupdate: capability probe after the config bundle", "ok", ok, "total", total, "degraded", strings.Join(names, ","))
 		}}
-	jobsRunner := signedjobs.NewRunner(client, gate, signedjobs.ExecutorChain{wipeExec, decommExec, updateExec, dockerExec, pveExec, kernelExec, bundleExec}, cfg.Hub.HostID, logger)
+	// v0.158.0 (felhom.eu `16` §9): the internal data place on an old box — the root helper, then the mount (late-bound
+	// to the local API server built below).
+	internalPlaceExec := signedjobs.NewInternalPlaceExecutor(suRunner,
+		localapi.InternalPlaceFileIn(intentStateDir), logger)
+	jobsRunner := signedjobs.NewRunner(client, gate, signedjobs.ExecutorChain{wipeExec, decommExec, updateExec, dockerExec, pveExec, kernelExec, bundleExec, internalPlaceExec}, cfg.Hub.HostID, logger)
 	loop.SetEnvelopeObserver(hub.MultiObserver(desiredSyncer, jobsRunner))
 
 	// Controller-driven escrow ceremony (v0.88.0): static config facts + the LATE-BOUND DR gate —
@@ -1190,6 +1227,9 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 			return cfg.Escrow.PBSStorageID
 		},
 		HubConfigured: cfg.Hub.URL != "" && cfg.Hub.HostID != "" && cfg.Hub.APIKey != "",
+		// R-978: the ceremony hands its package back; the claim uploads it with the same PUT the root
+		// subprocess used, under the same per-host key, before the code is handed out.
+		Upload: func(ctx context.Context, pkg []byte) error { return putEscrowBody(ctx, cfg, pkg) },
 		DRConfigured: func() bool {
 			if drConfigured != nil {
 				return drConfigured()
@@ -1199,6 +1239,8 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 	}
 	localSrv := buildLocalAPIServer(cfg, px, backupStore, heavyOps, observer, driveKnown, hostOps, gate, collector, client, intentRec, guestBindStore, formatJobStore, logRing, escrowCeremonyCfg, logger, &localTokens)
 	if localSrv != nil {
+		backupCapRefusal = localSrv.BackupCapRefusal // R-973
+		internalPlaceExec.SetEnsure(localSrv.EnsureInternalPlace)
 		// R-998 (doc 14): POST /host/poweroff — guest shutdown, then host power-off; guards in hostpoweroff.go.
 		pwMode := proxmox.RunnerMode(cfg.Privileged.Mode)
 		if pwMode == "" {
@@ -1417,7 +1459,13 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 		// filesystem UUID and re-mounts (idempotent `enable --now`) so a letter reshuffle is a no-op.
 		// Type-asserted (the concrete op exposes it; the interface stays lean).
 		mountReasserter := hostOps // concrete *storage.SudoHostOps — exposes ReassertEnrolledMounts
+		// v0.158.0 (felhom.eu `16` §5): the internal data place is mounted like an enrolled drive — first, so the
+		// re-assert below sees its unit.
+		localSrv.EnsureInternalPlace(ctx)
 		if mountReasserter != nil {
+			// R-1124: a decommissioned drive that is gone loses its host mount unit first (start-up and every tick), so
+			// no later boot waits for it and the re-assert below no longer lists it.
+			mountReasserter.ForgetRetiredDrives(ctx, intentReader)
 			mountReasserter.ReassertEnrolledMounts(ctx)
 		}
 		// F9: on startup (the host's bring-up/reconcile trigger), re-assert any enrolled guest data-drive
@@ -1454,7 +1502,9 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 				case <-t.C:
 					// Re-assert mounts first (handles a USB that enumerated late after a host reboot, or a
 					// unit re-disabled at runtime), then re-assert the guest binds against the now-live mounts.
+					localSrv.EnsureInternalPlace(ctx) // `16` §5: a record written since the last tick (the signed job)
 					if mountReasserter != nil {
+						mountReasserter.ForgetRetiredDrives(ctx, intentReader) // R-1124: decommissioned + gone → unit removed
 						mountReasserter.ReassertEnrolledMounts(ctx)
 					}
 					localSrv.ReassertGuestBinds(ctx)
@@ -1532,6 +1582,8 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 		// the dangling snapshot → start iff onboot). Runs BEFORE the backup loop starts, so a present
 		// backup lock is stale by definition (guarded by a no-vzdump-running check; fail-safe otherwise).
 		localSrv.RecoverStaleLockedGuests(ctx)
+		// R-1006: then remove a dead half-written backup a power cut left (the lock recovery first: same order as a human).
+		localSrv.SweepVzdumpPartials(ctx)
 		// F-REBOOT: the startup recovery above only covers a guest left LOCKED by an interrupted
 		// backup. A guest that simply ends up stopped-and-unlocked (a `pct reboot` whose shutdown
 		// half completed and whose start half never fired — Campaign 8 fault 11, 9m47s of total
@@ -1892,11 +1944,10 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 	}
 	apiTiers := make([]localapi.BackupTier, 0, len(backupTiers))
 	var runner *backup.BackupRunner
+	var tierRunners []*backup.BackupRunner
+	var tierTargets []string
 	for _, t := range backupTiers {
-		prune := ""
-		if t.KeepLast > 0 {
-			prune = fmt.Sprintf("keep-last=%d", t.KeepLast)
-		}
+		prune := t.PruneSpec() // primary: keep-daily=N (R-1149); an extra tier: keep-last=N; 0 → no prune
 		// Pruning a PBS target is allowed ONLY for an additional tier with an explicit keep_last
 		// (the primary's target AND retention both default, so it could prune the DR by accident).
 		allowPBSPrune := !t.Primary && t.KeepLast > 0
@@ -1904,6 +1955,8 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 		if t.Primary {
 			runner = r
 		}
+		tierRunners = append(tierRunners, r)
+		tierTargets = append(tierTargets, t.TargetID)
 		apiTiers = append(apiTiers, localapi.BackupTier{
 			TargetID:    t.TargetID,
 			Cadence:     t.Cadence,
@@ -1912,7 +1965,7 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 			Service:     r,
 		})
 		logger.Info("backup tier armed", "target", t.TargetID, "cadence", t.Cadence.String(),
-			"keep_last", t.KeepLast, "wait_timeout", t.WaitTimeout.String(),
+			"keep_last", t.KeepLast, "prune", prune, "wait_timeout", t.WaitTimeout.String(),
 			"prune_pbs_allowed", allowPBSPrune, "primary", t.Primary)
 	}
 	// Guest data-drive passthrough (slice 10 P2): a root-CLI runner for the `pct set` bind + chown
@@ -1993,6 +2046,7 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 		DriveTargets:     driveTargets,                    // Impl-2a: registry+units drives for the /disks view (union w/ Observe storages)
 		Smart:            storage.NewSmartReader(hostOps), // v0.95.0 Fix B: SMART for the union-path drives
 		HostReader:       storage.NewProcHostReader(),     // Impl-2b: durableIDForMount raw-mount fallback + role gate
+		Identity:         storage.NewIdentityResolver(),   // v0.157.0: `16` §3 identity on /disks + POST /disks/same-disk
 		Tokens:           tokens,
 		BackupCadence:    cfg.Backup.BackupCadence(),
 		// Disk management (slice 8C): the privileged host surface + the data-bearing wipe gate.
@@ -2014,10 +2068,12 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 		// F2-b: recover a guest left with a stale vzdump lock by a reboot-during-backup. Reads + start
 		// go through the API client; the `pct unlock` is the one fenced root-CLI op (no API equivalent).
 		// A1 (v0.62.0): the scan is restricted to felhom-pool members (ownership proven, not assumed).
-		StaleLock:  localapi.NewStaleLockController(px, &proxmox.ExecRunner{Mode: gaMode, SudoPath: cfg.Privileged.SudoPath}, reconcile.DefaultPool, logger),
-		Intent:     intent,     // slice 10 P3: record enroll/eject intent for self-heal
-		GuestBinds: guestBinds, // F9: per-guest bind record for the startup re-assert
-		FormatJobs: formatJobs, // F20-BUG3: detached-format job record + restart recovery
+		StaleLock: localapi.NewStaleLockController(px, &proxmox.ExecRunner{Mode: gaMode, SudoPath: cfg.Privileged.SudoPath}, reconcile.DefaultPool, logger),
+		Intent:    intent, // slice 10 P3: record enroll/eject intent for self-heal
+		// v0.158.0 (felhom.eu `16` §5): the internal data place record, beside the intent store.
+		InternalPlaceFile: localapi.InternalPlaceFileIn(agentStateDir(cfg)),
+		GuestBinds:        guestBinds, // F9: per-guest bind record for the startup re-assert
+		FormatJobs:        formatJobs, // F20-BUG3: detached-format job record + restart recovery
 
 		// Host metrics (slice 9): the shared collector serves GET /host/metrics — a fresh host +
 		// per-storage view to the customer's monitoring page (reuses the slice-4 collector).
@@ -2032,6 +2088,30 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 		logger.Warn("daemon: local-api disabled (server build)", "err", err)
 		return nil
 	}
+	// R-1006: a dead half-written vzdump (a power cut mid-backup) is removed at start-up and before every backup of the
+	// guest on that target, through the root helper; the in-flight answer is R-1008's (a task from before the boot is dead).
+	cleaner := &vzpartial.Cleaner{
+		API:      px,
+		Runner:   &proxmox.ExecRunner{Mode: gaMode, SudoPath: cfg.Privileged.SudoPath},
+		InFlight: srv.VzdumpInFlight,
+		Boot: func() (int64, bool) {
+			b, err := strconv.ParseInt(poweroff.ProcBtime(), 10, 64)
+			return b, err == nil && b > 0
+		},
+		Logger: logger,
+	}
+	// R-1012: an orphan thin device (a full host root during a snapshot) is NAMED in the backup failure the controller
+	// mails to the operator, through the read-only root helper. Never removed — the repair stays a human step.
+	thinChecker := &thincheck.Checker{API: px, Runner: &proxmox.ExecRunner{Mode: gaMode, SudoPath: cfg.Privileged.SudoPath}, Logger: logger}
+	for _, r := range tierRunners {
+		r.SetBeforeBackup(func(ctx context.Context, storage string, vmid int) { cleaner.Clean(ctx, storage, vmid) })
+		r.SetThinCheck(func(ctx context.Context, vmid int) string { return thincheck.Describe(thinChecker.Check(ctx, vmid)) })
+	}
+	srv.SetPartialSweep(func(ctx context.Context, vmid int) {
+		for _, t := range tierTargets {
+			cleaner.Clean(ctx, t, vmid)
+		}
+	})
 	return srv
 }
 
@@ -2161,6 +2241,7 @@ func runSelftestHub(ctx context.Context, cfg config.Config, logger *slog.Logger)
 		return 1
 	}
 	observer := storage.NewObserver(px, storage.NewProcHostReader(), newHostOps(cfg, logger), logger)
+	observer.SetIdentity(storage.NewIdentityResolver()) // `16` §3 physical-disk identity
 	// Wire the LIVE PBS reporter here too (v0.39.0): selftest=hub is a separate one-shot process — no
 	// verify loop runs — so a nil reporter previously yielded pbs_snapshots:[] and an absent dr_recipe
 	// pbs coord. The live reporter lists snapshots directly (fresh store, last-known-good fallback) so
@@ -2217,6 +2298,7 @@ func runSelftestStorage(ctx context.Context, cfg config.Config, logger *slog.Log
 	hostReader := storage.NewProcHostReader()
 	hostOps := newHostOps(cfg, logger)
 	observer := storage.NewObserver(px, hostReader, hostOps, logger)
+	observer.SetIdentity(storage.NewIdentityResolver()) // `16` §3 physical-disk identity
 
 	octx, cancel := context.WithTimeout(ctx, 60*time.Second)
 	fmt.Printf("=== felhom-agent %s selftest=storage ===\n", version)
@@ -2775,6 +2857,20 @@ type escrowCeremonyOpts struct {
 	offline            bool
 	upload             bool
 	identityBundlePath string
+	// deferUpload (R-978): with upload, build the hub upload body and hand it back in
+	// CeremonyOutput.PendingUpload instead of sending it. Set for json mode — the controller-driven
+	// ceremony, whose daemon uploads the package only when the customer claims the new code.
+	deferUpload bool
+}
+
+// escrowOptsFromCLI maps the escrow-create flags onto the ceremony options. json mode defers the upload
+// (R-978, pinned by TestEscrowOptsFromCLI_JSONModeDefersUpload); text mode — an operator at a terminal,
+// who sees R at once — keeps the immediate upload.
+func escrowOptsFromCLI(storage string, paperkey, offline, upload bool, identityBundlePath string, jsonMode bool) escrowCeremonyOpts {
+	return escrowCeremonyOpts{
+		storage: storage, paperkey: paperkey, offline: offline, upload: upload,
+		identityBundlePath: identityBundlePath, deferUpload: upload && jsonMode,
+	}
 }
 
 // escrowCeremonyOutcome is the shared core's result. R is the ONLY secret; Sum mirrors the
@@ -2925,16 +3021,36 @@ func escrowCeremony(ctx context.Context, cfg config.Config, logger *slog.Logger,
 		ResticPwSealed:    resticStaged,
 	}
 	if opts.upload {
-		if err := uploadEscrowBlob(ctx, cfg, res, resticPwSHA256); err != nil {
+		pending, uploaded, err := deliverEscrowPackage(ctx, cfg, res, resticPwSHA256, opts.deferUpload, putEscrowBody)
+		if err != nil {
 			// R is minted and the blob self-verified — only the hub leg failed. kind "upload" lets
 			// the text shell keep the pre-extraction order (R surfaced, THEN the failure).
 			return out, &escrowCeremonyErr{kind: "upload", err: err}
 		}
-		out.Sum.Uploaded = true
+		out.Sum.PendingUpload, out.Sum.Uploaded = pending, uploaded
 	}
 	return out, nil
 }
 
+// deliverEscrowPackage builds the hub upload body and either sends it now (text mode) or returns it for
+// the caller to send later (deferUpload — json mode, R-978: the daemon sends it when the customer claims
+// the code). A hub that is not configured fails in both modes, as before: a package that can never be
+// delivered is not a ceremony. put is the sender seam (putEscrowBody in production).
+func deliverEscrowPackage(ctx context.Context, cfg config.Config, res escrow.CreateResult, resticPwSHA256 string, deferUpload bool,
+	put func(ctx context.Context, cfg config.Config, body []byte) error) (pending []byte, uploaded bool, err error) {
+	if cfg.Hub.URL == "" || cfg.Hub.HostID == "" || cfg.Hub.APIKey == "" {
+		return nil, false, fmt.Errorf("hub not configured (url/host_id/api_key)")
+	}
+	body := escrowUploadBody(res, resticPwSHA256)
+	if deferUpload {
+		return body, false, nil
+	}
+	if err := put(ctx, cfg, body); err != nil {
+		return nil, false, err
+	}
+	return nil, true, nil
+}
+
 // printEscrowTextBanner prints the text-mode header line (shared by the success and the
 // post-banner failure paths so the pre-extraction output stays byte-identical — including the
 // posture: the historical banner always printed the DEFAULT posture, not the result's).
@@ -2998,10 +3114,7 @@ func runSelftestEscrowCreate(ctx context.Context, cfg config.Config, logger *slo
 		return 2
 	}
 
-	out, cerr := escrowCeremony(ctx, cfg, logger, escrowCeremonyOpts{
-		storage: storage, paperkey: paperkey, offline: offline, upload: upload,
-		identityBundlePath: identityBundlePath,
-	})
+	out, cerr := escrowCeremony(ctx, cfg, logger, escrowOptsFromCLI(storage, paperkey, offline, upload, identityBundlePath, jsonMode))
 	if cerr != nil {
 		switch cerr.kind {
 		case "usage":
@@ -3039,6 +3152,9 @@ func runSelftestEscrowCreate(ctx context.Context, cfg config.Config, logger *slo
 		if out.Sum.Uploaded {
 			fmt.Fprintln(os.Stderr, "  uploaded the opaque blob(s) to the hub (host record); the hub cannot open them")
 		}
+		if len(out.Sum.PendingUpload) > 0 {
+			fmt.Fprintln(os.Stderr, "  package prepared, NOT uploaded: the agent uploads it when the new code is claimed (R-978)")
+		}
 		wire := out.Sum
 		wire.RecoveryCode = out.R
 		if err := json.NewEncoder(os.Stdout).Encode(wire); err != nil {
@@ -3192,13 +3308,10 @@ type escrowUploadRequest struct {
 	ResticPwSHA256 string `json:"restic_pw_sha256,omitempty"`
 }
 
-// uploadEscrowBlob PUTs the opaque blob (and, for 10D, the identity blob) to
-// the hub, authed with the per-host key. The hub stores ciphertext + non-secret fields; no usable
-// secret leaves the agent.
-func uploadEscrowBlob(ctx context.Context, cfg config.Config, res escrow.CreateResult, resticPwSHA256 string) error {
-	if cfg.Hub.URL == "" || cfg.Hub.HostID == "" || cfg.Hub.APIKey == "" {
-		return fmt.Errorf("hub not configured (url/host_id/api_key)")
-	}
+// escrowUploadBody is the hub upload body for one ceremony's result: the opaque blob (and, for 10D, the
+// identity blob) plus non-secret fields. Ciphertext only — useless without R; no usable secret leaves
+// the agent. putEscrowBody sends it, authed with the per-host key.
+func escrowUploadBody(res escrow.CreateResult, resticPwSHA256 string) []byte {
 	upReq := escrowUploadRequest{
 		BlobB64:        base64.StdEncoding.EncodeToString(res.Blob),
 		KeyFingerprint: res.KeyFingerprint,
@@ -3210,6 +3323,15 @@ func uploadEscrowBlob(ctx context.Context, cfg config.Config, res escrow.CreateR
 		upReq.IdentityBlobB64 = base64.StdEncoding.EncodeToString(res.IdentityBlob)
 	}
 	body, _ := json.Marshal(upReq)
+	return body
+}
+
+// putEscrowBody PUTs an escrow upload body to this host's hub record (the per-host key). Used by the
+// text-mode ceremony at once, and by the daemon when the customer claims the code (R-978).
+func putEscrowBody(ctx context.Context, cfg config.Config, body []byte) error {
+	if cfg.Hub.URL == "" || cfg.Hub.HostID == "" || cfg.Hub.APIKey == "" {
+		return fmt.Errorf("hub not configured (url/host_id/api_key)")
+	}
 	url := strings.TrimRight(cfg.Hub.URL, "/") + "/api/v1/hosts/" + cfg.Hub.HostID + "/escrow"
 	req, err := http.NewRequestWithContext(ctx, http.MethodPut, url, bytes.NewReader(body))
 	if err != nil {
@@ -3874,3 +3996,12 @@ func (f *factsReporter) SystemFacts(ctx context.Context) (int, json.RawMessage,
 	f.facts, f.err = f.leg.Facts(ctx, f.vmid)
 	return f.vmid, f.facts, f.err
 }
+
+// agentStateDir is the agent's durable state dir: beside the nonce store when one is configured, else the default.
+// The intent store, the guest-bind record and the internal place record (v0.158.0) all live here.
+func agentStateDir(cfg config.Config) string {
+	if cfg.Authz.NonceStorePath != "" {
+		return filepath.Dir(cfg.Authz.NonceStorePath)
+	}
+	return "/var/lib/felhom-agent"
+}

cmd/felhom-opsign/main.go

@@ -43,7 +43,7 @@ func main() {
 
 func run() error {
 	var (
-		op        = flag.String("op", "", "op class to sign, e.g. storage_wipe | guest_destroy | decommission | agent_update | os_docker_step | os_pve_step | os_kernel_step | agent_config_update")
+		op        = flag.String("op", "", "op class to sign, e.g. storage_wipe | guest_destroy | decommission | agent_update | os_docker_step | os_pve_step | os_kernel_step | agent_config_update | add_internal_place")
 		host      = flag.String("host", "", "target host_id (anti-retarget — the op runs ONLY on this host)")
 		guest     = flag.String("guest", "", "target guest_id (\"\" = host-scoped op)")
 		keyID     = flag.String("key-id", "", "key id of the signing key (must match a pinned agent signer)")

configs/felhom-agent.sudoers

@@ -25,6 +25,7 @@
 
 Cmnd_Alias FELHOM_MOUNT = \
   /usr/local/sbin/felhom-priv-apply ^unit mnt-[A-Za-z0-9_.\\-]+\.(mount|automount)$, \
+  /usr/local/sbin/felhom-priv-apply ^unit-forget mnt-[A-Za-z0-9_.\\-]+\.mount$, \
   /usr/bin/systemctl daemon-reload, \
   /usr/bin/systemctl ^enable --now -- mnt-[A-Za-z0-9_.\\-]+\.mount$, \
   /usr/bin/systemctl ^disable -- mnt-[A-Za-z0-9_.\\-]+\.mount$, \
@@ -88,6 +89,27 @@ Cmnd_Alias FELHOM_GUESTHOOK = \
 Cmnd_Alias FELHOM_POWEROFF = \
   /usr/bin/systemctl poweroff
 
+# Dead half-written vzdump removal (R-1006, power-cut-fixes-2026-10-14): a power cut mid-backup leaves
+# vzdump-lxc-<vmid>-<date>.tar.dat + .tmp/ on the dir storage; nothing removed them and every later backup was skipped
+# „not enough space". ONE root helper (configs/felhom-vzdump-partial-clean, from the config bundle) re-checks everything
+# as root — a dir storage from storage.cfg, the exact name of THAT vmid, not in use, dead — and refuses otherwise. The
+# arguments here are its exact shape (storage id, vmid, one half-written name); the vmid-in-name match is the helper's.
+Cmnd_Alias FELHOM_VZPARTIAL = \
+  /usr/local/sbin/felhom-vzdump-partial-clean ^[A-Za-z][A-Za-z0-9_-]{0\,63} [1-9][0-9]{2\,8} vzdump-(lxc|qemu)-[1-9][0-9]{2\,8}-[0-9]{4}_[0-9]{2}_[0-9]{2}-[0-9]{2}_[0-9]{2}_[0-9]{2}\.(tar\.dat|vma\.dat|tmp)$
+
+# Internal data place („Belső tárhely", felhom.eu architecture/16 §5, §7; agent v0.158.0). ONE root helper
+# (configs/felhom-internal-place, from the config bundle) creates <vg>/felhom-belso in the local-lvm thin pool by the
+# ruled size rule and formats ONLY that volume, ONLY when it carries no signature. The agent calls it from the signed
+# `add_internal_place` job (an old box), always with 0 GiB planned: its guest's volumes already exist. `plan` changes
+# nothing. The installer runs the helper as root itself (a new box).
+# R-1012 (agent v0.159.0): the READ-ONLY orphan-thin check before and after a whole-guest backup — lists the pool's thin
+# device ids (through a metadata snapshot it reserves and releases) next to LVM's. It sends no other pool message; the
+# repair stays a human step. Pinned by configs/test_felhom_thin_check.py.
+Cmnd_Alias FELHOM_THINCHECK = \
+  /usr/local/sbin/felhom-thin-check ^[A-Za-z0-9_+][A-Za-z0-9_.+-]{0\,63} [A-Za-z0-9_+][A-Za-z0-9_.+-]{0\,63}$
+Cmnd_Alias FELHOM_INTERNALPLACE = \
+  /usr/local/sbin/felhom-internal-place ^(plan|create) 0$
+
 # Intermediary mount model (the drive hot-swap re-architecture). The agent keeps a SHARED host parent
 # /mnt/felhom-drives (self-bind + make-shared) and binds/unbinds each drive's felhom-data namespace UNDERNEATH it so the
 # change propagates into the running guest live. The boot-persistence script + unit are FIXED files from the config
@@ -320,4 +342,4 @@ Cmnd_Alias FELHOM_GUESTNET = \
   /usr/sbin/pct ^exec [0-9]+ -- pgrep -x dhclient$, \
   /usr/sbin/pct ^exec [0-9]+ -- dhclient -pf /run/dhclient\.eth0\.pid -lf /var/lib/dhcp/dhclient\.eth0\.leases eth0$
 
-felhom-agent ALL=(root) NOPASSWD: FELHOM_MOUNT, FELHOM_DISK, FELHOM_PROVISION, FELHOM_FORMAT, FELHOM_DNSMASQ, FELHOM_GUESTHOOK, FELHOM_INTERMEDIARY, FELHOM_CONTROLLERSWAP, FELHOM_STALELOCK, FELHOM_FSTRIM, FELHOM_NETMOUNT, FELHOM_WG, FELHOM_SELFUPDATE, FELHOM_SSHD, FELHOM_OOB, FELHOM_PBSDR, FELHOM_BACKUPTARGET, FELHOM_SELFHEAL, FELHOM_ESCROW, FELHOM_GUESTNET, FELHOM_SCRATCH_TEARDOWN, FELHOM_OSAPPLY, FELHOM_POWEROFF
+felhom-agent ALL=(root) NOPASSWD: FELHOM_MOUNT, FELHOM_DISK, FELHOM_PROVISION, FELHOM_FORMAT, FELHOM_DNSMASQ, FELHOM_GUESTHOOK, FELHOM_INTERMEDIARY, FELHOM_CONTROLLERSWAP, FELHOM_STALELOCK, FELHOM_FSTRIM, FELHOM_NETMOUNT, FELHOM_WG, FELHOM_SELFUPDATE, FELHOM_SSHD, FELHOM_OOB, FELHOM_PBSDR, FELHOM_BACKUPTARGET, FELHOM_SELFHEAL, FELHOM_ESCROW, FELHOM_GUESTNET, FELHOM_SCRATCH_TEARDOWN, FELHOM_OSAPPLY, FELHOM_POWEROFF, FELHOM_VZPARTIAL, FELHOM_INTERNALPLACE, FELHOM_THINCHECK

configs/felhom-internal-place

@@ -0,0 +1,118 @@
+#!/bin/bash
+# felhom-internal-place — make the one-disk box's internal data place („Belső tárhely"), felhom.eu
+# architecture/16-storage-physical-identity.md §5, §7 (sizes ruled by the operator 2026-10-16). Installed as
+# /usr/local/sbin/felhom-internal-place (config bundle). Called by felhom-host-install.sh (a new box) and by the agent's
+# signed `add_internal_place` job (an old box) — ONE implementation of the size rule and of the only mkfs it does.
+#
+#   felhom-internal-place plan   <planned GiB>   → prints the numbers, changes nothing
+#   felhom-internal-place create <planned GiB>   → makes the place (or finds it), prints the result
+#
+# <planned GiB> is thin space still to be promised to the guest (rootfs + data volume) that does not exist yet:
+# the installer passes 32 + the data volume; an old box passes 0 (its guest's volumes are already counted).
+#
+# The rule (16 §7, ruled): reserve = max(16 GiB, 10 % of the pool), at most 48 GiB; internal = pool − reserve −
+# every thin volume already promised in the pool − <planned>, made only when ≥ 20 GiB. So nothing in the pool is ever
+# promised beyond pool − reserve: a pool at 100 % stops every volume on the box.
+#
+# It trusts nothing it is told. Every rule is checked here, as root, and any doubt REFUSES (exit 1):
+#   - the pool is the `lvmthin: local-lvm` section of /etc/pve/storage.cfg (its vgname/thinpool), nothing else;
+#   - the only volume it creates, formats or reads is <vg>/felhom-belso;
+#   - it formats ONLY a felhom-belso that carries no signature at all (blkid -p finds nothing) — the one it just made,
+#     or one a crash left between lvcreate and mkfs; any other content REFUSES; it never resizes or removes anything.
+# Output, one line: „created uuid=<fs-uuid> gib=<n>", „formatted uuid=… gib=…" (a crash left it empty), „existing uuid=<fs-uuid> gib=<n>", „skipped <why>" (exit 0),
+# or „plan pool=… reserve=… promised=… planned=… internal=…". Pinned by configs/test_felhom_internal_place.py.
+set -u
+STORAGE_CFG=${FELHOM_STORAGE_CFG:-/etc/pve/storage.cfg}
+LV_NAME=felhom-belso
+LABEL=felhom-belso
+MIN_GIB=20
+
+die() { echo "felhom-internal-place: REFUSED: $*" >&2; exit 1; }
+
+pool_of_local_lvm() { # prints "<vg> <thinpool>" of the lvmthin storage local-lvm, or nothing
+	awk '
+		/^[a-z]+:[ \t]/ { split($0, h, /:[ \t]+/); type = h[1]; id = h[2]; sub(/[ \t]+$/, "", id); in_s = (type == "lvmthin" && id == "local-lvm"); next }
+		in_s && /^[ \t]+vgname[ \t]/ { vg = $2 }
+		in_s && /^[ \t]+thinpool[ \t]/ { tp = $2 }
+		END { if (vg != "" && tp != "") print vg, tp }
+	' "$STORAGE_CFG"
+}
+
+gib_floor() { echo $(( $1 / 1073741824 )); }
+
+[[ $# -eq 2 ]] || die "usage: felhom-internal-place plan|create <planned GiB>"
+MODE=$1; PLANNED=$2
+[[ "$MODE" == plan || "$MODE" == create ]] || die "mode must be plan or create"
+[[ "$PLANNED" =~ ^(0|[1-9][0-9]{0,5})$ ]] || die "planned GiB must be a whole number"
+
+read -r VG POOL < <(pool_of_local_lvm) || true
+[[ -n "${VG:-}" && -n "${POOL:-}" ]] || die "no lvmthin storage local-lvm in $STORAGE_CFG"
+[[ "$VG" =~ ^[A-Za-z0-9_.+-]+$ && "$POOL" =~ ^[A-Za-z0-9_.+-]+$ ]] || die "unexpected vg/pool name"
+
+pool_b=$(lvs --noheadings --nosuffix --units b -o lv_size "$VG/$POOL" 2>/dev/null | tr -d ' ')
+[[ "$pool_b" =~ ^[0-9]+$ ]] || die "cannot read the size of $VG/$POOL"
+pool_gib=$(gib_floor "$pool_b")
+
+# Every thin volume promised in the pool (its virtual size) — except the place itself and vzdump's short-lived
+# snapshots (snap_*_vzdump), which exist only during a whole-box copy and are what the reserve is for.
+# FAIL CLOSED: a failed or empty listing must never read as „nothing promised" (that would size the place too big).
+listing=$(lvs --noheadings --nosuffix --units b -o lv_name,pool_lv,lv_size "$VG" 2>/dev/null) || die "cannot list the volumes of $VG"
+pool_seen=false; promised_b=0; have_lv=false
+while read -r name pl size; do
+	[[ -n "${name:-}" ]] || continue
+	[[ "$name" == "$POOL" ]] && pool_seen=true
+	[[ "$pl" == "$POOL" ]] || continue
+	if [[ "$name" == "$LV_NAME" ]]; then have_lv=true; continue; fi
+	[[ "$name" =~ ^snap_.*_vzdump$ ]] && continue
+	[[ "$size" =~ ^[0-9]+$ ]] || die "cannot read the size of $VG/$name"
+	promised_b=$(( promised_b + size ))
+done <<< "$listing"
+$pool_seen || die "the listing of $VG does not contain the pool $POOL — refusing to guess what is promised"
+promised_gib=$(( (promised_b + 1073741823) / 1073741824 ))   # rounded UP: never promise a partial GiB twice
+
+reserve=$(( pool_gib / 10 )); (( reserve < 16 )) && reserve=16; (( reserve > 48 )) && reserve=48
+internal=$(( pool_gib - reserve - promised_gib - PLANNED ))
+
+if [[ "$MODE" == plan ]]; then
+	echo "plan pool=$pool_gib reserve=$reserve promised=$promised_gib planned=$PLANNED internal=$internal exists=$have_lv"
+	exit 0
+fi
+
+DEV=/dev/$VG/$LV_NAME
+is_block() { [[ "$(stat -L -c %F "$1" 2>/dev/null)" == "block special file" ]]; }
+fs_uuid() { blkid -p -s UUID -o value "$DEV" 2>/dev/null; }
+fs_type() { blkid -p -s TYPE -o value "$DEV" 2>/dev/null; }
+fs_label() { blkid -p -s LABEL -o value "$DEV" 2>/dev/null; }
+# has_signature: FAIL CLOSED. blkid -p exits 0 when it found something and 2 when it found NOTHING; any other result (an
+# error, an ambivalent probe, blkid missing) must never read as „empty" — that is the only gate in front of mkfs.
+has_signature() {
+	local rc=0
+	blkid -p -o value "$DEV" >/dev/null 2>&1 || rc=$?
+	case $rc in
+		0) return 0 ;;
+		2) return 1 ;;
+		*) die "cannot probe $DEV for a signature (blkid exit $rc) — refusing" ;;
+	esac
+}
+lv_gib() { gib_floor "$(lvs --noheadings --nosuffix --units b -o lv_size "$VG/$LV_NAME" 2>/dev/null | tr -d ' ')"; }
+
+if $have_lv; then
+	is_block "$DEV" || die "$DEV is not an active block device"
+	t=$(fs_type)
+	if [[ -n "$t" ]]; then
+		[[ "$t" == ext4 && "$(fs_label)" == "$LABEL" ]] || die "$DEV holds something that is not Felhom's ($t) — left alone"
+		echo "existing uuid=$(fs_uuid) gib=$(lv_gib)"; exit 0
+	fi
+	# A felhom-belso with NO signature: a crash between lvcreate and mkfs. Nothing can be on it — format it.
+	has_signature && die "$DEV carries a signature blkid cannot name — left alone"
+else
+	(( internal >= MIN_GIB )) || { echo "skipped only $internal GiB free for it (pool $pool_gib, reserve $reserve, promised $promised_gib, planned $PLANNED; needs $MIN_GIB)"; exit 0; }
+	lvcreate -q -y -V "${internal}G" -T "$VG/$POOL" -n "$LV_NAME" >/dev/null || die "lvcreate failed"
+	is_block "$DEV" || die "$DEV did not appear after lvcreate"
+	has_signature && die "the new $DEV already carries a signature — left alone"
+	made=created
+fi
+mkfs.ext4 -q -L "$LABEL" "$DEV" >/dev/null || die "mkfs.ext4 failed on $DEV"
+u=$(fs_uuid)
+[[ "$u" =~ ^[0-9a-fA-F-]{36}$ ]] || die "no filesystem UUID after mkfs on $DEV"
+echo "${made:-formatted} uuid=$u gib=$(lv_gib)"

configs/felhom-mkfs-guarded.sh

@@ -58,14 +58,26 @@ fi
 
 # 3) mounted OUTSIDE Felhom's own drive area = a live foreign filesystem → catastrophic. Mounts under
 #    /mnt/felhom-drives are our own drives (the agent detaches before a re-init) → allowed.
-while read -r mp; do
-  [[ -z "$mp" ]] && continue
-  case "$mp" in
-    /mnt/felhom-drives|/mnt/felhom-drives/*) : ;;
-    *) die "device (or a partition) is mounted at $mp ($dev)" ;;
-  esac
-done < <(lsblk -nro MOUNTPOINT "$whole" 2>/dev/null || true)
-
+#    Agent v0.158.0: read from the kernel's mount table for EVERY node of the disk (lsblk's MOUNTPOINT shows one mount
+#    per node; a bind elsewhere hid behind it), and a disk whose nodes cannot be listed REFUSES (it passed before).
+#    Deliberately NO exemption for a sibling that is Felhom's own enrolled drive (raw /mnt/<name> + /mnt/felhom-drives):
+#    the only corroboration available here is a mount the agent itself can make, so trusting it would let a compromised
+#    agent unlock a host owner's disk. Consequence (felhom.eu `16` §13): the free part next to an in-use Felhom drive
+#    cannot be formatted; the dashboard does not offer it.
+PROC_MOUNTS=/proc/mounts
+node_mounts() { awk -v d="$1" '$1 == d { print $2 }' "$PROC_MOUNTS"; }
+disk_nodes="$(lsblk -nrpo NAME "$whole" 2>/dev/null || true)"
+[[ -n "$disk_nodes" ]] || die "cannot list the nodes of $whole — refusing"
+while read -r node; do
+  [[ -z "$node" ]] && continue
+  while read -r mp; do
+    [[ -z "$mp" ]] && continue
+    case "$mp" in
+      /mnt/felhom-drives|/mnt/felhom-drives/*) : ;;
+      *) die "device (or a partition) is mounted at $mp ($node)" ;;
+    esac
+  done < <(node_mounts "$node")
+done <<< "$disk_nodes"
 # 4) read-only device (audit D1): a device the kernel marks RO is never a formattable data disk.
 wbase="${whole#/dev/}"
 rof="/sys/block/$wbase/ro"

configs/felhom-os-apply

@@ -189,6 +189,12 @@ BUNDLE_FILES = [
     ("/usr/local/sbin/felhom-selfupdate-guarded", "felhom-selfupdate-guarded", 0o755, "sh", "replace"),
     ("/usr/local/sbin/felhom-pbs-apply", "felhom-pbs-apply", 0o755, "bash", "replace"),
     ("/usr/local/sbin/felhom-backup-target-apply", "felhom-backup-target-apply", 0o755, "bash", "replace"),
+    # R-1006 (agent v0.156.0): the only delete of backup files — a dead half-written vzdump, re-checked as root.
+    ("/usr/local/sbin/felhom-vzdump-partial-clean", "felhom-vzdump-partial-clean", 0o755, "bash", "replace"),
+    # R-1012 (agent v0.159.0): the read-only orphan-thin check around a whole-guest backup (names an orphan, never removes).
+    ("/usr/local/sbin/felhom-thin-check", "felhom-thin-check", 0o755, "bash", "replace"),
+    # felhom.eu architecture/16 §5 (agent v0.158.0): the internal data place — the only lvcreate/mkfs of a system-disk volume.
+    ("/usr/local/sbin/felhom-internal-place", "felhom-internal-place", 0o755, "bash", "replace"),
     ("/usr/local/sbin/felhom-os-apply", "felhom-os-apply", 0o755, "python", "replace"),
     ("/usr/local/sbin/felhom-crash-guard", "felhom-crash-guard", 0o755, "python", "replace"),
     # R-861 (agent v0.146.0): the content checker for every agent-staged file a root program reads, and the two FIXED

configs/felhom-priv-apply

@@ -13,6 +13,8 @@ longer name the destination: each verb has a fixed source and a fixed (or strict
 
 Verbs (each one sudoers line, exact-match pattern):
   unit <name>              /var/lib/felhom-agent/units/<name> -> /etc/systemd/system/<name>   (.mount | .automount)
+  unit-forget <name>       REMOVES /etc/systemd/system/<name> (R-1124): only a felhom by-UUID drive .mount unit that is
+                           disabled, not mounted, and whose drive is absent (its /dev/disk/by-uuid link is gone)
   dnsmasq <tmp> <name>     /tmp/felhom-resolver-<digits>.conf -> /etc/dnsmasq.d/felhom-<...>.conf
   wg                       /var/lib/felhom-agent/wg/wg-felhom.conf -> /etc/wireguard/wg-felhom.conf (0600)
   sshd-config              /var/lib/felhom-agent/felhom-sshd/sshd_config -> /etc/felhom-sshd/sshd_config
@@ -41,7 +43,7 @@ WG_SRC, WG_DEST = STATE + "/wg/wg-felhom.conf", "/etc/wireguard/wg-felhom.conf"
 SSHD_SRC, SSHD_DEST = STATE + "/felhom-sshd/sshd_config", "/etc/felhom-sshd/sshd_config"
 KEY_SRC, KEY_DEST = STATE + "/felhom-sshd/authorized_keys.felhom-op", "/etc/felhom-sshd/authorized_keys/felhom-op"
 MAX_BYTES = 64 * 1024
-VERBS = ("unit", "dnsmasq", "wg", "sshd-config", "sshd-key", "controller-image")
+VERBS = ("unit", "unit-forget", "dnsmasq", "wg", "sshd-config", "sshd-key", "controller-image")
 # R-861 (a) A1 (`09` §3 decision 165): the SAME pattern as the agent's controllerImageRe (internal/localapi/
 # controllerswap.go) — a compromised agent cannot hand the guest's bootstrap any other image. Pinned by
 # configs/test_felhom_priv_apply.py ControllerImage.
@@ -51,6 +53,11 @@ CONTROLLER_IMAGE_MAX = 256
 VMID_RE = re.compile(r"^[0-9]{1,9}$")
 
 UNIT_NAME_RE = re.compile(r"^mnt-[A-Za-z0-9_.\\-]+\.(mount|automount)$")
+DRIVE_UNIT_NAME_RE = re.compile(r"^mnt-[A-Za-z0-9_.\\-]+\.mount$")
+DRIVE_UNIT_MARKER = "Managed by felhom-agent"
+NET_UNIT_MARKER = "felhom network storage"
+WANTS_DIR = UNIT_DIR + "/multi-user.target.wants"
+BY_UUID_DIR = "/dev/disk/by-uuid"
 DNSMASQ_TMP_RE = re.compile(r"^/tmp/felhom-resolver-[0-9]+\.conf$")
 DNSMASQ_NAME_RE = re.compile(r"^felhom-[a-z0-9][a-z0-9._-]*\.conf$")
 SEG = r"[A-Za-z0-9_-][A-Za-z0-9_.-]*"
@@ -135,6 +142,36 @@ class Host:
     def read_stdin(self, limit):
         return sys.stdin.buffer.read(limit + 1)
 
+    # ---- unit-forget (R-1124) ----
+    def read_installed(self, path):
+        """The installed unit: None when absent; Refused when it is not a regular file (a symlink is never followed)."""
+        try:
+            fd = os.open(path, os.O_RDONLY | os.O_NOFOLLOW | os.O_CLOEXEC)
+        except FileNotFoundError:
+            return None
+        except OSError as e:
+            raise Refused("F1", f"cannot open {path}: {e.strerror}")
+        with os.fdopen(fd, "rb") as f:
+            if not stat.S_ISREG(os.fstat(f.fileno()).st_mode):
+                raise Refused("F1", f"{path} is not a regular file")
+            return f.read(MAX_BYTES + 1)
+
+    def lexists(self, path):
+        return os.path.lexists(path)
+
+    def mountpoints(self):
+        """Every mount point in /proc/self/mountinfo (field 5, octal escapes decoded)."""
+        out = set()
+        with open("/proc/self/mountinfo", "r", encoding="utf-8", errors="replace") as f:
+            for line in f:
+                parts = line.split(" ")
+                if len(parts) > 4:
+                    out.add(re.sub(r"\\([0-7]{3})", lambda m: chr(int(m.group(1), 8)), parts[4]))
+        return out
+
+    def remove(self, path):
+        os.remove(path)
+
     def write_guest_image(self, vmid, data):
         """As root: `pct exec <vmid> -- tee <the fixed file>` with the checked ref on stdin (no shell)."""
         r = subprocess.run(["/usr/sbin/pct", "exec", str(vmid), "--", "tee", CONTROLLER_IMAGE_FILE],
@@ -424,6 +461,45 @@ def controller_image(rest, host):
     return 0
 
 
+def unit_forget(rest, host):
+    """R-1124: remove ONE felhom drive mount unit that the agent has already disabled, whose drive is gone. A drive that
+    is still present or still mounted keeps its unit — this verb can only forget, never unmount or take a drive away."""
+    try:
+        if len(rest) != 1 or not DRIVE_UNIT_NAME_RE.match(rest[0]) or "/" in rest[0]:
+            raise Refused("A1", "usage: felhom-priv-apply unit-forget mnt-<escaped path>.mount")
+        name = rest[0]
+        path = os.path.join(UNIT_DIR, name)
+        data = host.read_installed(path)
+        if data is None:
+            host.log(f"felhom-priv-apply: ABSENT unit-forget {path} (nothing to remove)")
+            return 0
+        text = text_of(data, path)
+        if DRIVE_UNIT_MARKER not in text or NET_UNIT_MARKER in text:
+            raise Refused("F2", f"{name} is not a felhom drive unit")
+        check_unit(name, text)
+        s = parse_ini(text, name)
+        what, where = s["Mount"].get("What", ""), s["Mount"]["Where"]
+        mu = re.match(r"^/dev/disk/by-uuid/(.+)$", what)
+        if not mu:
+            raise Refused("F2", f"{name} is not a by-UUID drive unit")
+        if host.lexists(os.path.join(WANTS_DIR, name)):
+            raise Refused("F3", f"{name} is still enabled (disable it first)")
+        if where in host.mountpoints():
+            raise Refused("F4", f"{where} is mounted")
+        if host.lexists(os.path.join(BY_UUID_DIR, mu.group(1))):
+            raise Refused("F5", f"the drive of {name} is present (filesystem UUID {mu.group(1)})")
+    except Refused as e:
+        host.log(f"felhom-priv-apply: REFUSED [{e.rule}] unit-forget {' '.join(rest)[:120]}: {e.reason}")
+        return 2 if e.rule == "A1" else 3
+    try:
+        host.remove(path)
+    except OSError as e:
+        host.log(f"felhom-priv-apply: FAILED unit-forget {path}: {e}")
+        return 4
+    host.log(f"felhom-priv-apply: REMOVED unit-forget {path} (where={where})")
+    return 0
+
+
 def main(argv, host=None):
     host = host or Host()
     if argv == ["--self-check"]:
@@ -445,6 +521,8 @@ def main(argv, host=None):
         return 0
     if argv and argv[0] == "controller-image":
         return controller_image(argv[1:], host)
+    if argv and argv[0] == "unit-forget":
+        return unit_forget(argv[1:], host)
     try:
         verb, src, dest, mode, checker = plan(argv)
         data = host.read_source(src)

configs/felhom-thin-check

@@ -0,0 +1,74 @@
+#!/bin/bash
+# felhom-thin-check — list the thin device ids a thin pool holds next to the ones LVM knows (R-1012). READ-ONLY: it
+# names an orphan, it never removes one. Installed as /usr/local/sbin/felhom-thin-check (config bundle).
+#
+#   felhom-thin-check <vg> <pool>
+#
+# A full host root while vzdump takes its snapshot can leave a thin snapshot inside the pool that LVM no longer knows
+# (demo-hp 2026-10-14: `thin_ls` device 4, LVM only 1 and 2); from then on every snapshot backup of that guest fails
+# („Creation of new snapshot 4 of device 2 failed") and nothing said why. The repair (`dmsetup message … delete <id>`)
+# stays a HUMAN step, decided by the operator — this helper only reads.
+#
+# Reading the live pool's device list needs a metadata snapshot: the kernel's documented way to read metadata while
+# the pool is in use (`reserve_metadata_snap`, `thin_ls --metadata-snap`, `release_metadata_snap`). It changes no data
+# and no mapping; it is held for the length of one `thin_ls -o DEV` (well under a second on these pools). Rules:
+#   1. the arguments are an LVM name pair; <vg>/<pool> is a thin pool (lv_attr starts with `t`);
+#   2. only one run at a time (a lock file); a metadata snapshot that someone ELSE holds is never touched — the run
+#      refuses (exit 3) and does not release it;
+#   3. LVM's list is read before AND after the pool's, so a volume created in between is not reported as an orphan
+#      (the caller treats an id as an orphan only when it is in the pool and in neither LVM list).
+# Output, one record per line:  lvm-before <id> <lv> | pool <id> | lvm-after <id> <lv> | ok
+# Exit: 0 read, 1 refused (arguments / not a thin pool), 3 a metadata snapshot is already held or cannot be reserved,
+# 4 the pool could not be read. Pinned by configs/test_felhom_thin_check.py.
+set -u
+LVS=/usr/sbin/lvs
+DMSETUP=/usr/sbin/dmsetup
+THIN_LS=/usr/sbin/thin_ls
+LOCK=/run/lock/felhom-thin-check.lock
+NAME_RE='^[A-Za-z0-9_+][A-Za-z0-9_.+-]{0,63}$'
+
+die() { echo "felhom-thin-check: REFUSED: $1" >&2; exit "${2:-1}"; }  # $1 the reason, $2 the exit code
+
+lvm_ids() { # $1 tag → "<tag> <id> <lv>" for every thin volume of this pool
+	"$LVS" --noheadings --separator , -o lv_name,thin_id --select "pool_lv=$POOL" -- "$VG" 2>/dev/null |
+		awk -F, -v tag="$1" '{ gsub(/ /, ""); if ($2 ~ /^[0-9]+$/) print tag, $2, $1 }'
+}
+
+release() { "$DMSETUP" message "$TPOOL" 0 release_metadata_snap >/dev/null 2>&1; }
+
+main() {
+	[[ $# -eq 2 ]] || die "usage: felhom-thin-check <vg> <pool>"
+	VG=$1 POOL=$2
+	local attr held ids
+	[[ "$VG" =~ $NAME_RE && "$POOL" =~ $NAME_RE ]] || die "not an LVM name pair"
+
+	attr=$("$LVS" --noheadings -o lv_attr -- "$VG/$POOL" 2>/dev/null | tr -d ' ')
+	[[ "$attr" == t* ]] || die "$VG/$POOL is not a thin pool"
+
+	DM="${VG//-/--}-${POOL//-/--}"
+	TPOOL="$DM-tpool"
+	TMETA="/dev/mapper/${DM}_tmeta"
+
+	exec 9>"$LOCK" || die "cannot open the lock file" 4
+	flock -w 30 9 || die "another check holds the lock" 3
+
+	# Field 7 of the thin-pool status is the held metadata root: "-" when no metadata snapshot exists.
+	held=$("$DMSETUP" status "$TPOOL" 2>/dev/null | awk '$3 == "thin-pool" { print $7 }')
+	[[ -n "$held" ]] || die "cannot read the status of $TPOOL" 4
+	[[ "$held" == "-" ]] || die "a metadata snapshot of $TPOOL is already held (by someone else) — not touched" 3
+
+	lvm_ids lvm-before
+	"$DMSETUP" message "$TPOOL" 0 reserve_metadata_snap >/dev/null 2>&1 || die "could not reserve a metadata snapshot of $TPOOL" 3
+	trap release EXIT
+	trap 'release; trap - EXIT; exit 4' INT TERM HUP
+	if ! ids=$("$THIN_LS" --metadata-snap --no-headers -o DEV "$TMETA" 2>/dev/null); then
+		die "thin_ls could not read $TMETA" 4
+	fi
+	release
+	trap - EXIT INT TERM HUP
+	awk '$1 ~ /^[0-9]+$/ { print "pool", $1 }' <<<"$ids"
+	lvm_ids lvm-after
+	echo ok
+}
+
+if [[ "${BASH_SOURCE[0]}" == "$0" ]]; then main "$@"; fi

configs/felhom-vzdump-partial-clean

@@ -0,0 +1,99 @@
+#!/bin/bash
+# felhom-vzdump-partial-clean — remove ONE dead, half-written vzdump backup (R-1006). The only delete the agent's
+# sudoers allows for backup files. Installed as /usr/local/sbin/felhom-vzdump-partial-clean (config bundle).
+#
+#   felhom-vzdump-partial-clean <storage> <vmid> <name>
+#
+# A power cut during the night backup leaves `vzdump-lxc-<vmid>-<date>.tar.dat` (+ its `.tmp/` directory) on the
+# storage; nothing removes it, every later backup is then skipped „not enough space", and a full host root breaks
+# LVM's own metadata writes (power-cut-fixes-2026-10-14 C4). This helper removes such a leftover — and NOTHING else.
+# It trusts nothing the agent says; every rule is checked here, as root, and any doubt REFUSES (exit 1):
+#   1. the arguments: a storage id, a vmid, and a name that is exactly
+#      vzdump-(lxc|qemu)-<THAT vmid>-YYYY_MM_DD-HH_MM_SS.(tar.dat|vma.dat|tmp) — never a finished archive
+#      (.tar.zst, .vma.zst, …), never a .log/.notes, never another vmid;
+#   2. the storage: its section in /etc/pve/storage.cfg is `dir: <storage>` with a `path`; the target is
+#      <path>/dump/<name>, not a symlink, and its real directory IS <path>/dump;
+#   3. not in use: no vzdump runs on the host at all (a task `…:vzdump:…` or a `vzdump` command — a job for several
+#      guests carries no vmid in its task name), and no
+#      process holds the target (or, for the .tmp directory, anything under it) open or as its working directory
+#      (C2: after a killed worker, tar and zstd kept writing into the .tar.dat for 4 min 40 s, none named „vzdump");
+#   4. dead: last modified before this boot (`btime` in /proc/stat), or untouched for at least QUIET_S seconds.
+# Prints one line on success: „removed <name> <bytes> B (<why>)". Pinned by configs/test_felhom_vzdump_partial_clean.py.
+set -u
+STORAGE_CFG=/etc/pve/storage.cfg
+PROC=/proc
+QUIET_S=120
+
+die() { echo "felhom-vzdump-partial-clean: REFUSED: $*" >&2; exit 1; }
+
+storage_path() { # $1 storage id → prints the dir storage's path, or nothing
+	awk -v want="$1" '
+		/^[a-z]+:[ \t]/ { split($0, h, /:[ \t]+/); type = h[1]; id = h[2]; sub(/[ \t]+$/, "", id); in_s = (id == want); next }
+		in_s && /^[ \t]+path[ \t]/ { if (type == "dir") { sub(/^[ \t]+path[ \t]+/, ""); sub(/[ \t]+$/, ""); print; exit } }
+	' "$STORAGE_CFG"
+}
+
+storage_type() {
+	awk -v want="$1" '/^[a-z]+:[ \t]/ { split($0, h, /:[ \t]+/); id = h[2]; sub(/[ \t]+$/, "", id); if (id == want) { print h[1]; exit } }' "$STORAGE_CFG"
+}
+
+boot_time() { awk '$1 == "btime" { print $2 }' "$PROC/stat"; }
+
+newest_mtime() { # a file's mtime, or the newest mtime of anything in a directory (the dir itself included)
+	if [[ -d "$1" ]]; then find "$1" -xdev -printf '%T@\n' 2>/dev/null | sort -n | tail -1 | cut -d. -f1
+	else stat -c %Y "$1"; fi
+}
+
+busy_reason() { # $1 vmid, $2 target → prints why it is in use, or nothing
+	local vmid=$1 target=$2 p link cmd
+	for p in "$PROC"/[0-9]*; do
+		[[ -r "$p/cmdline" ]] || continue
+		cmd=$(tr '\0' ' ' < "$p/cmdline" 2>/dev/null)
+		# ANY vzdump on the host refuses: a job for several guests runs as `task UPID:…:vzdump::…` (no vmid in it), and
+		# its tar may be between this guest's files. Waiting costs one backup run; deleting a live one costs the backup.
+		if [[ "$cmd" == *":vzdump:"* ]] || [[ "$cmd" =~ (^|/)vzdump([[:space:]]|$) ]]; then
+			echo "a vzdump runs on this host (pid ${p##*/})"; return
+		fi
+		for link in "$p/cwd" "$p"/fd/*; do
+			local to
+			to=$(readlink "$link" 2>/dev/null) || continue
+			if [[ "$to" == "$target" || "$to" == "$target/"* ]]; then
+				echo "pid ${p##*/} has it open"; return
+			fi
+		done
+	done
+}
+
+main() {
+	local storage=${1:-} vmid=${2:-} name=${3:-}
+	[[ $# -eq 3 ]] || die "usage: felhom-vzdump-partial-clean <storage> <vmid> <name>"
+	[[ "$storage" =~ ^[A-Za-z][A-Za-z0-9_-]{0,63}$ ]] || die "bad storage id"
+	[[ "$vmid" =~ ^[1-9][0-9]{2,8}$ ]] || die "bad vmid"
+	[[ "$name" =~ ^vzdump-(lxc|qemu)-${vmid}-[0-9]{4}_[0-9]{2}_[0-9]{2}-[0-9]{2}_[0-9]{2}_[0-9]{2}\.(tar\.dat|vma\.dat|tmp)$ ]] \
+		|| die "name is not a half-written vzdump of guest $vmid: $name"
+	[[ "$(storage_type "$storage")" == "dir" ]] || die "storage $storage is not a dir storage"
+	local path; path=$(storage_path "$storage")
+	[[ "$path" == /* ]] || die "storage $storage has no path"
+	local dump="$path/dump" target="$path/dump/$name"
+	[[ -e "$target" ]] || die "no such entry: $target"
+	[[ ! -L "$target" ]] || die "a symlink: $target"
+	[[ "$(realpath -e "$dump")" == "$(realpath -e "$(dirname "$target")")" ]] || die "not directly in $dump"
+	if [[ "$name" == *.tmp ]]; then [[ -d "$target" ]] || die ".tmp is not a directory"
+	else [[ -f "$target" ]] || die "not a regular file"; fi
+	local busy; busy=$(busy_reason "$vmid" "$target")
+	[[ -z "$busy" ]] || die "in use: $busy"
+	local bt mt now why
+	bt=$(boot_time); mt=$(newest_mtime "$target"); now=$(date +%s)
+	[[ "$bt" =~ ^[0-9]+$ && "$mt" =~ ^[0-9]+$ ]] || die "cannot read the boot time or the modification time"
+	if (( mt < bt )); then why="written before this boot"
+	elif (( now - mt >= QUIET_S )); then why="untouched for $(( now - mt )) s, no vzdump for $vmid, nothing holds it"
+	else die "modified $(( now - mt )) s ago — may still be written"; fi
+	local size; size=$(du -sb "$target" 2>/dev/null | cut -f1)
+	if [[ -d "$target" ]]; then rm -rf --one-file-system -- "$target"; else rm -f -- "$target"; fi
+	[[ ! -e "$target" ]] || die "remove failed: $target"
+	echo "removed $name ${size:-?} B ($why)"
+}
+
+if [[ "${BASH_SOURCE[0]}" == "$0" ]]; then
+	main "$@"
+fi

configs/test_felhom_internal_place.py

@@ -0,0 +1,255 @@
+#!/usr/bin/env python3
+"""Tests for configs/felhom-internal-place (felhom.eu architecture/16 §5, §7 — sizes ruled 2026-10-16).
+
+The helper is run for real under bash; `lvs`, `lvcreate`, `blkid`, `mkfs.ext4` and `stat` are PATH stubs that read and
+write one JSON state file, so nothing on this machine is touched. Every refusal has a positive control: the same setup
+without the one condition goes through — so a test cannot pass because the helper refuses everything.
+Run: python3 configs/test_felhom_internal_place.py   (also run by the Go suite, internal/internalplace)
+"""
+import json, os, pathlib, stat, subprocess, tempfile, unittest
+
+HELPER = pathlib.Path(__file__).resolve().parent / "felhom-internal-place"
+GIB = 1 << 30
+
+STUB_LVS = r'''#!/usr/bin/env python3
+import json, os, sys
+st = json.load(open(os.environ["IP_STATE"]))
+if st.get("lvs_fail_listing") and "lv_name,pool_lv,lv_size" in sys.argv: sys.exit(5)
+args = [a for a in sys.argv[1:] if not a.startswith("-") and a not in ("b",)]
+fields = sys.argv[sys.argv.index("-o") + 1].split(",")
+target = args[-1]
+listing = target == st["vg"]
+for lv in st["lvs"]:
+    full = st["vg"] + "/" + lv["name"]
+    if listing and st.get("hide_pool_in_listing") and lv.get("pool", "") == "":
+        continue   # a listing that silently lost the pool row (the pool's own size stays readable)
+    if listing or target == full:
+        row = {"lv_name": lv["name"], "pool_lv": lv.get("pool", ""), "lv_size": str(lv["size"])}
+        print("  " + " ".join(row[f] for f in fields))
+'''
+STUB_LVCREATE = r'''#!/usr/bin/env python3
+import json, os, sys
+p = os.environ["IP_STATE"]; st = json.load(open(p))
+a = sys.argv[1:]
+size = int(a[a.index("-V") + 1].rstrip("G")) * (1 << 30)
+vg, pool = a[a.index("-T") + 1].split("/")
+st["lvs"].append({"name": a[a.index("-n") + 1], "pool": pool, "size": size, "active": True})
+st.setdefault("calls", []).append("lvcreate " + " ".join(a))
+if st.get("dirty_new"): st["fs"]["felhom-belso"] = {"TYPE": "", "RAW": "junk"}
+json.dump(st, open(p, "w"))
+'''
+STUB_BLKID = r'''#!/usr/bin/env python3
+# like blkid -p: exit 0 when something is found, 2 when nothing is; st["blkid_rc"] forces an error exit
+import json, os, sys
+st = json.load(open(os.environ["IP_STATE"]))
+if st.get("blkid_rc"): sys.exit(st["blkid_rc"])
+fs = st["fs"].get(os.path.basename(sys.argv[-1]), {})
+if "-s" in sys.argv:
+    v = fs.get(sys.argv[sys.argv.index("-s") + 1], "")
+    if v: print(v)
+    sys.exit(0 if v else 2)
+vals = [fs.get(k, "") for k in ("TYPE", "LABEL", "UUID", "RAW") if fs.get(k)]
+if vals: print("\n".join(vals))
+sys.exit(0 if vals else 2)
+'''
+STUB_MKFS = r'''#!/usr/bin/env python3
+import json, os, sys
+p = os.environ["IP_STATE"]; st = json.load(open(p))
+dev = os.path.basename(sys.argv[-1])
+st["fs"][dev] = {"TYPE": "ext4", "LABEL": sys.argv[sys.argv.index("-L") + 1], "UUID": "0b6e4c5a-1d2e-4f3a-9b8c-7d6e5f4a3b2c"}
+st.setdefault("calls", []).append("mkfs.ext4 " + " ".join(sys.argv[1:]))
+json.dump(st, open(p, "w"))
+'''
+STUB_STAT = r'''#!/usr/bin/env python3
+import json, os, sys
+st = json.load(open(os.environ["IP_STATE"]))
+name = os.path.basename(sys.argv[-1])
+if any(lv["name"] == name and lv.get("active", True) for lv in st["lvs"]): print("block special file")
+else: sys.exit(1)
+'''
+
+CFG = "dir: local\n\tpath /var/lib/vz\n\tcontent iso,backup\n\nlvmthin: local-lvm\n\tthinpool data\n\tvgname pve\n\tcontent rootdir,images\n"
+
+
+class Box:
+    def __init__(self, pool_gib, lvs=(), fs=None, cfg=CFG, **extra):
+        self.t = tempfile.TemporaryDirectory()
+        r = pathlib.Path(self.t.name)
+        self.bin = r / "bin"; self.bin.mkdir()
+        for name, body in (("lvs", STUB_LVS), ("lvcreate", STUB_LVCREATE), ("blkid", STUB_BLKID),
+                           ("mkfs.ext4", STUB_MKFS), ("stat", STUB_STAT)):
+            f = self.bin / name; f.write_text(body); f.chmod(0o755)
+        self.cfg = r / "storage.cfg"; self.cfg.write_text(cfg)
+        self.state = r / "state.json"
+        st = {"vg": "pve", "lvs": [{"name": "data", "pool": "", "size": pool_gib * GIB}] + [dict(l) for l in lvs],
+              "fs": fs or {}}
+        st.update(extra)
+        self.state.write_text(json.dumps(st))
+
+    def run(self, *args):
+        env = {"PATH": f"{self.bin}:/usr/bin:/bin", "IP_STATE": str(self.state), "FELHOM_STORAGE_CFG": str(self.cfg)}
+        p = subprocess.run(["bash", str(HELPER), *args], env=env, capture_output=True, text=True)
+        return p.returncode, p.stdout.strip(), p.stderr.strip()
+
+    def st(self):
+        return json.loads(self.state.read_text())
+
+    def calls(self):
+        return self.st().get("calls", [])
+
+    def close(self):
+        self.t.cleanup()
+
+
+def thin(name, gib, pool="data"):
+    return {"name": name, "pool": pool, "size": gib * GIB}
+
+
+class Sizes(unittest.TestCase):
+    """The operator's table (SPACE-CHECK §4, ruled 2026-10-16), as the helper computes it on a fresh box."""
+
+    def plan(self, pool, planned, lvs=()):
+        b = Box(pool, lvs)
+        try:
+            rc, out, err = b.run("plan", str(planned))
+            self.assertEqual(rc, 0, err)
+            self.assertEqual(b.calls(), [], "plan must change nothing")
+            return dict(kv.split("=") for kv in out.split()[1:])
+        finally:
+            b.close()
+
+    def test_the_ruled_table(self):
+        # disk → pool (GiB), data volume (planned = 32 + mp0), expected internal
+        for pool, mp0, reserve, internal in ((140, 50, 16, 42), (349, 100, 34, 183), (800, 130, 48, 590), (1720, 130, 48, 1510)):
+            got = self.plan(pool, 32 + mp0)
+            self.assertEqual((int(got["reserve"]), int(got["internal"])), (reserve, internal), (pool, got))
+
+    def test_promised_volumes_count_but_vzdump_snapshots_and_other_pools_do_not(self):
+        got = self.plan(349, 0, [thin("vm-9201-disk-0", 32), thin("vm-9201-disk-1", 250),
+                                 thin("snap_vm-9201-disk-1_vzdump", 250), thin("vm-500-disk-0", 40, pool="other")])
+        self.assertEqual((int(got["promised"]), int(got["internal"])), (282, 33), got)
+
+
+class Create(unittest.TestCase):
+    def test_a_fresh_box_gets_the_place(self):
+        b = Box(349)
+        try:
+            rc, out, err = b.run("create", "132")
+            self.assertEqual(rc, 0, err)
+            self.assertRegex(out, r"^created uuid=[0-9a-f-]{36} gib=183$")
+            self.assertEqual(len([c for c in b.calls() if c.startswith("lvcreate")]), 1)
+            self.assertIn("-V 183G -T pve/data -n felhom-belso", b.calls()[0])
+            self.assertIn("-L felhom-belso", b.calls()[1])
+        finally:
+            b.close()
+
+    def test_too_little_space_makes_nothing(self):
+        b = Box(53)   # demo-hp's 128 GB disk
+        try:
+            rc, out, err = b.run("create", str(32 + 50))
+            self.assertEqual(rc, 0, err)
+            self.assertTrue(out.startswith("skipped "), out)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+
+    def test_an_existing_place_is_found_never_formatted(self):
+        fs = {"felhom-belso": {"TYPE": "ext4", "LABEL": "felhom-belso", "UUID": "11111111-2222-3333-4444-555555555555"}}
+        b = Box(349, [thin("felhom-belso", 183)], fs)
+        try:
+            rc, out, err = b.run("create", "0")
+            self.assertEqual(rc, 0, err)
+            self.assertEqual(out, "existing uuid=11111111-2222-3333-4444-555555555555 gib=183")
+            self.assertEqual(b.calls(), [], "an existing place must never be created or formatted again")
+        finally:
+            b.close()
+
+    def test_a_foreign_filesystem_on_the_name_is_refused(self):
+        # positive control: test_an_existing_place_is_found_never_formatted (same LV, Felhom's label) goes through
+        for fs in ({"TYPE": "xfs", "LABEL": "felhom-belso", "UUID": "x"}, {"TYPE": "ext4", "LABEL": "films", "UUID": "y"}):
+            b = Box(349, [thin("felhom-belso", 183)], {"felhom-belso": fs})
+            try:
+                rc, out, err = b.run("create", "0")
+                self.assertEqual(rc, 1, (fs, out))
+                self.assertIn("REFUSED", err)
+                self.assertEqual(b.calls(), [], fs)
+            finally:
+                b.close()
+
+    def test_a_crash_between_lvcreate_and_mkfs_is_finished(self):
+        b = Box(349, [thin("felhom-belso", 183)], {})
+        try:
+            rc, out, err = b.run("create", "0")
+            self.assertEqual(rc, 0, err)
+            self.assertRegex(out, r"^formatted uuid=")
+            self.assertEqual([c.split()[0] for c in b.calls()], ["mkfs.ext4"])
+        finally:
+            b.close()
+
+    def test_an_unnamed_signature_is_refused(self):
+        b = Box(349, [thin("felhom-belso", 183)], {"felhom-belso": {"RAW": "something"}})
+        try:
+            rc, out, err = b.run("create", "0")
+            self.assertEqual(rc, 1, out)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+
+    def test_a_new_volume_that_is_not_empty_is_not_formatted(self):
+        b = Box(349, dirty_new=True)
+        try:
+            rc, out, err = b.run("create", "132")
+            self.assertEqual(rc, 1, out)
+            self.assertNotIn("mkfs", " ".join(b.calls()))
+        finally:
+            b.close()
+
+    def test_bad_arguments_and_no_local_lvm_are_refused(self):
+        b = Box(349, cfg="dir: local\n\tpath /var/lib/vz\n")
+        try:
+            self.assertEqual(b.run("create", "0")[0], 1)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+        b = Box(349)
+        try:
+            for args in (("create", "-1"), ("create", "1x"), ("wipe", "0"), ("create",)):
+                self.assertEqual(b.run(*args)[0], 1, args)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+
+
+class FailsClosed(unittest.TestCase):
+    """A read that fails must never read as „empty" or „nothing promised". Positive control: Create (same setups)."""
+
+    def test_a_failed_volume_listing_creates_nothing(self):
+        b = Box(349, lvs_fail_listing=True)
+        try:
+            rc, out, err = b.run("create", "132")
+            self.assertEqual(rc, 1, out)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+
+    def test_a_listing_without_the_pool_creates_nothing(self):
+        b = Box(349, [thin("vm-9201-disk-1", 250)], hide_pool_in_listing=True)
+        try:
+            self.assertEqual(b.run("create", "132")[0], 1)
+            self.assertEqual(b.calls(), [])
+        finally:
+            b.close()
+
+    def test_a_blkid_error_never_reads_as_empty(self):
+        for setup in ({"lvs": [thin("felhom-belso", 183)], "fs": {}}, {}):
+            b = Box(349, **({"lvs": setup["lvs"], "fs": setup["fs"]} if setup else {}), blkid_rc=4)
+            try:
+                rc, out, err = b.run("create", "0" if setup else "132")
+                self.assertEqual(rc, 1, (setup, out))
+                self.assertNotIn("mkfs", " ".join(b.calls()), setup)
+            finally:
+                b.close()
+
+
+if __name__ == "__main__":
+    unittest.main()

configs/test_felhom_mkfs_guarded.py

@@ -0,0 +1,145 @@
+#!/usr/bin/env python3
+"""Tests for configs/felhom-mkfs-guarded.sh — the ONLY mkfs path the agent's sudoers permits.
+
+First tests of this wrapper (agent v0.158.0), written for partitions as places (felhom.eu architecture/16 §6). Every
+mount of every node of the disk outside /mnt/felhom-drives refuses — INCLUDING a sibling that is Felhom's own drive
+(its raw /mnt/<name> mount): the agent can make that corroborating mount itself, so the root wrapper must not trust
+it (security review of 912f7c1). An unlistable disk refuses.
+
+HOW IT RUNS ANYWHERE: the wrapper is copied with five literal host paths replaced (each replacement is asserted, so a
+drifted wrapper fails here instead of silently testing nothing): /proc/mounts → a fake mount table, the block-device
+test → true (no real /dev nodes here), the pvs candidates → a stub, /sys/block → a fake tree, the mkfs binaries → a
+recorder. `lsblk` is a PATH stub answering the three calls the wrapper makes from one JSON disk description.
+Every refusal has a positive control (the same disk without the one condition is formatted).
+Run: python3 configs/test_felhom_mkfs_guarded.py   (also run by the Go suite, internal/storage)
+"""
+import json, os, pathlib, subprocess, tempfile, unittest
+
+WRAPPER = pathlib.Path(__file__).resolve().parent / "felhom-mkfs-guarded.sh"
+
+STUB_LSBLK = r'''#!/usr/bin/env python3
+import json, os, sys
+st = json.load(open(os.environ["MG_STATE"]))
+a = sys.argv[1:]; dev = a[-1]
+nodes = st["nodes"]   # [{"name": "/dev/sdb", "pk": "", "fstype": ""}, ...]
+if a[0] == "-ndo" and a[1] == "PKNAME":
+    for n in nodes:
+        if n["name"] == dev and n.get("pk"): print(n["pk"])
+elif a[0] == "-nrpo" and a[1] == "NAME":
+    if not st.get("lsblk_empty"):
+        for n in nodes: print(n["name"])
+elif a[0] == "-nro" and a[1] == "FSTYPE":
+    for n in nodes: print(n.get("fstype", ""))
+elif a[0] == "-nro" and a[1] == "MOUNTPOINT":   # the pre-v0.158.0 wrapper's call: one mountpoint per node, like lsblk
+    mt = [l.split()[:2] for l in open(os.environ["MG_MOUNTS"]) if l.strip()]
+    for n in nodes:
+        mps = [m for d, m in mt if d == n["name"]]
+        print(mps[0] if mps else "")
+'''
+STUB_MKFS = '#!/bin/sh\necho "$0 $*" >> "$MG_CALLS"\n'
+STUB_PVS = '#!/bin/sh\nprintf "%s\\n" $MG_PVS\n'
+
+
+class Disk:
+    def __init__(self, nodes, mounts, ro="0", pvs="", lsblk_empty=False):
+        self.t = tempfile.TemporaryDirectory()
+        r = pathlib.Path(self.t.name)
+        self.bin = r / "bin"; self.bin.mkdir()
+        (self.bin / "lsblk").write_text(STUB_LSBLK); (self.bin / "lsblk").chmod(0o755)
+        for n in ("mkfs.ext4", "mkfs.xfs"):
+            (self.bin / n).write_text(STUB_MKFS); (self.bin / n).chmod(0o755)
+        (self.bin / "pvs").write_text(STUB_PVS); (self.bin / "pvs").chmod(0o755)
+        self.state = r / "state.json"
+        self.state.write_text(json.dumps({"nodes": nodes, "lsblk_empty": lsblk_empty}))
+        self.mounts = r / "mounts"
+        self.mounts.write_text("".join(f"{d} {m} ext4 rw 0 0\n" for d, m in mounts))
+        self.sys = r / "sys"
+        for n in nodes:
+            if not n.get("pk"):
+                d = self.sys / os.path.basename(n["name"]); d.mkdir(parents=True); (d / "ro").write_text(ro + "\n")
+        self.calls = r / "calls"
+        self.pvs = pvs
+        src = WRAPPER.read_text()
+        subs = [
+            ("/proc/mounts", str(self.mounts)),
+            ('[[ -b "$dev" ]]', "true"),
+            ("for c in /usr/sbin/pvs /sbin/pvs; do", f"for c in {self.bin}/pvs; do"),
+            ('rof="/sys/block/', f'rof="{self.sys}/'),
+            ("exec /usr/sbin/mkfs.ext4", f"exec {self.bin}/mkfs.ext4"),
+            ("exec /usr/sbin/mkfs.xfs ", f"exec {self.bin}/mkfs.xfs "),
+        ]
+        for a, b in subs:
+            assert a in src, f"the wrapper no longer contains {a!r} — update this test's substitutions"
+            src = src.replace(a, b)
+        self.copy = r / "mkfs-guarded"; self.copy.write_text(src)
+
+    def run(self, dev, fstype="ext4"):
+        env = {"PATH": f"{self.bin}:/usr/bin:/bin", "MG_STATE": str(self.state), "MG_CALLS": str(self.calls),
+               "MG_MOUNTS": str(self.mounts),
+               "MG_PVS": self.pvs}
+        p = subprocess.run(["bash", str(self.copy), dev, fstype], env=env, capture_output=True, text=True)
+        made = self.calls.read_text().split("\n")[0] if self.calls.exists() else ""
+        return p.returncode, p.stderr.strip(), made
+
+    def close(self):
+        self.t.cleanup()
+
+
+def disk_sdb(sdb1_fstype="ext4", sdb2_fstype=""):
+    return [{"name": "/dev/sdb", "pk": "", "fstype": ""},
+            {"name": "/dev/sdb1", "pk": "sdb", "fstype": sdb1_fstype},
+            {"name": "/dev/sdb2", "pk": "sdb", "fstype": sdb2_fstype}]
+
+
+ROOT = ("/dev/sda2", "/")
+OWN_SDB1 = [("/dev/sdb1", "/mnt/hdd_1"), ("/dev/sdb1", "/mnt/felhom-drives/hdd_1")]
+
+
+class Partitions(unittest.TestCase):
+    def check(self, mounts, dev="/dev/sdb2", nodes=None, want_ok=True, **kw):
+        d = Disk(nodes or disk_sdb(), [ROOT] + mounts, **kw)
+        try:
+            rc, err, made = d.run(dev)
+            if want_ok:
+                self.assertEqual(rc, 0, err)
+                self.assertTrue(made.endswith(f"mkfs.ext4 -F {dev}"), made)
+            else:
+                self.assertEqual(rc, 1, (mounts, made))
+                self.assertIn("REFUSED", err)
+                self.assertEqual(made, "", "a refused format must not run mkfs")
+            return err
+        finally:
+            d.close()
+
+    def test_a_free_part_on_an_unmounted_disk_is_formatted(self):
+        self.check([])   # positive control for every refusal below
+
+    def test_the_free_part_next_to_felhoms_own_drive_is_refused(self):
+        # The agent could make the „corroborating" managed mount itself: the root wrapper does not trust it.
+        self.check(OWN_SDB1, want_ok=False)
+
+    def test_a_sibling_mounted_raw_only_is_refused(self):
+        self.check([("/dev/sdb1", "/mnt/hdd_1")], want_ok=False)
+
+    def test_a_sibling_mounted_elsewhere_is_refused_even_with_a_managed_mount(self):
+        for elsewhere in ("/srv/data", "/media/usb", "/mnt/a/b", "/home/x"):
+            self.check([("/dev/sdb1", elsewhere), ("/dev/sdb1", "/mnt/felhom-drives/hdd_1")], want_ok=False)
+
+    def test_the_target_itself_mounted_raw_is_refused(self):
+        self.check([("/dev/sdb2", "/mnt/films"), ("/dev/sdb2", "/mnt/felhom-drives/films")], want_ok=False)
+
+    def test_the_target_mounted_only_under_felhom_drives_is_formatted(self):
+        self.check([("/dev/sdb2", "/mnt/felhom-drives/films")])
+
+    def test_an_unlistable_disk_is_refused(self):
+        self.check([], want_ok=False, lsblk_empty=True)
+
+    def test_the_catastrophic_cases_still_refuse(self):
+        self.check([("/dev/sdb1", "/")], want_ok=False)                     # system disk
+        self.check([], nodes=disk_sdb(sdb1_fstype="LVM2_member"), want_ok=False)  # a member signature
+        self.check([], pvs="/dev/sdb1", want_ok=False)                      # an LVM PV (pvs)
+        self.check([], ro="1", want_ok=False)                               # read-only
+
+
+if __name__ == "__main__":
+    unittest.main()

configs/test_felhom_priv_apply.py

@@ -376,5 +376,67 @@ class ControllerImage(unittest.TestCase):
         self.assertIn("controller-image", buf.getvalue())
 
 
+class UnitForget(unittest.TestCase):
+    """R-1124: the agent forgets a decommissioned drive whose disk is gone — the root half removes its unit file, and
+    ONLY a felhom by-UUID drive unit that is disabled, not mounted, and whose /dev/disk/by-uuid link is absent.
+    RED-PROOF: on the pre-R-1124 wrapper `unit-forget` is not a verb (A1 usage, rc 2) — the removal case fails."""
+    NAME = "mnt-hdd_1.mount"
+    PATH = "/etc/systemd/system/mnt-hdd_1.mount"
+    UUID_LINK = "/dev/disk/by-uuid/91d2dc2d-2d28-4929-9bdd-3e11fa2f41ae"
+
+    def host(self, text=LOCAL_UNIT, kind="file", links=(), mounted=()):
+        h = FakeHost()
+        h.installed = {} if text is None else {self.PATH: (kind, text.encode())}
+        h.links, h.mounted, h.removed = set(links), set(mounted), []
+        h.read_installed = lambda p: (None if p not in h.installed else
+                                      (_ for _ in ()).throw(pa.Refused("F1", f"{p} is not a regular file"))
+                                      if h.installed[p][0] != "file" else h.installed[p][1])
+        h.lexists = lambda p: p in h.links
+        h.mountpoints = lambda: set(h.mounted)
+        h.remove = lambda p: h.removed.append(p)
+        return h
+
+    def test_an_absent_disabled_unmounted_drive_unit_is_removed(self):
+        h = self.host()
+        self.assertEqual(run(h, "unit-forget", self.NAME), 0, h.logs)
+        self.assertEqual(h.removed, [self.PATH])
+
+    def test_already_gone_is_fine(self):
+        h = self.host(text=None)
+        self.assertEqual(run(h, "unit-forget", self.NAME), 0, h.logs)
+        self.assertEqual(h.removed, [])
+
+    def refused(self, h, rule, *argv):
+        rc = run(h, *(argv or ("unit-forget", self.NAME)))
+        self.assertIn(rc, (2, 3), h.logs)
+        self.assertEqual(h.removed, [], h.logs)
+        self.assertTrue(any(f"[{rule}]" in l for l in h.logs), h.logs)
+
+    def test_F5_a_present_drive_keeps_its_unit(self):
+        self.refused(self.host(links=[self.UUID_LINK]), "F5")
+
+    def test_F4_a_mounted_drive_keeps_its_unit(self):
+        self.refused(self.host(mounted=["/mnt/hdd_1"]), "F4")
+
+    def test_F3_an_enabled_unit_is_not_removed(self):
+        self.refused(self.host(links=["/etc/systemd/system/multi-user.target.wants/" + self.NAME]), "F3")
+
+    def test_F2_a_network_unit_or_a_foreign_unit_is_not_removed(self):
+        self.refused(self.host(text=LOCAL_UNIT.replace("Managed by felhom-agent", "Written by someone")), "F2")
+        net = LOCAL_UNIT.replace("# Managed by felhom-agent", "# Managed by felhom-agent felhom network storage")
+        self.refused(self.host(text=net), "F2")
+
+    def test_F1_a_symlinked_unit_is_not_followed(self):
+        self.refused(self.host(kind="symlink"), "F1")
+
+    def test_A1_names(self):
+        for argv in (("unit-forget", "mnt-hdd_1.automount"), ("unit-forget", "sshd.service"),
+                     ("unit-forget", "../mnt-hdd_1.mount"), ("unit-forget",), ("unit-forget", self.NAME, "x")):
+            self.refused(self.host(), "A1", *argv)
+
+    def test_U3_a_unit_whose_name_lies_about_where_is_not_removed(self):
+        self.refused(self.host(text=LOCAL_UNIT.replace("Where=/mnt/hdd_1", "Where=/mnt/other")), "U3")
+
+
 if __name__ == "__main__":
     unittest.main(verbosity=2)

configs/test_felhom_thin_check.py

@@ -0,0 +1,118 @@
+#!/usr/bin/env python3
+"""Tests for configs/felhom-thin-check (R-1012): it lists the thin pool's device ids next to LVM's and NEVER removes
+anything. Fake `lvs`, `dmsetup` and `thin_ls` record every call; nothing real is touched. Each refusal has a positive
+control (the healthy read), so a test cannot pass because the helper refuses everything.
+Run: python3 configs/test_felhom_thin_check.py   (also run by internal/thincheck's Go test)
+"""
+import os, pathlib, re, subprocess, tempfile, unittest
+
+HELPER = pathlib.Path(__file__).resolve().parent / "felhom-thin-check"
+
+
+class Box:
+    def __init__(self, pool_ids=(1, 2), lvm=(("vm-9201-disk-0", 1), ("vm-9201-disk-1", 2)), attr="twi-aotz--",
+                 held="-", reserve_rc=0, thin_ls_rc=0):
+        self.t = tempfile.TemporaryDirectory()
+        r = pathlib.Path(self.t.name)
+        self.log = r / "calls"
+        self.log.write_text("")
+        lvm_lines = "\\n".join(f"  {n},{i}" for n, i in lvm)
+        pool_lines = "\\n".join(str(i) for i in pool_ids)
+        self.bins = {}
+        for name, body in {
+            "lvs": f'''case "$*" in *lv_attr*) echo "  {attr}";; *thin_id*) printf "{lvm_lines}\\n";; esac''',
+            "dmsetup": f'''case "$1" in
+  status) echo "0 112975872 thin-pool 585 6860/262144 517080/882624 {held} rw discard_passdown queue_if_no_space - 1024";;
+  message) [ "$4" = reserve_metadata_snap ] && exit {reserve_rc}; exit 0;;
+esac''',
+            "thin_ls": f'''[ {thin_ls_rc} -ne 0 ] && exit {thin_ls_rc}; printf "{pool_lines}\\n"''',
+        }.items():
+            p = r / name
+            p.write_text(f'#!/bin/bash\necho "{name} $*" >> "{self.log}"\n{body}\n')
+            p.chmod(0o755)
+            self.bins[name] = p
+        self.lock = r / "lock"
+
+    def run(self, *args):
+        script = (f'source "{HELPER}"; LVS="{self.bins["lvs"]}"; DMSETUP="{self.bins["dmsetup"]}"; '
+                  f'THIN_LS="{self.bins["thin_ls"]}"; LOCK="{self.lock}"; main "$@"')
+        r = subprocess.run(["bash", "-c", script, "x", *args], capture_output=True, text=True)
+        return r.returncode, r.stdout, r.stderr
+
+    def calls(self):
+        return self.log.read_text().splitlines()
+
+
+def messages(calls):
+    return [c.split()[-1] for c in calls if c.startswith("dmsetup message")]
+
+
+class Reads(unittest.TestCase):
+    def test_healthy_pool_lists_both_sides_and_releases(self):
+        b = Box()
+        rc, out, err = b.run("pve", "data")
+        self.assertEqual(rc, 0, err)
+        self.assertEqual(out.splitlines(), ["lvm-before 1 vm-9201-disk-0", "lvm-before 2 vm-9201-disk-1",
+                                            "pool 1", "pool 2",
+                                            "lvm-after 1 vm-9201-disk-0", "lvm-after 2 vm-9201-disk-1", "ok"])
+        self.assertEqual(messages(b.calls()), ["reserve_metadata_snap", "release_metadata_snap"])
+        self.assertIn("thin_ls --metadata-snap --no-headers -o DEV /dev/mapper/pve-data_tmeta", b.calls())
+
+    def test_the_2026_10_14_orphan_is_listed(self):
+        rc, out, _ = Box(pool_ids=(1, 2, 4)).run("pve", "data")
+        self.assertEqual(rc, 0)
+        self.assertIn("pool 4", out.splitlines())
+
+    def test_dashed_names_map_to_the_device_mapper_name(self):
+        b = Box()
+        rc, _, err = b.run("felhom-scratch", "thin-pool")
+        self.assertEqual(rc, 0, err)
+        self.assertIn("dmsetup status felhom--scratch-thin--pool-tpool", b.calls())
+
+
+class Refuses(unittest.TestCase):
+    def refused(self, box, rc_want, *args):
+        rc, out, err = box.run(*args)
+        self.assertEqual(rc, rc_want, (out, err))
+        self.assertNotIn("ok", out.splitlines())
+        return box.calls()
+
+    def test_arguments(self):
+        for args in (("pve",), ("pve", "data", "x"), ("pve;id", "data"), ("-pve", "data"), ("pve", "../data"),
+                     ("pve", "da ta")):
+            calls = self.refused(Box(), 1, *args)
+            self.assertEqual(calls, [], args)
+
+    def test_not_a_thin_pool(self):
+        calls = self.refused(Box(attr="Vwi-aotz--"), 1, "pve", "vm-9201-disk-1")
+        self.assertEqual(messages(calls), [])
+
+    def test_a_snapshot_someone_else_holds_is_never_touched(self):
+        b = Box(held="1234")
+        rc, _, err = b.run("pve", "data")
+        self.assertEqual(rc, 3)
+        self.assertEqual(messages(b.calls()), [])  # neither reserve nor release
+        # the reason line ends with the reason, not the exit code (found live on demo-hp 2026-10-17)
+        self.assertTrue(err.strip().endswith("not touched"), err)
+
+    def test_reserve_failure_does_not_release(self):
+        calls = self.refused(Box(reserve_rc=1), 3, "pve", "data")
+        self.assertEqual(messages(calls), ["reserve_metadata_snap"])
+
+    def test_thin_ls_failure_still_releases(self):
+        calls = self.refused(Box(thin_ls_rc=1), 4, "pve", "data")
+        self.assertEqual(messages(calls), ["reserve_metadata_snap", "release_metadata_snap"])
+
+
+class NeverDeletes(unittest.TestCase):
+    def test_the_only_pool_messages_are_reserve_and_release(self):
+        src = HELPER.read_text()
+        sent = re.findall(r'"\$DMSETUP" message "\$TPOOL" 0 (\S+)', src)
+        self.assertEqual(sorted(set(sent)), ["release_metadata_snap", "reserve_metadata_snap"])
+        for word in ("delete", "lvremove", "thin_repair", "create_thin", "create_snap", " rm "):
+            code = "\n".join(l for l in src.splitlines() if not l.lstrip().startswith("#"))
+            self.assertNotIn(word, code, word)
+
+
+if __name__ == "__main__":
+    unittest.main(verbosity=2)

configs/test_felhom_vzdump_partial_clean.py

@@ -0,0 +1,198 @@
+#!/usr/bin/env python3
+"""Tests for configs/felhom-vzdump-partial-clean (R-1006): it removes a DEAD half-written vzdump backup and refuses
+everything else. Each refusal has a positive control: the same setup without the one condition is removed — so a
+test cannot pass because the helper refuses everything. A fake storage.cfg and a fake /proc; nothing real is touched.
+Run: python3 configs/test_felhom_vzdump_partial_clean.py   (also run by internal/backup's Go test)
+"""
+import os, pathlib, subprocess, tempfile, time, unittest
+
+HELPER = pathlib.Path(__file__).resolve().parent / "felhom-vzdump-partial-clean"
+BOOT = 1_791_967_481
+NAME = "vzdump-lxc-9201-2026_10_14-10_42_25"
+
+
+class Box:
+    """A temp tree: a dir storage `local` (path vz), an lvmthin `local-lvm`, a dir storage `other`, and a fake /proc."""
+
+    def __init__(self):
+        self.t = tempfile.TemporaryDirectory()
+        r = pathlib.Path(self.t.name)
+        self.vz, self.other, self.nas, self.proc = r / "vz", r / "other", r / "nas", r / "proc"
+        for d in (self.vz / "dump", self.other / "dump", self.nas / "dump", self.proc):
+            d.mkdir(parents=True)
+        (self.proc / "stat").write_text(f"cpu  1 2 3\nbtime {BOOT}\n")
+        self.cfg = r / "storage.cfg"
+        self.cfg.write_text(f"dir: local\n\tpath {self.vz}\n\tcontent iso,backup\n\n"
+                            f"lvmthin: local-lvm\n\tthinpool data\n\tvgname pve\n\n"
+                            f"dir: other\n\tpath {self.other}\n\tcontent backup\n\n"
+                            f"nfs: nas\n\tpath {self.nas}\n\tserver 192.0.2.1\n\texport /x\n\tcontent backup\n")
+        self.pid = 1000
+
+    def partial(self, name=NAME + ".tar.dat", mtime=BOOT - 100, where=None, size=4096):
+        p = (where or self.vz / "dump") / name
+        if name.endswith(".tmp"):
+            (p / "etc" / "vzdump").mkdir(parents=True)
+            (p / "etc" / "vzdump" / "pct.conf").write_text("x")
+            for q in [p / "etc" / "vzdump" / "pct.conf", p / "etc" / "vzdump", p / "etc", p]:
+                os.utime(q, (mtime, mtime))
+        else:
+            p.write_bytes(b"\0" * size)
+            os.utime(p, (mtime, mtime))
+        return p
+
+    def process(self, cmdline, fds=(), cwd=None):
+        self.pid += 1
+        d = self.proc / str(self.pid)
+        (d / "fd").mkdir(parents=True)
+        (d / "cmdline").write_bytes(b"\0".join(a.encode() for a in cmdline) + b"\0")
+        for i, f in enumerate(fds):
+            os.symlink(str(f), d / "fd" / str(i + 3))
+        os.symlink(str(cwd or "/"), d / "cwd")
+
+    def run(self, storage, vmid, name, quiet=120):
+        script = (f'source "{HELPER}"; STORAGE_CFG="{self.cfg}"; PROC="{self.proc}"; QUIET_S={quiet}; '
+                  f'main "$@"')
+        r = subprocess.run(["bash", "-c", script, "x", storage, str(vmid), name], capture_output=True, text=True)
+        return r.returncode, r.stdout + r.stderr
+
+
+class Removes(unittest.TestCase):
+    def test_dead_tar_dat_before_boot(self):
+        b = Box(); p = b.partial()
+        rc, out = b.run("local", 9201, p.name)
+        self.assertEqual(rc, 0, out); self.assertFalse(p.exists()); self.assertIn("written before this boot", out)
+
+    def test_dead_tmp_dir_before_boot(self):
+        b = Box(); p = b.partial(NAME + ".tmp")
+        rc, out = b.run("local", 9201, p.name)
+        self.assertEqual(rc, 0, out); self.assertFalse(p.exists())
+
+    def test_quiet_partial_after_boot(self):
+        b = Box(); p = b.partial(mtime=int(time.time()) - 300)
+        rc, out = b.run("local", 9201, p.name)
+        self.assertEqual(rc, 0, out); self.assertFalse(p.exists()); self.assertIn("untouched", out)
+
+    def test_qemu_vma_dat(self):
+        b = Box(); p = b.partial("vzdump-qemu-341-2026_10_14-10_42_25.vma.dat")
+        rc, out = b.run("local", 341, p.name)
+        self.assertEqual(rc, 0, out); self.assertFalse(p.exists())
+
+
+class Keeps(unittest.TestCase):
+    """Each case: (setup that must be REFUSED, the same setup that must be REMOVED)."""
+
+    def pair(self, refuse, control, quiet=120):
+        b = Box(); target, args = refuse(b)
+        rc, out = b.run(*args, quiet=quiet)
+        self.assertNotEqual(rc, 0, f"must refuse: {out}")
+        self.assertIn("REFUSED", out)
+        self.assertTrue(os.path.lexists(target), "a refused entry must still exist")
+        c = Box(); ctarget, cargs = control(c)
+        crc, cout = c.run(*cargs, quiet=quiet)
+        self.assertEqual(crc, 0, f"positive control must be removed: {cout}")
+        self.assertFalse(os.path.lexists(ctarget))
+
+    def test_complete_archive_of_same_vmid(self):
+        self.pair(lambda b: (b.partial(NAME + ".tar.zst"), ("local", 9201, NAME + ".tar.zst")),
+                  lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_log_and_notes(self):
+        for suffix in (".log", ".tar.zst.notes"):
+            self.pair(lambda b, s=suffix: (b.partial(NAME + s), ("local", 9201, NAME + s)),
+                      lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_partial_of_another_vmid(self):
+        other = "vzdump-lxc-9202-2026_10_14-10_42_25.tar.dat"
+        self.pair(lambda b: (b.partial(other), ("local", 9201, other)),
+                  lambda b: (b.partial(other), ("local", 9202, other)))
+
+    def test_partial_on_a_non_dir_storage(self):
+        self.pair(lambda b: (b.partial(), ("local-lvm", 9201, NAME + ".tar.dat")),
+                  lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_partial_on_a_network_storage_with_a_path(self):
+        # an nfs storage HAS a path: only the „dir only" rule (both its guards) can refuse it
+        self.pair(lambda b: (b.partial(where=b.nas / "dump"), ("nas", 9201, NAME + ".tar.dat")),
+                  lambda b: (b.partial(where=b.other / "dump"), ("other", 9201, NAME + ".tar.dat")))
+
+    def test_partial_on_another_storage_than_named(self):
+        self.pair(lambda b: (b.partial(where=b.other / "dump"), ("local", 9201, NAME + ".tar.dat")),
+                  lambda b: (b.partial(where=b.other / "dump"), ("other", 9201, NAME + ".tar.dat")))
+
+    def test_unknown_storage(self):
+        self.pair(lambda b: (b.partial(), ("nosuch", 9201, NAME + ".tar.dat")),
+                  lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_vzdump_task_in_flight(self):
+        def refuse(b):
+            p = b.partial(); b.process(["task UPID:demo-hp:001:002:6ACF4071:vzdump:9201:felhom-agent@pve!agent:"]); return p, ("local", 9201, p.name)
+        def control(b):
+            p = b.partial(); b.process(["task UPID:demo-hp:001:002:6ACF4071:vzrestore:9202:root@pam:"]); return p, ("local", 9201, p.name)
+        self.pair(refuse, control)
+
+    def test_any_vzdump_job_even_without_the_vmid(self):
+        # a job for several guests: `vzdump::` — and a vzdump for ANOTHER guest — both refuse
+        for cmd in (["task UPID:demo-hp:001:002:6ACF4071:vzdump::root@pam:"],
+                    ["task UPID:demo-hp:001:002:6ACF4071:vzdump:9202:felhom-agent@pve!agent:"]):
+            def refuse(b, c=cmd):
+                p = b.partial(); b.process(c); return p, ("local", 9201, p.name)
+            def control(b):
+                p = b.partial(); b.process(["task UPID:demo-hp:001:002:6ACF4071:vzstart:9202:root@pam:"]); return p, ("local", 9201, p.name)
+            self.pair(refuse, control)
+
+    def test_vzdump_command_for_the_vmid(self):
+        def refuse(b):
+            p = b.partial(); b.process(["/usr/bin/perl", "/usr/bin/vzdump", "9201", "--storage", "local"]); return p, ("local", 9201, p.name)
+        def control(b):
+            p = b.partial(); b.process(["/usr/bin/perl", "/usr/bin/vzdumpfoo", "9201"]); return p, ("local", 9201, p.name)
+        self.pair(refuse, control)
+
+    def test_file_held_open_by_an_orphan_writer(self):
+        def refuse(b):
+            p = b.partial(); b.process(["zstd", "--threads=1"], fds=[p]); return p, ("local", 9201, p.name)
+        def control(b):
+            p = b.partial(); b.process(["zstd", "--threads=1"], fds=[b.vz / "dump" / "something-else"]); return p, ("local", 9201, p.name)
+        self.pair(refuse, control)
+
+    def test_tmp_dir_with_a_process_inside(self):
+        def refuse(b):
+            p = b.partial(NAME + ".tmp"); b.process(["tar", "cpf", "-"], cwd=p / "etc"); return p, ("local", 9201, p.name)
+        def control(b):
+            p = b.partial(NAME + ".tmp"); b.process(["tar", "cpf", "-"], cwd=b.vz); return p, ("local", 9201, p.name)
+        self.pair(refuse, control)
+
+    def test_recently_written_after_boot(self):
+        self.pair(lambda b: (b.partial(mtime=int(time.time()) - 10), ("local", 9201, NAME + ".tar.dat")),
+                  lambda b: (b.partial(mtime=int(time.time()) - 300), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_symlink(self):
+        def refuse(b):
+            real = b.partial(where=b.vz)
+            link = b.vz / "dump" / (NAME + ".tar.dat")
+            os.symlink(real, link)
+            return link, ("local", 9201, link.name)
+        # quiet=0: the link was made just now, so only the symlink rule may refuse it
+        self.pair(refuse, lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")), quiet=0)
+
+    def test_near_miss_names(self):
+        for bad in (NAME + ".tar.dat.bak", "vzdump-lxc-92011-2026_10_14-10_42_25.tar.dat", NAME + ".tar.da",
+                    "x" + NAME + ".tar.dat", NAME + ".TAR.DAT", "vzdump-lxc-9201-2026_10_14-10_42.tar.dat"):
+            self.pair(lambda b, n=bad: (b.partial(n), ("local", 9201, n)),
+                      lambda b: (b.partial(), ("local", 9201, NAME + ".tar.dat")))
+
+    def test_path_traversal_and_bad_args(self):
+        b = Box(); p = b.partial()
+        for args in (("local", 9201, "../dump/" + p.name), ("local", 9201, ""), ("loc al", 9201, p.name),
+                     ("local", "9201x", p.name), ("local", 12, p.name)):
+            rc, out = b.run(*args)
+            self.assertNotEqual(rc, 0, f"{args}: {out}")
+        self.assertTrue(p.exists())
+
+    def test_wrong_kind(self):
+        def refuse(b):  # a FILE named .tmp
+            p = b.vz / "dump" / (NAME + ".tmp"); p.write_bytes(b"x"); os.utime(p, (BOOT - 100, BOOT - 100)); return p, ("local", 9201, p.name)
+        self.pair(refuse, lambda b: (b.partial(NAME + ".tmp"), ("local", 9201, NAME + ".tmp")))
+
+
+if __name__ == "__main__":
+    unittest.main()

go.mod

@@ -1,7 +1,8 @@
 module gitea.dooplex.hu/admin/felhom-agent
 
-go 1.25.0
+go 1.27.2
 
-require golang.org/x/crypto v0.52.0
-
-require golang.org/x/sys v0.45.0 // indirect
+require (
+	golang.org/x/crypto v0.58.0
+	golang.org/x/sys v0.49.0
+)

go.sum

@@ -1,6 +1,6 @@
-golang.org/x/crypto v0.52.0 h1:RMs7fP2rXdep0CftQlK8Uf+kibLm7qkCcradZWYz988=
-golang.org/x/crypto v0.52.0/go.mod h1:1QgfPxDqh0T2M/elOJtp9RvuR95kVjir0e6/BvEmGbc=
-golang.org/x/sys v0.45.0 h1:dO4czNzziLiiXplLQgBCEpCvXQ3dnkn0SdaZSYdQ+FY=
-golang.org/x/sys v0.45.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
-golang.org/x/term v0.43.0 h1:S4RLU2sB31O/NCl+zFN9Aru9A/Cq2aqKpTZJ6B+DwT4=
-golang.org/x/term v0.43.0/go.mod h1:lrhlHNdQJHO+1qVYiHfFKVuVioJIheAc3fBSMFYEIsk=
+golang.org/x/crypto v0.58.0 h1:COkYLr4k7nDI4r6QtMML2AfzeIswWHAJthLS6K/M/54=
+golang.org/x/crypto v0.58.0/go.mod h1:Xh+kl5A+M0gMiWDytH2Hhr5Z7WbNCpv/eB/EFgfSk+w=
+golang.org/x/sys v0.49.0 h1:XbzkgYJHdqh/8m2Uu0W/dQv8nktxx4BFHp1M0gROTXA=
+golang.org/x/sys v0.49.0/go.mod h1:hNLxWAXmnKAxqDtdwIYC4bM9oQPEecfsnNMuSxOs3og=
+golang.org/x/term v0.47.0 h1:cpyOwQkr2Nzxe2p6KnrTG7Ei2wtzjsuSjhpXXepZ/BM=
+golang.org/x/term v0.47.0/go.mod h1:efAfAuG6RKABvg1ArGebRkvX6LXul14SM67o2e4mUCY=

internal/backup/backup_state.go

@@ -39,13 +39,25 @@ type savedSuccess struct {
 	target string
 	vmid   int
 	at     time.Time
+	// recent (R-973) holds the start times of the newest successes (at most maxRecentSuccesses, newest
+	// last) — what the local-API backup cadence cap counts, so an agent restart does not reset the cap.
+	recent []time.Time
+	// refusedAt (R-973) is the newest time the cap refused a backup of this guest on this tier.
+	refusedAt time.Time
 }
 
-// backupSuccessJSON is one entry on disk.
+// maxRecentSuccesses bounds the per-tier success history kept for the cadence cap (R-973). The cap
+// allows at most 3 runs in a window, so 8 is ample.
+const maxRecentSuccesses = 8
+
+// backupSuccessJSON is one entry on disk. Recent and LastRefusedAt are additive (R-973): a file written
+// before them reads as "no history", and an older agent ignores them.
 type backupSuccessJSON struct {
-	Target    string `json:"target"`
-	VMID      int    `json:"vmid"`
-	StartedAt string `json:"started_at"`
+	Target        string   `json:"target"`
+	VMID          int      `json:"vmid"`
+	StartedAt     string   `json:"started_at"`
+	Recent        []string `json:"recent,omitempty"`
+	LastRefusedAt string   `json:"last_refused_at,omitempty"`
 }
 
 func backupStateKey(target string, vmid int) string { return target + "/" + strconv.Itoa(vmid) }
@@ -67,7 +79,19 @@ func NewBackupSuccessState(path string) *BackupSuccessState {
 		if perr != nil {
 			continue // one unreadable entry must not lose the others
 		}
-		s.last[backupStateKey(e.Target, e.VMID)] = savedSuccess{target: e.Target, vmid: e.VMID, at: t.UTC()}
+		ss := savedSuccess{target: e.Target, vmid: e.VMID, at: t.UTC()}
+		for _, r := range e.Recent {
+			if rt, rerr := time.Parse(time.RFC3339, r); rerr == nil {
+				ss.recent = append(ss.recent, rt.UTC())
+			}
+		}
+		if len(ss.recent) == 0 {
+			ss.recent = []time.Time{ss.at} // a pre-R-973 file: its one known success counts
+		}
+		if rt, rerr := time.Parse(time.RFC3339, e.LastRefusedAt); rerr == nil {
+			ss.refusedAt = rt.UTC()
+		}
+		s.last[backupStateKey(e.Target, e.VMID)] = ss
 	}
 	return s
 }
@@ -85,13 +109,72 @@ func (s *BackupSuccessState) RecordBackupSuccess(target string, b hub.Backup) er
 	s.mu.Lock()
 	defer s.mu.Unlock()
 	k := backupStateKey(target, b.VMID)
-	if old, ok := s.last[k]; ok && !t.After(old.at) {
+	old, had := s.last[k]
+	if had && !t.After(old.at) {
 		return nil
 	}
-	s.last[k] = savedSuccess{target: target, vmid: b.VMID, at: t.UTC()}
+	ss := savedSuccess{target: target, vmid: b.VMID, at: t.UTC(), refusedAt: old.refusedAt}
+	ss.recent = append(append([]time.Time(nil), old.recent...), t.UTC())
+	if len(ss.recent) > maxRecentSuccesses {
+		ss.recent = ss.recent[len(ss.recent)-maxRecentSuccesses:]
+	}
+	s.last[k] = ss
 	return s.saveLocked()
 }
 
+// RecentBackupSuccesses returns the saved success start times for this tier and guest that are not
+// older than since (R-973 — the cadence cap's restart-proof count).
+func (s *BackupSuccessState) RecentBackupSuccesses(target string, vmid int, since time.Time) []time.Time {
+	if s == nil {
+		return nil
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	var out []time.Time
+	for _, t := range s.last[backupStateKey(target, vmid)].recent {
+		if !t.Before(since) {
+			out = append(out, t)
+		}
+	}
+	return out
+}
+
+// RecordBackupRefusal saves that the cadence cap refused a backup of this guest on this tier at at
+// (R-973), so the operator signal it raises survives an agent restart.
+func (s *BackupSuccessState) RecordBackupRefusal(target string, vmid int, at time.Time) error {
+	if s == nil {
+		return nil
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	k := backupStateKey(target, vmid)
+	ss, ok := s.last[k]
+	if !ok {
+		return nil // nothing to cap against without a success; the cap never refuses then
+	}
+	if !at.After(ss.refusedAt) {
+		return nil
+	}
+	ss.refusedAt = at.UTC()
+	s.last[k] = ss
+	return s.saveLocked()
+}
+
+// LastBackupRefusal returns the newest saved cadence-cap refusal across every tier and guest.
+func (s *BackupSuccessState) LastBackupRefusal() (target string, vmid int, at time.Time, ok bool) {
+	if s == nil {
+		return "", 0, time.Time{}, false
+	}
+	s.mu.Lock()
+	defer s.mu.Unlock()
+	for _, e := range s.last {
+		if !e.refusedAt.IsZero() && e.refusedAt.After(at) {
+			target, vmid, at, ok = e.target, e.vmid, e.refusedAt, true
+		}
+	}
+	return target, vmid, at, ok
+}
+
 // LastKnownSuccess returns the newest saved success for this tier and guest (ok=false = none on file).
 func (s *BackupSuccessState) LastKnownSuccess(target string, vmid int) (time.Time, bool) {
 	if s == nil {
@@ -129,7 +212,14 @@ func (s *BackupSuccessState) KnownBackupSuccesses() []hub.Backup {
 func (s *BackupSuccessState) saveLocked() error {
 	entries := make([]backupSuccessJSON, 0, len(s.last))
 	for _, e := range s.last {
-		entries = append(entries, backupSuccessJSON{Target: e.target, VMID: e.vmid, StartedAt: e.at.Format(time.RFC3339)})
+		j := backupSuccessJSON{Target: e.target, VMID: e.vmid, StartedAt: e.at.Format(time.RFC3339)}
+		for _, r := range e.recent {
+			j.Recent = append(j.Recent, r.Format(time.RFC3339))
+		}
+		if !e.refusedAt.IsZero() {
+			j.LastRefusedAt = e.refusedAt.Format(time.RFC3339)
+		}
+		entries = append(entries, j)
 	}
 	// Deterministic file content (Go's map order is random).
 	sort.Slice(entries, func(i, j int) bool {

internal/backup/backup_state_test.go

@@ -74,3 +74,40 @@ func TestBackupSuccessState_KnownBackupSuccessesSurviveAReopen(t *testing.T) {
 		t.Fatal("uncovered_volumes must marshal as [], not null")
 	}
 }
+
+// R-973: the cadence cap's history and refusals survive a reopen; the history is bounded; a file
+// written before R-973 counts its one known success (never "no history" for a tier that has a copy).
+func TestBackupSuccessState_R973_RecentHistoryAndRefusalSurviveAReopen(t *testing.T) {
+	path := filepath.Join(t.TempDir(), "backup-success-state.json")
+	s := NewBackupSuccessState(path)
+	base := time.Date(2026, 10, 18, 2, 30, 0, 0, time.UTC)
+	for i := 0; i < 10; i++ {
+		at := base.Add(time.Duration(i) * time.Hour)
+		if err := s.RecordBackupSuccess("local", hub.Backup{VMID: 9201, Success: true, StartedAt: at.Format(time.RFC3339)}); err != nil {
+			t.Fatal(err)
+		}
+	}
+	refused := base.Add(11 * time.Hour)
+	if err := s.RecordBackupRefusal("local", 9201, refused); err != nil {
+		t.Fatal(err)
+	}
+	r := NewBackupSuccessState(path)
+	all := r.RecentBackupSuccesses("local", 9201, time.Time{})
+	if len(all) != maxRecentSuccesses || !all[len(all)-1].Equal(base.Add(9*time.Hour)) {
+		t.Fatalf("recent history after a reopen: %v (want the newest %d, ending at +9h)", all, maxRecentSuccesses)
+	}
+	if got := r.RecentBackupSuccesses("local", 9201, base.Add(7*time.Hour)); len(got) != 3 {
+		t.Fatalf("successes since +7h: %d, want 3", len(got))
+	}
+	if tg, vm, at, ok := r.LastBackupRefusal(); !ok || tg != "local" || vm != 9201 || !at.Equal(refused) {
+		t.Fatalf("refusal after a reopen: %q %d %v %v", tg, vm, at, ok)
+	}
+
+	old := filepath.Join(t.TempDir(), "old.json")
+	if err := os.WriteFile(old, []byte(`[{"target":"felhom-pbs","vmid":9201,"started_at":"2026-10-17T03:00:00Z"}]`), 0o600); err != nil {
+		t.Fatal(err)
+	}
+	if got := NewBackupSuccessState(old).RecentBackupSuccesses("felhom-pbs", 9201, time.Time{}); len(got) != 1 {
+		t.Fatalf("a pre-R-973 file must count its one success: %v", got)
+	}
+}

internal/backup/backup_test.go

@@ -2,9 +2,9 @@ package backup
 
 import (
 	"context"
-	"fmt"
 	"encoding/json"
 	"errors"
+	"fmt"
 	"io"
 	"log/slog"
 	"testing"

internal/backup/prune_test.go

@@ -93,7 +93,7 @@ func TestPrune_KeepLastClamp(t *testing.T) {
 			t.Fatalf("keep-last must NEVER be < 1 (would prune the fresh backup), got %d for %d", b.KeepLast(), c.set)
 		}
 	}
-	if spec := (config.BackupConfig{}).PruneBackupsSpec(); spec != "keep-last=3" {
-		t.Fatalf("default PruneBackupsSpec = %q, want keep-last=3", spec)
+	if spec := (config.BackupConfig{}).PruneBackupsSpec(); spec != "keep-daily=3" {
+		t.Fatalf("default PruneBackupsSpec = %q, want keep-daily=3 (R-1149)", spec)
 	}
 }

internal/backup/r1012_thincheck_test.go

@@ -0,0 +1,67 @@
+package backup
+
+import (
+	"context"
+	"errors"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1012: a snapshot backup that fails carries the orphan finding in its error — the text the controller mails to
+// the operator (`whole_guest_backup_failed`). The check runs ONCE, after the failure — never before the run.
+// RED-PROOF (P1/redproof-thincheck.txt): with explainFailure's append removed this test fails. RED-PROOF (2026-10-17
+// night, felhom-agent REPORT.md): with the v0.159.0 pre-run check restored, calls = 2 and this test fails.
+func TestBackup_SnapshotFailure_NamesTheOrphan(t *testing.T) {
+	const why = "probable cause: thin pool pve/data holds device 4 that LVM does not know"
+	for name, api := range map[string]*fakeBackupAPI{
+		"task fails":  {vzdumpUPID: "UPID:x", waitErr: errors.New("task failed: job errors")},
+		"start fails": {vzdumpErr: errors.New("vzdump: 500")},
+	} {
+		t.Run(name, func(t *testing.T) {
+			r := NewBackupRunner(api, "local", "", "", "", quiet())
+			calls := 0
+			r.SetThinCheck(func(context.Context, int) string { calls++; return why })
+			rec, err := r.Backup(context.Background(), 9201)
+			if err == nil || rec.Success {
+				t.Fatalf("want a failure, got %+v", rec)
+			}
+			if !strings.Contains(rec.Error, why) {
+				t.Fatalf("the failure does not name the orphan: %q", rec.Error)
+			}
+			if calls != 1 {
+				t.Fatalf("thin check ran %d times, want 1 (after the failure only)", calls)
+			}
+		})
+	}
+}
+
+// A successful backup never runs the check (it takes a metadata snapshot of the live pool, and nobody needs the answer)
+// and keeps a clean record; a stop-mode backup takes no snapshot, so the check never runs; an unwired runner behaves
+// exactly as before. RED-PROOF (2026-10-17 night): with the v0.159.0 pre-run check restored, the success leg fails
+// with calls=1.
+func TestBackup_ThinCheck_SilentWhenNotApplicable(t *testing.T) {
+	ok := &fakeBackupAPI{vzdumpUPID: "UPID:x",
+		content: []proxmox.StorageContent{{VolID: "local:backup/vzdump-lxc-9201-x", Content: "backup", VMID: 9201, Size: 100, CTime: 1}}}
+	r := NewBackupRunner(ok, "local", "", "", "", quiet())
+	calls := 0
+	r.SetThinCheck(func(context.Context, int) string { calls++; return "probable cause: x" })
+	rec, err := r.Backup(context.Background(), 9201)
+	if err != nil || !rec.Success || rec.Error != "" || calls != 0 {
+		t.Fatalf("success: err=%v rec=%+v calls=%d (want no check on a successful backup, clean record)", err, rec, calls)
+	}
+
+	stop := NewBackupRunner(&fakeBackupAPI{vzdumpUPID: "UPID:x", waitErr: errors.New("boom")}, "local", proxmox.ModeStop, "", "", quiet())
+	calls = 0
+	stop.SetThinCheck(func(context.Context, int) string { calls++; return "probable cause: x" })
+	rec, _ = stop.Backup(context.Background(), 9201)
+	if calls != 0 || strings.Contains(rec.Error, "probable cause") {
+		t.Fatalf("stop mode: calls=%d error=%q", calls, rec.Error)
+	}
+
+	plain := NewBackupRunner(&fakeBackupAPI{vzdumpUPID: "UPID:x", waitErr: errors.New("boom")}, "local", "", "", "", quiet())
+	if rec, _ := plain.Backup(context.Background(), 9201); rec.Error != "boom" {
+		t.Fatalf("unwired: error = %q, want the task error unchanged", rec.Error)
+	}
+}

internal/backup/r685_space_test.go

@@ -81,3 +81,28 @@ func TestR685_FailsOpen(t *testing.T) {
 		}
 	}
 }
+
+// TestR1006_LeftoverCleanupRunsBeforeThePreflight — the night of 2026-10-14: a 12.2 GB half-written archive from a power
+// cut held the space, so every later backup was skipped. The cleanup step runs BEFORE the preflight reads the free space:
+// here it „removes" the leftover (free 4 → 16 GiB) and the backup then fits and runs.
+//
+// RED-PROOF: drop the r.beforeBackup call from backup() (or move it after spaceFits) → the backup is skipped → fails.
+func TestR1006_LeftoverCleanupRunsBeforeThePreflight(t *testing.T) {
+	api := spaceAPI(4*gib, 8182759056, "dir")
+	r := NewBackupRunner(api, "local", proxmox.ModeSnapshot, "", "keep-last=1", quiet())
+	var got []string
+	r.SetBeforeBackup(func(_ context.Context, storage string, vmid int) {
+		got = append(got, storage)
+		if vmid != 9201 {
+			t.Errorf("cleanup asked for vmid %d", vmid)
+		}
+		api.storages[0].Avail = 16 * gib
+	})
+	_, _ = r.Backup(context.Background(), 9201)
+	if len(got) != 1 || got[0] != "local" {
+		t.Fatalf("the cleanup must run once for this target, got %v", got)
+	}
+	if len(api.vzdumps) != 1 {
+		t.Fatalf("after the cleanup freed the space the backup must run, vzdumps=%d", len(api.vzdumps))
+	}
+}

internal/backup/r689_golden_test.go

@@ -48,7 +48,7 @@ func TestR689_GuestBackupArchiveShapes(t *testing.T) {
 		{proxmox.StorageContent{VolID: "felhom-pbs:backup/vm/300/2026-07-28T05:31:14Z", VMID: 300}, true},
 		{proxmox.StorageContent{VolID: "local:backup/felhom-golden-0.236.0.tar.zst"}, false},
 		{proxmox.StorageContent{VolID: "local:backup/vzdump-lxc-9201-x.tar.zst", VMID: 9202}, false}, // vmid disagrees with the name
-		{proxmox.StorageContent{VolID: "felhom-pbs:backup/ct/9201/2026-07-28T05:31:14Z"}, false},       // no vmid reported
+		{proxmox.StorageContent{VolID: "felhom-pbs:backup/ct/9201/2026-07-28T05:31:14Z"}, false},     // no vmid reported
 	} {
 		if ok, why := guestBackupArchive(c.e); ok != c.ok {
 			t.Errorf("%s vmid=%d: ok=%v (%s), want %v", c.e.VolID, c.e.VMID, ok, why, c.ok)

internal/backup/runner.go

@@ -58,8 +58,16 @@ type BackupRunner struct {
 	// tier cannot have that accident — its keep_last defaults to 0 (never prune), so any value
 	// there is a deliberate act.
 	allowPBSPrune bool
-	logger        *slog.Logger
-	now           func() time.Time
+	// beforeBackup runs just before the space preflight (R-1006): it removes a dead half-written vzdump of this guest on
+	// this target, so a leftover from a power cut cannot hold the space this backup needs. nil = nothing.
+	beforeBackup func(ctx context.Context, storage string, vmid int)
+	// thinCheck (R-1012) names an orphan thin device in the guest's pools in plain words ("" = none / not checked). It
+	// runs ONLY when a snapshot-mode vzdump has failed (start or task), so the failure the controller mails to the
+	// operator says why; a successful backup never runs it (it takes a metadata snapshot of the live pool). It never
+	// removes anything. nil = not wired.
+	thinCheck func(ctx context.Context, vmid int) string
+	logger    *slog.Logger
+	now       func() time.Time
 	// rejected remembers the volids already announced by warnRejectedArchiveOnce, so an incomplete
 	// archive is reported ONCE rather than on every 5-minute due-check. Bounded in practice: one
 	// entry per aborted upload, and a process restart clears it. Guarded by rejectedMu because the
@@ -102,6 +110,27 @@ func NewBackupRunnerFull(api BackupAPI, target string, mode proxmox.BackupMode,
 		now: func() time.Time { return time.Now().UTC() }}
 }
 
+// SetBeforeBackup sets the step run just before the space preflight (R-1006: the dead-leftover cleanup).
+func (r *BackupRunner) SetBeforeBackup(fn func(ctx context.Context, storage string, vmid int)) {
+	r.beforeBackup = fn
+}
+
+// SetThinCheck wires the R-1012 orphan check (see BackupRunner.thinCheck).
+func (r *BackupRunner) SetThinCheck(fn func(ctx context.Context, vmid int) string) { r.thinCheck = fn }
+
+// explainFailure appends the orphan finding to a failed snapshot-mode backup's error — the only place the check runs, so
+// it sees the pool as the failure left it. Bounded, so a stuck helper cannot hold the job.
+func (r *BackupRunner) explainFailure(ctx context.Context, vmid int, rec *hub.Backup) {
+	if r.thinCheck == nil || r.mode != proxmox.ModeSnapshot {
+		return
+	}
+	cctx, cancel := context.WithTimeout(context.WithoutCancel(ctx), 2*time.Minute)
+	defer cancel()
+	if why := r.thinCheck(cctx, vmid); why != "" {
+		rec.Error = rec.Error + " — " + why
+	}
+}
+
 // localPruneSpec returns the `--prune-backups` spec to apply to THIS backup, or "" to skip pruning. It
 // applies the configured retention ONLY when the target is a non-PBS storage — PBS offsite retention is a
 // separate lifecycle and must never be pruned by the per-run flag (§9). Fail-safe: if the target's type
@@ -183,6 +212,9 @@ func (r *BackupRunner) backup(ctx context.Context, vmid int, onSnapshot func())
 	// R-685 (v0.134.0): will the new archive FIT on a local target? Asked before anything runs, so a
 	// target that cannot hold it is a named SKIP with the numbers, not a nightly "No space left on
 	// device" that only the vzdump log explains (demo-hp, every night from 2026-09-23 — R-684).
+	if r.beforeBackup != nil {
+		r.beforeBackup(ctx, r.target, vmid)
+	}
 	if ok, why := r.spaceFits(ctx, vmid); !ok {
 		rec.Error = BackupSkipNoSpacePrefix + why
 		rec.DurationSeconds = time.Since(start).Seconds()
@@ -190,12 +222,15 @@ func (r *BackupRunner) backup(ctx context.Context, vmid int, onSnapshot func())
 		return rec, fmt.Errorf("backup: %s", rec.Error)
 	}
 
+	// R-1012: no orphan check here. It takes a metadata snapshot of the live pool, and its answer is only needed when the
+	// snapshot step has failed — explainFailure runs it then (pinned by TestBackup_ThinCheck_SilentWhenNotApplicable).
 	upid, err := r.api.Vzdump(ctx, proxmox.VzdumpOptions{
 		VMID: vmid, Storage: r.target, Mode: r.mode, Notes: r.notes,
 		PruneBackups: r.localPruneSpec(ctx), // local target → keep-last=N; PBS/unknown → "" (no prune)
 	})
 	if err != nil {
 		rec.Error = err.Error()
+		r.explainFailure(ctx, vmid, &rec)
 		rec.DurationSeconds = time.Since(start).Seconds()
 		return rec, fmt.Errorf("backup: vzdump vmid %d: %w", vmid, err)
 	}
@@ -209,6 +244,7 @@ func (r *BackupRunner) backup(ctx context.Context, vmid int, onSnapshot func())
 		}
 		if _, err := r.api.WaitTask(ctx, upid, proxmox.WaitOptions{Timeout: r.waitTimeout}); err != nil {
 			rec.Error = err.Error()
+			r.explainFailure(ctx, vmid, &rec)
 			rec.DurationSeconds = time.Since(start).Seconds()
 			return rec, fmt.Errorf("backup: vzdump task vmid %d: %w", vmid, err)
 		}

internal/capability/manifest.go

@@ -91,9 +91,11 @@ var manifest = []Capability{
 	{"disk-mkfs-xfs", "guarded format (xfs)", "/usr/local/sbin/felhom-mkfs-guarded", []string{"/dev/sda", "xfs"}, false, ""},
 	{"disk-smart", "disk SMART health read", "/usr/sbin/smartctl", []string{"-a", "-j", "/dev/sda"}, false, ""},
 	{"disk-lvs", "thin-pool usage read", "/usr/sbin/lvs", []string{"--reportformat", "json", "--units", "b", "-o", "lv_name,data_percent,metadata_percent", "--", "pve/data"}, false, ""},
+	{"thin-orphan-check", "orphan thin device check after a failed whole-guest backup (read-only, R-1012)", "/usr/local/sbin/felhom-thin-check", []string{"pve", "data"}, false, ""},
 
 	// ---- Storage mount units (watchdog re-mount) ----
 	{"mount-unit-install", "fs-UUID mount unit install (root content check, R-861)", "/usr/local/sbin/felhom-priv-apply", []string{"unit", "mnt-felhom\\x2dx.mount"}, false, ""},
+	{"mount-unit-forget", "decommissioned-drive unit removal (root check, R-1124)", "/usr/local/sbin/felhom-priv-apply", []string{"unit-forget", "mnt-felhom\\x2dx.mount"}, false, ""},
 	{"mount-daemon-reload", "systemd reload after unit write", "/usr/bin/systemctl", []string{"daemon-reload"}, false, ""},
 	{"mount-unit-enable", "mount unit enable", "/usr/bin/systemctl", []string{"enable", "--now", "--", "mnt-felhom\\x2dx.mount"}, false, ""},
 	{"mount-unit-disable", "mount unit disable", "/usr/bin/systemctl", []string{"disable", "--", "mnt-felhom\\x2dx.mount"}, false, ""},
@@ -117,6 +119,9 @@ var manifest = []Capability{
 	{"guest-reboot", "enroll activate-binds reboot", "/usr/sbin/pct", []string{"reboot", "9201"}, false, ""},
 	{"guest-shutdown", "household power-off: guest shutdown (R-998)", "/usr/sbin/pct", []string{"shutdown", "9201", "--timeout", "120"}, false, ""},
 	{"host-poweroff", "household power-off: host power-off (R-998)", "/usr/bin/systemctl", []string{"poweroff"}, false, ""},
+	{"vzdump-partial-clean", "dead half-written vzdump removal (R-1006)", "/usr/local/sbin/felhom-vzdump-partial-clean", []string{"local", "9201", "vzdump-lxc-9201-2026_10_14-04_42_18.tar.dat"}, false, ""},
+	{"internal-place-plan", "internal data place: the size plan (16 §7)", "/usr/local/sbin/felhom-internal-place", []string{"plan", "0"}, false, ""},
+	{"internal-place-create", "internal data place: create on an old box (signed add_internal_place)", "/usr/local/sbin/felhom-internal-place", []string{"create", "0"}, false, ""},
 
 	// ---- LAN split-horizon resolver (dnsmasq) ----
 	{"dnsmasq-install", "dnsmasq package install", "/usr/bin/apt-get", []string{"install", "-y", "-q", "dnsmasq"}, false, ""},

internal/capability/probe.go

@@ -47,7 +47,7 @@ type Runner interface {
 // downgrades a HEALTHY probe to StatusInactive (broken plumbing stays degraded regardless).
 type Prober struct {
 	Runner     Runner
-	Exists     func(path string) bool  // nil → os.Stat
+	Exists     func(path string) bool // nil → os.Stat
 	GateActive func(gate string) bool // nil → every gate treated active
 }
 

internal/capability/r861_injection_test.go

@@ -63,6 +63,20 @@ var r861Injections = []string{
 	"/usr/bin/systemctl reboot",
 	"/usr/bin/systemctl halt",
 	"/usr/bin/systemctl kexec",
+	// R-1006: the partial-cleanup helper takes exactly one storage, one vmid, one half-written name.
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201 vzdump-lxc-9201-2026_10_14-04_42_18.tar.zst",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201 vzdump-lxc-9201-2026_10_14-04_42_18.log",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201 ../../etc/vzdump-lxc-9201-2026_10_14-04_42_18.tar.dat",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201 vzdump-lxc-9201-2026_10_14-04_42_18.tar.dat /etc/shadow",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201 vzdump-lxc-9201-2026_10_14-04_42_18.tar.dat;id",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local;id 9201 vzdump-lxc-9201-2026_10_14-04_42_18.tar.dat",
+	"/usr/local/sbin/felhom-vzdump-partial-clean local 9201",
+	"/usr/local/sbin/felhom-vzdump-partial-clean --help",
+	"/usr/local/sbin/felhom-internal-place create 132",
+	"/usr/local/sbin/felhom-internal-place create 0 extra",
+	"/usr/local/sbin/felhom-internal-place create 0;id",
+	"/usr/local/sbin/felhom-internal-place wipe 0",
+	"/usr/local/sbin/felhom-internal-place create",
 }
 
 func TestSudoersRefusesTheR861Injections(t *testing.T) {

internal/config/backup_tiers_test.go

@@ -180,3 +180,35 @@ func TestBackupTiers_WaitTimeoutIsPerTier(t *testing.T) {
 		t.Fatalf("wait bounds are shared between tiers — the whole point is that they differ: %+v", tiers)
 	}
 }
+
+// R-1149 (operator decision D17, 2026-10-19): the local whole-guest tier keeps one restore point per day for N days
+// (`--prune-backups keep-daily=N`), not the newest N; an additional tier keeps its explicit count.
+//
+// COMPANION RED-PROOF (observed 2026-10-19): make PruneSpec return keep-last for the primary → this fails with
+// `primary tier prune = "keep-last=3", want "keep-daily=3"`. Restored.
+func TestBackupTierPruneSpec_R1149_LocalKeepsOnePerDay(t *testing.T) {
+	b := BackupConfig{
+		LocalBackupTarget:    "local",
+		LocalBackupRetention: 3,
+		ExtraTargets: []BackupTargetConfig{
+			{TargetID: "felhom-pbs", CadenceSeconds: 604800, KeepLast: 2},
+			{TargetID: "nas", CadenceSeconds: 86400},
+		},
+	}
+	tiers, _ := b.BackupTiers()
+	if len(tiers) != 3 {
+		t.Fatalf("want 3 tiers, got %d", len(tiers))
+	}
+	for _, c := range []struct {
+		i    int
+		want string
+	}{{0, "keep-daily=3"}, {1, "keep-last=2"}, {2, ""}} {
+		if got := tiers[c.i].PruneSpec(); got != c.want {
+			t.Fatalf("%s tier prune = %q, want %q", map[bool]string{true: "primary", false: tiers[c.i].TargetID}[tiers[c.i].Primary], got, c.want)
+		}
+	}
+	// The unset retention defaults to 3 days, never to „prune everything".
+	if got := (BackupConfig{}).PruneBackupsSpec(); got != "keep-daily=3" {
+		t.Fatalf("default local prune = %q, want keep-daily=3", got)
+	}
+}

internal/config/config.go

@@ -391,7 +391,8 @@ type BackupConfig struct {
 	// hub-served per-guest policy is slice 10; this is the agent-local cadence.
 	BackupCadenceSeconds int `json:"backup_cadence_seconds"`
 
-	// LocalBackupRetention is keep-last=N for the per-run `--prune-backups` on a LOCAL vzdump target —
+	// LocalBackupRetention is N for the per-run `--prune-backups keep-daily=N` on a LOCAL vzdump target (keep-last=N
+	// before R-1149) —
 	// so the agent's own local whole-guest backups can't pile up and refill root (the felhom-pve incident;
 	// the host_disk + storage_fill checkers are the detectors, this is the preventive default). 0/unset →
 	// default 3; ALWAYS clamped to ≥1 by KeepLast() so a mis-config can never prune the fresh backup.
@@ -538,9 +539,26 @@ func (b BackupConfig) KeepLast() int {
 	return b.LocalBackupRetention
 }
 
-// PruneBackupsSpec returns the PVE `--prune-backups` value for the local vzdump (e.g. "keep-last=3").
+// PruneBackupsSpec returns the PVE `--prune-backups` value for the local vzdump: "keep-daily=N" (R-1149, operator
+// decision D17 2026-10-19 — one restore point per day for N days; `07` §6.1). keep-last=N had the same disk cost but
+// let N extra runs on one day push out every older point; keep-daily keeps the newest copy of each of the N newest
+// days that have one, so a flood inside one day replaces only that day's point. Pinned by
+// TestBackupTierPruneSpec_R1149_LocalKeepsOnePerDay and, as the consequence, TestBackupCap_R1149_LocalFloodKeepsOlderDays.
 func (b BackupConfig) PruneBackupsSpec() string {
-	return fmt.Sprintf("keep-last=%d", b.KeepLast())
+	return fmt.Sprintf("keep-daily=%d", b.KeepLast())
+}
+
+// PruneSpec is this tier's per-run `--prune-backups` value, or "" for never prune. The PRIMARY (local) tier keeps one
+// point per day (keep-daily, R-1149, from KeepLast() — always ≥1); an additional tier keeps its explicit keep_last
+// count (a weekly off-site tier's „last 2" is two weeks, and PBS prunes stay opt-in per tier, R-82).
+func (t BackupTier) PruneSpec() string {
+	if t.KeepLast <= 0 {
+		return ""
+	}
+	if t.Primary {
+		return fmt.Sprintf("keep-daily=%d", t.KeepLast)
+	}
+	return fmt.Sprintf("keep-last=%d", t.KeepLast)
 }
 
 // BackupCadence returns the per-guest /backup/due window: positive as-is, else 24h default.

internal/escrow/ceremony.go

@@ -1,5 +1,7 @@
 package escrow
 
+import "encoding/json"
+
 // Controller-driven ceremony contract (v0.88.0, TASK 2026-07-13; mechanics validated by
 // felhom.eu/documentation/audits/SPIKE-controller-escrow-2026-07-13.md). The agent's local API
 // re-invokes the agent binary as root via `sudo -n` with ONE fixed argument vector; sudoers
@@ -43,7 +45,15 @@ type CeremonyOutput struct {
 	IdentityBlobBytes int     `json:"identity_blob_bytes"`
 	ResticPwSealed    bool    `json:"restic_pw_sealed"`
 	Uploaded          bool    `json:"uploaded"`
+	// PendingUpload (version 2, R-978) is the hub upload body — the R-wrapped blobs (ciphertext), the key
+	// fingerprint and the sealed password's hash — that json mode hands back INSTEAD of uploading. The agent
+	// daemon uploads it only when the customer claims the new recovery code, so a ceremony that is started
+	// and abandoned never replaces the package the customer's previous code opens. Not a secret without R,
+	// but never logged. Uploaded is false whenever this is set.
+	PendingUpload json.RawMessage `json:"pending_upload,omitempty"`
 }
 
-// CeremonyOutputVersion is the current CeremonyOutput.Version value.
-const CeremonyOutputVersion = 1
+// CeremonyOutputVersion is the current CeremonyOutput.Version value. 2 (R-978): json mode no longer
+// uploads; it returns PendingUpload. The daemon refuses any other version (a binary/daemon mismatch
+// fails the ceremony instead of losing a package).
+const CeremonyOutputVersion = 2

internal/escrow/consume.go

@@ -21,10 +21,10 @@ import (
 // The four inputs (the consumption contract — spike findings):
 //   - blob:                the opaque R-wrapped escrow blob (hub-served restore directive in 10D).
 //   - recoveryCode (R):    BY HAND from the customer — NEVER hub-sourced (zero-knowledge holds: a
-//                          hub compromise alone cannot decrypt). A secret: never logged/persisted.
+//     hub compromise alone cannot decrypt). A secret: never logged/persisted.
 //   - expectedFingerprint: hub-served — the gate target (F-C4).
 //   - keyDest:             where the PBS restore reads the key (`--keyfile` or the default
-//                          `$XDG_CONFIG_HOME/proxmox-backup/encryption-key.json`).
+//     `$XDG_CONFIG_HOME/proxmox-backup/encryption-key.json`).
 //
 // Order (spike F-C2/F-C3/F-C4/F-C6): Unwrap → fingerprint-gate → install. On ANY failure there is
 // NO partial install (nothing at keyDest) and no key material leaks: the recovered key lives only

internal/guesthook/heal_test.go

@@ -50,8 +50,8 @@ func TestParseConfMounts(t *testing.T) {
 func TestMissingBindSources(t *testing.T) {
 	mounts := ParseConfMounts(sampleConf)
 	present := map[string]bool{
-		"/mnt/felhom-usb/felhom-data":                  true, // usb attached
-		"/var/lib/felhom-agent/guests/9201/bootstrap":  true, // bootstrap always present
+		"/mnt/felhom-usb/felhom-data":                 true, // usb attached
+		"/var/lib/felhom-agent/guests/9201/bootstrap": true, // bootstrap always present
 		// "/mnt/felhom-flash/felhom-data" is ABSENT (drive unplugged)
 	}
 	got := MissingBindSources(mounts, func(p string) bool { return present[p] })

internal/hub/cloudflared.go

@@ -14,6 +14,13 @@ const (
 	TunnelRunning    = "running"     // the cloudflared container runs AND its readiness check says CONNECTED
 	TunnelNotRunning = "not_running" // stopped / exited / absent, or running but NOT connected (Detail says which)
 	TunnelUnknown    = "unknown"     // the probe could not ask (guest down, pct/sudo error, health still starting)
+	// TunnelAbsent (R-1070, agent Unreleased 2026-10-19): the guest answered and has NO cloudflared container — the
+	// controller makes none when no tunnel token is configured (a LAN-only or test box). Not a failure by itself: the
+	// hub never alarms on it, and the OS-update host rule reads it as „not applicable" only when the container was
+	// also absent before the update (HostHealthVerdict). Docker 29 prints „error: no such object" in lower case, so
+	// before this state existed the case-sensitive match fell through to `unknown` — which failed the host rule on
+	// every box without a tunnel (Tester4, 2026-10-15).
+	TunnelAbsent = "absent"
 )
 
 // CloudflaredProber reports the box's tunnel. Injectable so tests use a fake and never exec.
@@ -53,7 +60,7 @@ func (p GuestTunnelProber) Status(ctx context.Context) (string, string) {
 		return TunnelUnknown, "no running customer guest"
 	}
 	worst, wdetail := "", ""
-	rank := map[string]int{TunnelRunning: 0, TunnelUnknown: 1, TunnelNotRunning: 2}
+	rank := map[string]int{TunnelRunning: 0, TunnelAbsent: 1, TunnelUnknown: 2, TunnelNotRunning: 3}
 	for _, v := range vmids {
 		out, errOut, err := p.Runner.Run(ctx, "/usr/sbin/pct", "exec", fmt.Sprint(v), "--", "docker", "inspect", "-f", tunnelInspect, "cloudflared")
 		st, d := ClassifyTunnel(string(out), string(errOut), err)
@@ -71,8 +78,8 @@ func (p GuestTunnelProber) Status(ctx context.Context) (string, string) {
 func ClassifyTunnel(stdout, stderr string, err error) (string, string) {
 	out := strings.TrimSpace(stdout)
 	if err != nil || out == "" {
-		if strings.Contains(stderr, "No such object") || strings.Contains(stderr, "No such container") {
-			return TunnelNotRunning, "no cloudflared container in the guest"
+		if e := strings.ToLower(stderr); strings.Contains(e, "no such object") || strings.Contains(e, "no such container") {
+			return TunnelAbsent, "no cloudflared container in the guest (no tunnel set up on this box)"
 		}
 		return TunnelUnknown, "could not ask the guest: " + firstLine(stderr, err)
 	}

internal/hub/cloudflared_test.go

@@ -20,7 +20,10 @@ func TestClassifyTunnel(t *testing.T) {
 		{"connected", "running|0|healthy\n", "", nil, TunnelRunning, "connected"},
 		{"running but not connected", "running|0|unhealthy\n", "", nil, TunnelNotRunning, "NOT connected"},
 		{"stopped", "exited|137|unhealthy\n", "", nil, TunnelNotRunning, "exit code 137"},
-		{"absent", "", "Error: No such object: cloudflared", errors.New("exit status 1"), TunnelNotRunning, "no cloudflared container"},
+		{"absent", "", "Error: No such object: cloudflared", errors.New("exit status 1"), TunnelAbsent, "no cloudflared container"},
+		// R-1070: Docker 29's wording, measured on demo-hp 2026-10-19 (`docker inspect` of a missing name).
+		{"absent, docker 29", "", "error: no such object: cloudflared", errors.New("exit status 1"), TunnelAbsent, "no tunnel set up"},
+		{"absent, no such container", "", "Error response from daemon: No such container: cloudflared", errors.New("exit status 1"), TunnelAbsent, "no cloudflared container"},
 		{"still starting", "running|0|starting\n", "", nil, TunnelUnknown, "starting"},
 		{"no health check (older controller)", "running|0|none\n", "", nil, TunnelRunning, "connection not checked"},
 		{"guest not running", "", "CT 9201 not running", errors.New("exit status 255"), TunnelUnknown, "could not ask"},

internal/hub/contract_test.go

@@ -39,11 +39,14 @@ func TestHostReport_ContractMatchesGolden(t *testing.T) {
 				State: StorageStateAttached, Reachable: true, ClassHint: "fast",
 				ThinPool: &ThinPoolFill{DataUsedFraction: 0.42},
 				Smart:    SmartSummary{Health: SmartUnknown},
+				// v0.157.0 (`16` §3): the identity keys ride on both rows so the key-set guard sees them.
+				Kind: "system", PhysicalDisks: []string{"serial:ata:SanDisk_X600:182195804614"}, Identity: "proven",
 			},
 			{
 				Name: "usb-backup", Type: StorageTypeUSB, DurableID: "uuid:x",
 				State: StorageStateAttached, Reachable: true, MountPath: "/mnt/usb-backup", TotalBytes: 2000000000000,
 				Smart: SmartSummary{Health: SmartUnknown},
+				Kind:  "disk", PhysicalDisks: []string{"serial:usb:External:000000000000"}, Identity: "unknown", IdentityReason: "usb_bridge",
 			},
 			// A pbs target so the recipe's pbs coord has a storage.cfg row to resolve its namespace
 			// from (R-106). Without one the fixture produces namespace_state=unknown while the golden

internal/hub/loop_logtail_test.go

@@ -77,10 +77,10 @@ func TestLoop_FailedTailPushIsReArmedByNextEnvelope(t *testing.T) {
 		env *ControlEnvelope
 		err error
 	}{
-		{env: &ControlEnvelope{LogTailRequested: true}},          // arm
-		{err: errors.New("hub 5xx")},                             // the carrying push FAILS
-		{env: &ControlEnvelope{LogTailRequested: true}},          // hub still pending → re-arm
-		{env: &ControlEnvelope{}},                                // fulfilled
+		{env: &ControlEnvelope{LogTailRequested: true}}, // arm
+		{err: errors.New("hub 5xx")},                    // the carrying push FAILS
+		{env: &ControlEnvelope{LogTailRequested: true}}, // hub still pending → re-arm
+		{env: &ControlEnvelope{}},                       // fulfilled
 	}}
 	l := tailLoop(rep)
 	ctx := context.Background()

internal/hub/report.go

@@ -359,7 +359,7 @@ type GuestSpec struct {
 
 // Cloudflared is the box's tunnel (R-841, agent v0.141.0): the cloudflared container in the customer guest.
 type Cloudflared struct {
-	Status string `json:"status"`           // running | not_running | unknown (TunnelRunning …)
+	Status string `json:"status"`           // running | not_running | unknown | absent (TunnelRunning …; absent since R-1070)
 	Detail string `json:"detail,omitempty"` // why not_running / unknown, or "connected"
 }
 
@@ -431,6 +431,15 @@ type StorageTarget struct {
 	// metadata fill is null until Phase B's privileged `lvs` read.
 	ThinPool *ThinPoolFill `json:"thin_pool,omitempty"`
 
+	// Physical-disk identity (felhom.eu architecture/16-storage-physical-identity.md §3, agent v0.157.0):
+	// the place's kind, the keys of the whole physical disks under it, and whether the agent can prove
+	// them (proven | unknown + identity_reason vm|network|usb_bridge|weak_wwn|no_serial). Agent-computed
+	// from /sys + udev only; empty when not computed (an absent device, an older agent). Additive.
+	Kind           string   `json:"kind,omitempty"`
+	PhysicalDisks  []string `json:"physical_disks,omitempty"`
+	Identity       string   `json:"identity,omitempty"`
+	IdentityReason string   `json:"identity_reason,omitempty"`
+
 	// Smart is the disk-health summary, populated in Phase B via smartctl (sudoers). In
 	// Phase A it is {health:"UNKNOWN", <counters null>} — the keys are committed now so the
 	// contract is stable across both phases.

internal/hub/testdata/host-report.golden.json

@@ -46,6 +46,9 @@
       "backing_device": "",
       "class_hint": "fast",
       "role": "",
+      "kind": "system",
+      "physical_disks": ["serial:ata:SanDisk_X600:182195804614"],
+      "identity": "proven",
       "thin_pool": { "data_used_fraction": 0.42, "metadata_used_fraction": null },
       "smart": {
         "health": "UNKNOWN",
@@ -74,6 +77,10 @@
       "backing_device": "/dev/sdb1",
       "class_hint": "slow",
       "role": "",
+      "kind": "disk",
+      "physical_disks": ["serial:usb:External:000000000000"],
+      "identity": "unknown",
+      "identity_reason": "usb_bridge",
       "smart": {
         "health": "UNKNOWN",
         "temperature_c": null,

internal/localapi/backup_cap.go

@@ -0,0 +1,146 @@
+package localapi
+
+import (
+	"fmt"
+	"net/http"
+	"sort"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
+)
+
+// The whole-guest backup cadence cap (R-973, security review 2026-10-12).
+//
+// POST /backup is reachable from inside the guest, and every successful run prunes the tier
+// (`--prune-backups keep-last=N` on a local tier; ep0's server-side keep-last 2 on the off-site tier). So
+// a compromised controller calling it in a loop pushed every pre-compromise whole-guest restore point out
+// within minutes — the threat `07` §8 row 10 counts the off-site tier as the answer to. The agent now
+// refuses a run that the schedule in `07` §6 would never ask for:
+//
+//   - a tier whose cadence is at most 24 h (the local tier): at most 3 successful runs in any rolling
+//     24 h. The schedule is one night run; a „Mentés most" press makes a LOCAL copy and, since R-899, the
+//     night still takes its own after it, which can land inside the same 24 h as the previous night (the
+//     window opens anywhere in [W+2h, W+6h)). So a normal day needs at most 3: two nights plus one press.
+//     A second press inside 24 h is refused — the margin is spent on the night, which must never be refused
+//     because of the household's presses.
+//   - a longer tier (the 7-day off-site tier): at most 1 successful run per max(cadence − 24 h, cadence/2)
+//     — 6 days for 7 days. The controller asks /backup/due first and runs only when the newest copy is older
+//     than the cadence, and the button never targets this tier (`07` §6, controller v0.301.0), so a
+//     legitimate run never comes sooner; 24 h of slack absorbs clock and archive-time differences.
+//
+// Only SUCCESSES count — a failed run prunes nothing and must stay retryable. The count survives an agent
+// restart: it reads the success history saved on disk (LastKnownBackupStore, the R-894 file) as well as
+// this process's own record. A refusal answers 429, is logged at WARN, is saved, and raises the operator
+// signal BackupCapRefusal feeds (a critical capability in the host report → the hub's operator-only
+// `whole_guest_backup_refused` event since hub v0.159.0, R-1150). Pinned by TestBackupCap_* (backup_cap_test.go).
+
+// backupCapRefusalHold is how long one refusal keeps the operator signal raised.
+const backupCapRefusalHold = 24 * time.Hour
+
+// backupCadenceCap returns the rolling window and the most successful runs allowed in it for a tier.
+func backupCadenceCap(cadence time.Duration) (window time.Duration, max int) {
+	if cadence <= 0 {
+		cadence = defaultBackupCadence
+	}
+	if cadence <= 24*time.Hour {
+		return 24 * time.Hour, 3
+	}
+	w := cadence - 24*time.Hour
+	if w < cadence/2 {
+		w = cadence / 2
+	}
+	return w, 1
+}
+
+// recentCapRuns returns the start times of the successful runs on this tier for this guest inside the
+// window ending now — this process's record and the saved one, each counted once. Caller holds no lock.
+func (s *Server) recentCapRuns(key backupJobKey, since time.Time) []time.Time {
+	seen := map[int64]bool{}
+	var out []time.Time
+	add := func(t time.Time) {
+		if t.Before(since) || seen[t.Unix()] {
+			return
+		}
+		seen[t.Unix()] = true
+		out = append(out, t)
+	}
+	s.capMu.Lock()
+	for _, t := range s.capRuns[key] {
+		add(t)
+	}
+	s.capMu.Unlock()
+	if s.lastKnown != nil {
+		for _, t := range s.lastKnown.RecentBackupSuccesses(key.target, key.vmid, since) {
+			add(t)
+		}
+	}
+	sort.Slice(out, func(i, j int) bool { return out[i].Before(out[j]) })
+	return out
+}
+
+// recordCapRun notes a successful run for the cap (the saved half is written by RecordBackupSuccess).
+func (s *Server) recordCapRun(key backupJobKey, b hub.Backup) {
+	t, err := time.Parse(time.RFC3339, b.StartedAt)
+	if err != nil {
+		t = s.now()
+	}
+	s.capMu.Lock()
+	defer s.capMu.Unlock()
+	if s.capRuns == nil {
+		s.capRuns = map[backupJobKey][]time.Time{}
+	}
+	runs := append(s.capRuns[key], t.UTC())
+	if len(runs) > 8 {
+		runs = runs[len(runs)-8:]
+	}
+	s.capRuns[key] = runs
+}
+
+// capRefuses applies the cap to a new run request. It writes the 429 and returns true when refused.
+func (s *Server) capRefuses(w http.ResponseWriter, key backupJobKey, tier BackupTier) bool {
+	window, max := backupCadenceCap(tier.Cadence)
+	now := s.now()
+	runs := s.recentCapRuns(key, now.Add(-window))
+	if len(runs) < max {
+		return false
+	}
+	next := runs[len(runs)-max].Add(window)
+	s.capMu.Lock()
+	s.capRefusal = capRefusal{target: key.target, vmid: key.vmid, at: now.UTC()}
+	s.capMu.Unlock()
+	if s.lastKnown != nil {
+		if err := s.lastKnown.RecordBackupRefusal(key.target, key.vmid, now); err != nil {
+			s.logger.Warn("local-api: could not save the backup refusal on disk (R-973)", "vmid", key.vmid, "target", key.target, "err", err)
+		}
+	}
+	s.logger.Warn("local-api: backup REFUSED — more whole-guest backups than the schedule allows (R-973 cadence cap; check the guest)",
+		"vmid", key.vmid, "target", key.target, "runs_in_window", len(runs), "allowed", max,
+		"window_h", int(window.Hours()), "next_allowed", next.UTC().Format(time.RFC3339))
+	writeStatus(w, http.StatusTooManyRequests, false, nil, fmt.Sprintf(
+		"too many whole-guest backups on target %q: %d successful in the last %d h, the schedule allows %d — refused (the next one is allowed after %s)",
+		key.target, len(runs), int(window.Hours()), max, next.UTC().Format(time.RFC3339)))
+	return true
+}
+
+type capRefusal struct {
+	target string
+	vmid   int
+	at     time.Time
+}
+
+// BackupCapRefusal reports the newest cadence-cap refusal still inside the 24 h hold (this process's or
+// the saved one) — the source of the operator signal in the host report (R-973). ok=false = none.
+func (s *Server) BackupCapRefusal() (target string, vmid int, at time.Time, ok bool) {
+	s.capMu.Lock()
+	r := s.capRefusal
+	s.capMu.Unlock()
+	if s.lastKnown != nil {
+		if t, v, a, found := s.lastKnown.LastBackupRefusal(); found && a.After(r.at) {
+			r = capRefusal{target: t, vmid: v, at: a}
+		}
+	}
+	if r.at.IsZero() || s.now().Sub(r.at) > backupCapRefusalHold {
+		return "", 0, time.Time{}, false
+	}
+	return r.target, r.vmid, r.at, true
+}

internal/localapi/backup_cap_test.go

@@ -0,0 +1,377 @@
+package localapi
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"log/slog"
+	"net/http"
+	"path/filepath"
+	"sync"
+	"testing"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/backup"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/config"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
+)
+
+// R-973 — the whole-guest backup cadence cap. A compromised controller calling POST /backup in a loop
+// pushed out every pre-compromise restore point: each success prunes the tier. These tests model the tier
+// the way Proxmox and ep0 keep it (keep-last N, pruned after each success) and ask what is LEFT.
+
+// keepLastTier is a whole-guest tier: every successful run adds an archive stamped with the clock and
+// prunes to keepLast, as vzdump's --prune-backups keep-last (local) and ep0's keep-last 2 (off-site) do.
+type keepLastTier struct {
+	mu       sync.Mutex
+	clock    *testClock
+	keepLast int
+	// spec, when set, is the `--prune-backups` value the agent passes (R-1149) and the tier prunes the way Proxmox
+	// reads it: keep-last=N → the newest N; keep-daily=N → the newest copy of each of the N newest days with one.
+	spec     string
+	t        *testing.T
+	archives []time.Time
+	fail     bool
+	runs     int
+}
+
+// pruneBySpec applies a Proxmox `--prune-backups` spec to archives (oldest first) and returns what is kept.
+func pruneBySpec(t *testing.T, archives []time.Time, spec string) []time.Time {
+	var n int
+	var mode string
+	if _, err := fmt.Sscanf(spec, "keep-last=%d", &n); err == nil {
+		mode = "last"
+	} else if _, err := fmt.Sscanf(spec, "keep-daily=%d", &n); err == nil {
+		mode = "daily"
+	} else {
+		t.Fatalf("the model does not know the prune spec %q", spec)
+	}
+	var kept []time.Time
+	days := map[string]bool{}
+	for i := len(archives) - 1; i >= 0; i-- { // newest first, as Proxmox walks them
+		a := archives[i]
+		switch mode {
+		case "last":
+			if len(kept) < n {
+				kept = append([]time.Time{a}, kept...)
+			}
+		case "daily":
+			d := a.Format("2006-01-02")
+			if !days[d] && len(days) < n {
+				days[d] = true
+				kept = append([]time.Time{a}, kept...)
+			}
+		}
+	}
+	return kept
+}
+
+func (k *keepLastTier) BackupWithSnapshotHook(_ context.Context, vmid int, _ func()) (hub.Backup, error) {
+	k.mu.Lock()
+	defer k.mu.Unlock()
+	k.runs++
+	now := k.clock.Now()
+	if k.fail {
+		return hub.Backup{VMID: vmid, StartedAt: now.Format(time.RFC3339)}, fmt.Errorf("vzdump failed")
+	}
+	k.archives = append(k.archives, now)
+	if k.spec != "" {
+		k.archives = pruneBySpec(k.t, k.archives, k.spec)
+	} else if len(k.archives) > k.keepLast {
+		k.archives = k.archives[len(k.archives)-k.keepLast:]
+	}
+	return hub.Backup{VMID: vmid, Success: true, Archive: "x", StartedAt: now.Format(time.RFC3339)}, nil
+}
+
+// oldestBefore reports whether the tier still holds an archive made before t (a pre-compromise point).
+func (k *keepLastTier) oldestBefore(t time.Time) bool {
+	k.mu.Lock()
+	defer k.mu.Unlock()
+	return len(k.archives) > 0 && k.archives[0].Before(t)
+}
+
+func (k *keepLastTier) count() (runs, kept int) {
+	k.mu.Lock()
+	defer k.mu.Unlock()
+	return k.runs, len(k.archives)
+}
+
+type testClock struct {
+	mu sync.Mutex
+	t  time.Time
+}
+
+func (c *testClock) Now() time.Time          { c.mu.Lock(); defer c.mu.Unlock(); return c.t }
+func (c *testClock) Advance(d time.Duration) { c.mu.Lock(); c.t = c.t.Add(d); c.mu.Unlock() }
+
+// capServer builds the two-tier box of `07` §6: local 24 h keep-last 3, off-site 7 d keep-last 2, the
+// success history on disk at path (a second capServer on the same path is an agent restart).
+func capServer(t *testing.T, path string, clock *testClock, local, offsite *keepLastTier) *Server {
+	t.Helper()
+	srv, err := NewServer(Options{
+		ListenAddr: "127.0.0.1:0", Guests: &fakeGuests{}, Backups: &fakeBackups{}, Store: &fakeStore{},
+		Storage: fakeStorage{targets: []hub.StorageTarget{{Name: "local"}, {Name: "felhom-pbs"}}},
+		Tokens:  staticTokens{"A": 8200},
+		BackupTiers: []BackupTier{
+			{TargetID: "local", Cadence: 24 * time.Hour, Primary: true, Service: local},
+			{TargetID: "felhom-pbs", Cadence: 7 * 24 * time.Hour, Service: offsite},
+		},
+		LastKnownBackups: backup.NewBackupSuccessState(path),
+		Logger:           slog.New(slog.NewTextHandler(io.Discard, nil)),
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	srv.baseCtx = context.Background()
+	srv.now = clock.Now
+	return srv
+}
+
+// runBackup POSTs one backup and waits for the job to end. Returns the HTTP status of the POST.
+func runBackup(t *testing.T, srv *Server, target string) int {
+	t.Helper()
+	rr := do(t, srv.Handler(), "POST", "/backup?target="+target, "A", "")
+	if rr.Code == http.StatusAccepted {
+		waitFor(t, func() bool {
+			srv.jobsMu.Lock()
+			defer srv.jobsMu.Unlock()
+			j := srv.jobs[backupJobKey{vmid: 8200, target: target}]
+			return j != nil && !backupInFlight(j.Phase)
+		})
+	}
+	return rr.Code
+}
+
+// The off-site tier — the one `07` §8 row 10 counts on against ransomware. Before the cap ten calls in a
+// row left ep0 holding only the attacker's two copies. Now ONE runs (the tier was due), the other nine
+// are refused, and the newest pre-compromise copy survives; five days of flooding later it still does.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): make capRefuses always return false → this fails with
+// "the off-site tier lost every pre-compromise copy: 10 runs" . Restored.
+func TestBackupCap_OffsiteFloodKeepsAPreCompromiseCopy(t *testing.T) {
+	clock := &testClock{t: testNow}
+	local := &keepLastTier{clock: clock, keepLast: 3}
+	off := &keepLastTier{clock: clock, keepLast: 2}
+	path := filepath.Join(t.TempDir(), "backup-success-state.json")
+	srv := capServer(t, path, clock, local, off)
+
+	// The honest history: two weekly copies, the newest 7 days and 1 hour old (so the tier is due).
+	for _, age := range []time.Duration{14*24*time.Hour + time.Hour, 7*24*time.Hour + time.Hour} {
+		clock.t = testNow.Add(-age)
+		if code := runBackup(t, srv, "felhom-pbs"); code != http.StatusAccepted {
+			t.Fatalf("seeding the honest weekly copy: %d", code)
+		}
+	}
+	clock.t = testNow
+	compromise := testNow
+
+	refused := 0
+	for i := 0; i < 10; i++ {
+		if runBackup(t, srv, "felhom-pbs") == http.StatusTooManyRequests {
+			refused++
+		}
+		clock.Advance(time.Minute)
+	}
+	runs, _ := off.count()
+	if !off.oldestBefore(compromise) {
+		t.Fatalf("the off-site tier lost every pre-compromise copy: %d runs", runs-2)
+	}
+	if runs-2 != 1 || refused != 9 {
+		t.Fatalf("flood: %d runs, %d refused — want 1 run (the tier was due) and 9 refused", runs-2, refused)
+	}
+	for d := 0; d < 5; d++ { // five more days of flooding, every hour
+		for h := 0; h < 24; h++ {
+			clock.Advance(time.Hour)
+			if code := runBackup(t, srv, "felhom-pbs"); code != http.StatusTooManyRequests {
+				t.Fatalf("day %d hour %d: an off-site run was allowed %v after the last one", d, h, clock.Now().Sub(compromise))
+			}
+		}
+	}
+	if !off.oldestBefore(compromise) {
+		t.Fatal("five days of flooding pushed out the pre-compromise off-site copy")
+	}
+}
+
+// The local tier: a flood runs exactly 3 in a day (the schedule's own day: two nights + one press) and
+// every further request is refused. Honest about the limit: with keep-last 3 those 3 replace the 3 kept
+// points — the cap slows a flush from minutes to a day; it cannot keep a point a 3-per-day pace prunes.
+func TestBackupCap_LocalFloodRunsAtMostThreeADay(t *testing.T) {
+	clock := &testClock{t: testNow}
+	local := &keepLastTier{clock: clock, keepLast: 3}
+	srv := capServer(t, filepath.Join(t.TempDir(), "s.json"), clock, local, &keepLastTier{clock: clock, keepLast: 2})
+	refused := 0
+	for i := 0; i < 20; i++ {
+		if runBackup(t, srv, "local") == http.StatusTooManyRequests {
+			refused++
+		}
+		clock.Advance(10 * time.Minute)
+	}
+	if runs, _ := local.count(); runs != 3 || refused != 17 {
+		t.Fatalf("local flood over ~3 h: %d runs, %d refused — want 3 and 17", runs, refused)
+	}
+	clock.Advance(24 * time.Hour) // a day later the window has moved on
+	if code := runBackup(t, srv, "local"); code != http.StatusAccepted {
+		t.Fatalf("a run a day later must be allowed: %d", code)
+	}
+}
+
+// The schedule itself is never refused: a night at W+6h, a press at 10:00, the next night at W+2h (all
+// three inside one rolling 24 h) run; a second press that day is refused; nights keep running after it.
+func TestBackupCap_TheScheduledDayIsNeverRefused(t *testing.T) {
+	day := time.Date(2026, 6, 10, 0, 0, 0, 0, time.UTC)
+	clock := &testClock{t: day.Add(8*time.Hour + 29*time.Minute)} // W=02:30 → W+6h ≈ 08:29
+	local := &keepLastTier{clock: clock, keepLast: 3}
+	srv := capServer(t, filepath.Join(t.TempDir(), "s.json"), clock, local, &keepLastTier{clock: clock, keepLast: 2})
+	steps := []struct {
+		at   time.Time
+		want int
+		what string
+	}{
+		{day.Add(8*time.Hour + 29*time.Minute), http.StatusAccepted, "night 1 at W+6h"},
+		{day.Add(10 * time.Hour), http.StatusAccepted, "a press at 10:00"},
+		{day.Add(24*time.Hour + 4*time.Hour + 30*time.Minute), http.StatusAccepted, "night 2 at W+2h (20 h later)"},
+		{day.Add(24*time.Hour + 7*time.Hour), http.StatusTooManyRequests, "a second press inside the same 24 h"},
+		{day.Add(48*time.Hour + 4*time.Hour + 30*time.Minute), http.StatusAccepted, "night 3"},
+	}
+	for _, s := range steps {
+		clock.t = s.at
+		if got := runBackup(t, srv, "local"); got != s.want {
+			t.Fatalf("%s: got %d, want %d", s.what, got, s.want)
+		}
+	}
+}
+
+// A restart must not reset the cap: the history is read back from disk.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): make recentCapRuns ignore s.lastKnown → this fails with
+// "after a restart the 4th run in a day was allowed (got 202)". Restored.
+func TestBackupCap_SurvivesARestart(t *testing.T) {
+	clock := &testClock{t: testNow}
+	path := filepath.Join(t.TempDir(), "s.json")
+	local := &keepLastTier{clock: clock, keepLast: 3}
+	srv := capServer(t, path, clock, local, &keepLastTier{clock: clock, keepLast: 2})
+	for i := 0; i < 3; i++ {
+		if code := runBackup(t, srv, "local"); code != http.StatusAccepted {
+			t.Fatalf("run %d: %d", i+1, code)
+		}
+		clock.Advance(time.Minute)
+	}
+	srv2 := capServer(t, path, clock, local, &keepLastTier{clock: clock, keepLast: 2}) // the restart
+	if code := runBackup(t, srv2, "local"); code != http.StatusTooManyRequests {
+		t.Fatalf("after a restart the 4th run in a day was allowed (got %d)", code)
+	}
+}
+
+// Failed runs prune nothing and must stay retryable: they do not count.
+func TestBackupCap_FailuresDoNotCount(t *testing.T) {
+	clock := &testClock{t: testNow}
+	local := &keepLastTier{clock: clock, keepLast: 3, fail: true}
+	srv := capServer(t, filepath.Join(t.TempDir(), "s.json"), clock, local, &keepLastTier{clock: clock, keepLast: 2})
+	for i := 0; i < 6; i++ {
+		if code := runBackup(t, srv, "local"); code != http.StatusAccepted {
+			t.Fatalf("failed attempt %d was refused: %d", i+1, code)
+		}
+		clock.Advance(15 * time.Minute)
+	}
+	local.mu.Lock()
+	local.fail = false
+	local.mu.Unlock()
+	if code := runBackup(t, srv, "local"); code != http.StatusAccepted {
+		t.Fatalf("a success after six failures was refused: %d", code)
+	}
+}
+
+// A refusal raises the operator signal for 24 h, and the signal survives a restart.
+func TestBackupCap_RefusalRaisesTheOperatorSignal(t *testing.T) {
+	clock := &testClock{t: testNow}
+	path := filepath.Join(t.TempDir(), "s.json")
+	off := &keepLastTier{clock: clock, keepLast: 2}
+	srv := capServer(t, path, clock, &keepLastTier{clock: clock, keepLast: 3}, off)
+	if _, _, _, ok := srv.BackupCapRefusal(); ok {
+		t.Fatal("a fresh box must not raise the signal")
+	}
+	runBackup(t, srv, "felhom-pbs")
+	clock.Advance(time.Hour)
+	if code := runBackup(t, srv, "felhom-pbs"); code != http.StatusTooManyRequests {
+		t.Fatalf("second off-site run an hour later: %d, want 429", code)
+	}
+	target, vmid, at, ok := srv.BackupCapRefusal()
+	if !ok || target != "felhom-pbs" || vmid != 8200 || !at.Equal(clock.Now()) {
+		t.Fatalf("signal after a refusal: ok=%v target=%q vmid=%d at=%v", ok, target, vmid, at)
+	}
+	clock.Advance(23 * time.Hour)
+	srv2 := capServer(t, path, clock, &keepLastTier{clock: clock, keepLast: 3}, off) // a restart
+	if _, _, _, ok := srv2.BackupCapRefusal(); !ok {
+		t.Fatal("the signal did not survive a restart")
+	}
+	clock.Advance(2 * time.Hour)
+	if _, _, _, ok := srv2.BackupCapRefusal(); ok {
+		t.Fatal("the signal must clear 24 h after the last refusal")
+	}
+}
+
+func TestBackupCadenceCap(t *testing.T) {
+	cases := []struct {
+		cadence time.Duration
+		window  time.Duration
+		max     int
+	}{
+		{0, 24 * time.Hour, 3},
+		{24 * time.Hour, 24 * time.Hour, 3},
+		{12 * time.Hour, 24 * time.Hour, 3},
+		{7 * 24 * time.Hour, 6 * 24 * time.Hour, 1},
+		{36 * time.Hour, 18 * time.Hour, 1},
+		{30 * 24 * time.Hour, 29 * 24 * time.Hour, 1},
+	}
+	for _, c := range cases {
+		if w, m := backupCadenceCap(c.cadence); w != c.window || m != c.max {
+			t.Errorf("cadence %v: got %v/%d, want %v/%d", c.cadence, w, m, c.window, c.max)
+		}
+	}
+}
+
+// R-1149 — the consequence of keep-daily on the local tier, with the agent's REAL prune spec (config.BackupTiers →
+// PruneSpec, what runDaemon hands the runner). Three honest nights, then a compromise at 01:00 that floods
+// POST /backup for the rest of the day. The cap lets 3 runs through inside the rolling 24 h (two at once, one when
+// last night's success ages out). With keep-last 3 those three were the only copies left; with keep-daily 3 the two
+// earlier nights survive the whole day.
+//
+// COMPANION RED-PROOF (observed 2026-10-19): make BackupTier.PruneSpec return keep-last for the primary → this fails
+// with "after a day of flooding the local tier holds 0 copies made before the compromise (kept […]) — want the 2
+// earlier nights". Restored.
+func TestBackupCap_R1149_LocalFloodKeepsOlderDays(t *testing.T) {
+	tiers, _ := config.BackupConfig{LocalBackupTarget: "local", LocalBackupRetention: 3}.BackupTiers()
+	spec := tiers[0].PruneSpec()
+	day := time.Date(2026, 6, 10, 0, 0, 0, 0, time.UTC)
+	clock := &testClock{t: day}
+	local := &keepLastTier{clock: clock, keepLast: 3, spec: spec, t: t}
+	srv := capServer(t, filepath.Join(t.TempDir(), "s.json"), clock, local, &keepLastTier{clock: clock, keepLast: 2})
+	for d := 3; d >= 1; d-- { // the nights of D-3, D-2, D-1 at 03:00
+		clock.t = day.Add(-time.Duration(d)*24*time.Hour + 3*time.Hour)
+		if code := runBackup(t, srv, "local"); code != http.StatusAccepted {
+			t.Fatalf("seeding night D-%d: %d", d, code)
+		}
+	}
+	compromise := day.Add(time.Hour)
+	ran := 0
+	for clock.t = compromise; clock.t.Before(day.Add(24 * time.Hour)); clock.Advance(10 * time.Minute) {
+		if runBackup(t, srv, "local") == http.StatusAccepted {
+			ran++
+		}
+	}
+	if ran != 3 {
+		t.Fatalf("the flood ran %d backups on day D — want 3 (the cap)", ran)
+	}
+	local.mu.Lock()
+	kept := append([]time.Time(nil), local.archives...)
+	local.mu.Unlock()
+	pre := 0
+	for _, a := range kept {
+		if a.Before(compromise) {
+			pre++
+		}
+	}
+	if pre != 2 {
+		t.Fatalf("after a day of flooding the local tier holds %d copies made before the compromise (kept %v) — want the 2 earlier nights", pre, kept)
+	}
+}

internal/localapi/controllersupervisor_test.go

@@ -19,10 +19,10 @@ import (
 // again", and each guard is pinned by the case where acting would be wrong.
 
 type supExec struct {
-	mu        sync.Mutex
-	status    map[int]string // docker .State.Status per vmid; "" = container absent
-	inspectErr error         // non-nil = pct exec itself failed (unknown)
-	restarts  map[int]int
+	mu         sync.Mutex
+	status     map[int]string // docker .State.Status per vmid; "" = container absent
+	inspectErr error          // non-nil = pct exec itself failed (unknown)
+	restarts   map[int]int
 	// onRestart, when set, is the status the container reaches after a restart (a crash-looper
 	// stays "exited").
 	onRestart string
@@ -249,7 +249,6 @@ func TestControllerSupervisorStanza_WireShape(t *testing.T) {
 	}
 }
 
-
 // ---- R-539 (operator ruling 3 of 2026-09-16): the SLOW crash loop ----------------------------------
 
 // supKillOnce kills the controller and lets the supervisor restart it (two confirming sweeps), then

internal/localapi/disks.go

@@ -188,6 +188,13 @@ type DiskInfo struct {
 	// device that exposes no SMART (USB bridge, unread) is ABSENT, not a misleading zero-value UNKNOWN;
 	// the controller feature-detects presence and renders "Nincs adat" when nil (never alarms).
 	Smart *hub.SmartSummary `json:"smart,omitempty"`
+	// Physical-disk identity (v0.157.0, felhom.eu architecture/16-storage-physical-identity.md §3) — the
+	// same agent-computed verdict the hub report carries. Empty when not computed (an absent drive, no
+	// resolver). Additive; an older controller ignores it.
+	Kind           string   `json:"kind,omitempty"`
+	PhysicalDisks  []string `json:"physical_disks,omitempty"`
+	Identity       string   `json:"identity,omitempty"`
+	IdentityReason string   `json:"identity_reason,omitempty"`
 }
 
 // handleDisks lists the host's drives + data-bearing flags (read-only/benign).
@@ -227,6 +234,8 @@ func (s *Server) handleDisks(w http.ResponseWriter, r *http.Request, vmid int) {
 			// really is the target. Reported here so the controller can name the drive in an
 			// absent-target alarm and hide the destructive controls the agent would refuse anyway.
 			BackupTarget: t.Name == primaryTargetID,
+			// `16` §3: copied from the observer, which computed it from /sys + udev.
+			Kind: t.Kind, PhysicalDisks: t.PhysicalDisks, Identity: t.Identity, IdentityReason: t.IdentityReason,
 		}
 		// Intermediary model: the stable in-guest path + whether felhom-data is bound under the parent.
 		// Only user-data /mnt/<name> drives have a guest path (system/backup mounts never cross in).
@@ -398,6 +407,10 @@ func (s *Server) handleDisks(w http.ResponseWriter, r *http.Request, vmid int) {
 						di.Smart = &sm
 					}
 				}
+				// `16` §3: a registry drive's identity, judged on its OWN mount only (never root's).
+				if p, kind, ok := s.ownMountIdentity(d.MountPath); ok {
+					di.Kind, di.PhysicalDisks, di.Identity, di.IdentityReason = kind, p.PhysicalDisks, p.Identity, p.Reason
+				}
 				// R-118: statfs ONLY while the drive's device is present. With the device gone the raw
 				// mountpoint reverts to a bare directory on the ROOT filesystem, and statfs would report
 				// pve-root's size as this drive's (measured: a 4 GB drive advertising 46 GiB). Same trap

internal/localapi/escrow_ceremony.go

@@ -3,11 +3,15 @@ package localapi
 import (
 	"bytes"
 	"context"
+	"crypto/rand"
+	"crypto/subtle"
 	"encoding/json"
 	"fmt"
+	"math/big"
 	"net/http"
 	"os/exec"
 	"strconv"
+	"strings"
 	"time"
 
 	"gitea.dooplex.hu/admin/felhom-agent/internal/escrow"
@@ -24,9 +28,23 @@ import (
 //     be structurally incapable of carrying R).
 //   - One claim, then the holder is zeroed. Unclaimed past the TTL → zeroed + phase
 //     unclaimed_void ("R unclaimed → ceremony void → a re-run supersedes").
-//   - Crash-safety is IN-MEMORY BY DESIGN: an agent restart loses R, which is SAFE (the blob is
-//     on the hub; a re-run supersedes it). Status reporting "none" after a restart is the honest
-//     answer. No journal, deliberately — persistence is the one property R must never have.
+//   - Crash-safety is IN-MEMORY BY DESIGN: an agent restart loses R AND the package it opens,
+//     which is SAFE (R-978: the package was never uploaded, so the hub still holds the previous one
+//     and the customer's previous code still opens it; a re-run starts over). Status reporting "none"
+//     after a restart is the honest answer. No journal, deliberately — persistence is the one
+//     property R must never have.
+//   - R-978: the NEW package reaches the hub only at the claim, and the code is handed out only after
+//     the hub took it. Until then the previous package stays current — so a ceremony that is started
+//     and abandoned (never claimed, superseded, TTL, restart) leaves the customer's written-down code
+//     working. The TTL and a new start both wait while a claim's upload is in flight, so a package is
+//     never stored without its code being handed out by that same claim.
+//   - R-1151: a claim that asks for it (`{"defer_upload": true}`, controller ≥ 0.332.0) hands out the code
+//     and two word positions, and keeps the package until POST /escrow/ceremony/confirm brings those two
+//     words back — the household's typed-back check, now checked HERE, not only in the browser. A code that
+//     was shown and never typed back (the page closed) never replaces the hub's package; the confirm window
+//     (escrowConfirmTTL) ends it. Only the two words are kept for the check, never R. A claim WITHOUT the
+//     flag (a controller that has not updated) keeps the R-978 order: upload at the claim. Pinned by
+//     TestEscrowCeremony_R1151_*.
 //   - Nothing from stdout is ever logged; stderr (log-clean, spike §2.1) is tail-captured for
 //     failure diagnostics only.
 
@@ -50,8 +68,16 @@ type EscrowCeremonyConfig struct {
 	// DRConfigured answers "is the DR tier applied on this box?" (pbsdr.Manager.DRConfigured,
 	// late-bound). nil → reported not-applied.
 	DRConfigured func() bool
+	// Upload sends a ceremony's package (CeremonyOutput.PendingUpload, the hub upload body) to this
+	// host's hub record — called by the claim, BEFORE the code is handed out (R-978). nil → a claim of
+	// a pending package is refused (503): a code must never be handed out for a package the hub lacks.
+	Upload func(ctx context.Context, pkg []byte) error
 }
 
+// escrowUploadTimeout bounds the claim-time package upload (one small PUT; the subprocess's whole
+// ceremony incl. its upload was measured at ~2.4 s).
+const escrowUploadTimeout = 30 * time.Second
+
 // ceremonyRunner executes the fixed-argv sudo self-invocation. stdout is SECRET-BEARING until
 // parsed (it carries R inside the JSON object); the caller must zero it. Tests inject canned
 // spike-shaped output; production is runCeremonySubprocess.
@@ -78,8 +104,15 @@ const (
 	escrowPhaseDone    = "done"
 	escrowPhaseFailed  = "failed"
 	escrowPhaseVoid    = "unclaimed_void"
+	// escrowPhaseUnconfirmed (R-1151): the code was shown but its words never came back inside the confirm
+	// window — the package was dropped, the hub's previous package (and the previous code) stays current.
+	escrowPhaseUnconfirmed = "unconfirmed_void"
 )
 
+// escrowConfirmTTL is how long a claimed code waits for its typed-back words (R-1151). Writing ten words down
+// by hand takes minutes; past this the package is dropped and the household starts again.
+const escrowConfirmTTL = 30 * time.Minute
+
 // escrowCeremonyTimeout bounds the whole subprocess run. Spike-measured ceremony ≈ 2.4 s incl.
 // upload — 60 s is a ≥25× margin that still absorbs WAN upload latency.
 const escrowCeremonyTimeout = 60 * time.Second
@@ -129,6 +162,9 @@ func (s *Server) tryStartEscrowCeremony(job *escrowCeremonyJob) bool {
 	if s.escrowJob != nil && s.escrowJob.Phase == escrowPhaseRunning {
 		return false
 	}
+	if s.escrowClaiming {
+		return false // R-978: a claim is uploading this job's package — its code must reach the claimer
+	}
 	s.wipeEscrowRLocked()
 	cp := *job
 	s.escrowJob = &cp
@@ -137,7 +173,7 @@ func (s *Server) tryStartEscrowCeremony(job *escrowCeremonyJob) bool {
 
 // finishEscrowCeremony records the terminal phase and (on success) arms the one-shot R holder.
 // R is stored as a byte slice so the claim/TTL paths can zero it in place.
-func (s *Server) finishEscrowCeremony(job *escrowCeremonyJob, r string) {
+func (s *Server) finishEscrowCeremony(job *escrowCeremonyJob, r string, pending []byte) {
 	s.escrowMu.Lock()
 	defer s.escrowMu.Unlock()
 	job.UpdatedAt = s.nowFn().UTC().Format(time.RFC3339)
@@ -145,6 +181,7 @@ func (s *Server) finishEscrowCeremony(job *escrowCeremonyJob, r string) {
 	s.escrowJob = &cp
 	if job.Phase == escrowPhaseDone && r != "" {
 		s.escrowR = []byte(r)
+		s.escrowPending = append([]byte(nil), pending...)
 		s.escrowRClaimed = false
 		s.escrowRExpiry = s.nowFn().Add(escrowClaimTTL)
 		// Active TTL belt: zero the holder even if nobody ever polls again. The lazy check in
@@ -164,6 +201,53 @@ func (s *Server) finishEscrowCeremony(job *escrowCeremonyJob, r string) {
 func (s *Server) wipeEscrowRLocked() {
 	zeroBytes(s.escrowR)
 	s.escrowR = nil
+	s.escrowPending = nil // R-978: the package goes with its code — never uploaded once the code is gone
+	s.wipeEscrowConfirmLocked()
+}
+
+// wipeEscrowConfirmLocked drops the R-1151 typed-back state: the two kept words and the wait. Caller holds escrowMu.
+func (s *Server) wipeEscrowConfirmLocked() {
+	for i := range s.escrowConfirmWords {
+		zeroBytes(s.escrowConfirmWords[i])
+		s.escrowConfirmWords[i] = nil
+	}
+	s.escrowAwaitConfirm = false
+}
+
+// checkEscrowConfirmTTLLocked ends a typed-back wait that ran past escrowConfirmTTL (R-1151): the package is
+// dropped (never uploaded) and the job reads unconfirmed_void. Not while a confirm is uploading. Caller holds escrowMu.
+func (s *Server) checkEscrowConfirmTTLLocked() {
+	if !s.escrowAwaitConfirm || s.escrowClaiming || !s.nowFn().After(s.escrowConfirmExpiry) {
+		return
+	}
+	s.escrowPending = nil
+	s.wipeEscrowConfirmLocked()
+	if s.escrowJob != nil {
+		s.escrowJob.Phase = escrowPhaseUnconfirmed
+		s.escrowJob.UpdatedAt = s.nowFn().UTC().Format(time.RFC3339)
+		s.logger.Info("local-api: escrow code was not typed back in time — the new package was dropped; the previous one stays current (R-1151)",
+			"job_id", s.escrowJob.JobID)
+	}
+}
+
+// pickConfirmPositions draws two distinct 1-based word positions out of n (n ≥ 2), smaller first.
+func pickConfirmPositions(n int) ([2]int, error) {
+	a, err := rand.Int(rand.Reader, big.NewInt(int64(n)))
+	if err != nil {
+		return [2]int{}, err
+	}
+	b, err := rand.Int(rand.Reader, big.NewInt(int64(n-1)))
+	if err != nil {
+		return [2]int{}, err
+	}
+	i, j := int(a.Int64()), int(b.Int64())
+	if j >= i {
+		j++
+	}
+	if j < i {
+		i, j = j, i
+	}
+	return [2]int{i + 1, j + 1}, nil
 }
 
 // expireEscrowRLocked applies the TTL outcome: an armed, unclaimed holder is zeroed and the job
@@ -172,6 +256,9 @@ func (s *Server) expireEscrowRLocked() {
 	if s.escrowJob == nil || s.escrowJob.Phase != escrowPhaseDone || s.escrowRClaimed {
 		return
 	}
+	if s.escrowClaiming {
+		return // R-978: the claim is uploading the package; voiding now would store a package nobody gets the code to
+	}
 	if len(s.escrowR) == 0 {
 		return
 	}
@@ -217,7 +304,7 @@ func (s *Server) handleEscrowCeremonyStart(w http.ResponseWriter, r *http.Reques
 		UpdatedAt: s.nowFn().UTC().Format(time.RFC3339),
 	}
 	if !s.tryStartEscrowCeremony(job) {
-		writeErr(w, http.StatusConflict, "an escrow ceremony is already running")
+		writeErr(w, http.StatusConflict, "an escrow ceremony is already running, or its code is being claimed")
 		return
 	}
 	s.logger.Info("local-api: escrow ceremony started (controller-driven)", "vmid", vmid, "job_id", job.JobID)
@@ -247,7 +334,7 @@ func (s *Server) handleEscrowCeremonyStart(w http.ResponseWriter, r *http.Reques
 				detail += " | stderr: " + tail
 			}
 			job.Phase, job.Detail = escrowPhaseFailed, detail
-			s.finishEscrowCeremony(job, "")
+			s.finishEscrowCeremony(job, "", nil)
 			s.logger.Warn("local-api: escrow ceremony failed", "job_id", job.JobID,
 				"exit", exit, "duration_ms", time.Since(start).Milliseconds())
 			return
@@ -255,24 +342,34 @@ func (s *Server) handleEscrowCeremonyStart(w http.ResponseWriter, r *http.Reques
 		var out escrow.CeremonyOutput
 		if err := json.Unmarshal(stdout, &out); err != nil {
 			job.Phase, job.Detail = escrowPhaseFailed, "ceremony output is not the expected JSON object"
-			s.finishEscrowCeremony(job, "")
+			s.finishEscrowCeremony(job, "", nil)
 			s.logger.Warn("local-api: escrow ceremony output unparseable (never logged)", "job_id", job.JobID)
 			return
 		}
 		if out.Version != escrow.CeremonyOutputVersion || out.RecoveryCode == "" {
 			out.RecoveryCode = ""
 			job.Phase, job.Detail = escrowPhaseFailed, fmt.Sprintf("unexpected ceremony output (version %d)", out.Version)
-			s.finishEscrowCeremony(job, "")
+			s.finishEscrowCeremony(job, "", nil)
+			return
+		}
+		// R-978: the package must come back un-uploaded. An output that uploaded by itself (or carries no
+		// package) would mean the previous package is already gone, or the new one can never be stored.
+		if out.Uploaded || len(out.PendingUpload) == 0 {
+			out.RecoveryCode = ""
+			job.Phase, job.Detail = escrowPhaseFailed, "ceremony output carries no pending package"
+			s.finishEscrowCeremony(job, "", nil)
+			s.logger.Warn("local-api: escrow ceremony output has no pending package — failed (R-978)", "job_id", job.JobID,
+				"uploaded", out.Uploaded)
 			return
 		}
 		job.Phase = escrowPhaseDone
 		job.KeyFingerprint = out.KeyFingerprint
 		job.EntropyBits = out.EntropyBits
 		job.ResticPwSealed = out.ResticPwSealed
-		job.Uploaded = out.Uploaded
-		s.finishEscrowCeremony(job, out.RecoveryCode)
+		job.Uploaded = false // R-978: uploaded by the claim
+		s.finishEscrowCeremony(job, out.RecoveryCode, out.PendingUpload)
 		out.RecoveryCode = "" // drop the parsed reference promptly (GC caveat: best-effort)
-		s.logger.Info("local-api: escrow ceremony complete — R claimable (in-memory, one-shot)",
+		s.logger.Info("local-api: escrow ceremony complete — R claimable (in-memory, one-shot); package uploads at the claim, or at the typed-back confirm (R-1151)",
 			"job_id", job.JobID, "restic_pw_sealed", job.ResticPwSealed, "uploaded", job.Uploaded,
 			"claim_ttl_s", int(escrowClaimTTL.Seconds()), "duration_ms", time.Since(start).Milliseconds())
 	}()
@@ -290,6 +387,7 @@ func (s *Server) handleEscrowCeremonyStatus(w http.ResponseWriter, r *http.Reque
 	}
 	s.escrowMu.Lock()
 	s.checkEscrowTTLLocked()
+	s.checkEscrowConfirmTTLLocked()
 	if s.escrowJob == nil {
 		s.escrowMu.Unlock()
 		writeOK(w, map[string]any{"vmid": vmid, "phase": escrowPhaseNone})
@@ -304,6 +402,13 @@ func (s *Server) handleEscrowCeremonyStatus(w http.ResponseWriter, r *http.Reque
 		}
 	}
 	claimed := s.escrowRClaimed
+	awaiting := s.escrowAwaitConfirm
+	confirmIn := 0
+	if awaiting {
+		if d := s.escrowConfirmExpiry.Sub(s.nowFn()); d > 0 {
+			confirmIn = int(d.Seconds())
+		}
+	}
 	s.escrowMu.Unlock()
 
 	writeOK(w, map[string]any{
@@ -312,6 +417,7 @@ func (s *Server) handleEscrowCeremonyStatus(w http.ResponseWriter, r *http.Reque
 		"key_fingerprint": job.KeyFingerprint, "entropy_bits": job.EntropyBits,
 		"restic_pw_sealed": job.ResticPwSealed, "uploaded": job.Uploaded,
 		"claimable": claimable, "claimed": claimed, "claim_expires_in_sec": expiresIn,
+		"awaiting_confirm": awaiting, "confirm_expires_in_sec": confirmIn,
 		"detail": job.Detail,
 	})
 }
@@ -324,6 +430,14 @@ func (s *Server) handleEscrowCeremonyClaim(w http.ResponseWriter, r *http.Reques
 		writeErr(w, http.StatusServiceUnavailable, "escrow ceremony not configured on this host")
 		return
 	}
+	// R-1151: a controller ≥ 0.332.0 asks for the package to wait for the typed-back words. An older controller
+	// sends `{}` (or nothing) and gets the R-978 order. Unknown fields are refused (decodeBody).
+	var req struct {
+		DeferUpload bool `json:"defer_upload"`
+	}
+	if !decodeBody(w, r, &req) {
+		return
+	}
 	s.escrowMu.Lock()
 	s.checkEscrowTTLLocked()
 	switch {
@@ -339,11 +453,101 @@ func (s *Server) handleEscrowCeremonyClaim(w http.ResponseWriter, r *http.Reques
 		s.escrowMu.Unlock()
 		writeErr(w, http.StatusConflict, "ceremony failed — nothing to claim")
 		return
+	case s.escrowClaiming:
+		s.escrowMu.Unlock()
+		writeErr(w, http.StatusConflict, "the recovery code is being claimed — try again in a moment")
+		return
 	case s.escrowRClaimed || len(s.escrowR) == 0 || s.escrowJob.Phase == escrowPhaseVoid:
 		s.escrowMu.Unlock()
 		writeErr(w, http.StatusGone, "the recovery code is no longer available (already claimed or expired) — run a new ceremony; the new code supersedes")
 		return
 	}
+	// R-1151: the deferred order — hand out the code and two word positions; the package waits for
+	// POST /escrow/ceremony/confirm. Needs an uploader (a confirm could never store it otherwise) and a code of ≥ 2
+	// words (always 10; anything else falls back to the R-978 order rather than to no check).
+	if req.DeferUpload && len(s.escrowPending) > 0 && s.escrowCeremony.Upload != nil {
+		words := strings.Split(string(s.escrowR), escrow.RecoveryCodeSep)
+		if len(words) >= 2 {
+			pos, perr := pickConfirmPositions(len(words))
+			if perr == nil {
+				s.escrowConfirmPos = pos
+				s.escrowConfirmWords = [2][]byte{[]byte(words[pos[0]-1]), []byte(words[pos[1]-1])}
+				for i := range words {
+					words[i] = ""
+				}
+				s.escrowAwaitConfirm = true
+				s.escrowConfirmExpiry = s.nowFn().Add(escrowConfirmTTL)
+				recovery := string(s.escrowR)
+				zeroBytes(s.escrowR) // R only — the package and the two words stay for the confirm
+				s.escrowR = nil
+				s.escrowRClaimed = true
+				jobID := s.escrowJob.JobID
+				s.escrowMu.Unlock()
+				time.AfterFunc(escrowConfirmTTL+time.Second, func() {
+					s.escrowMu.Lock()
+					defer s.escrowMu.Unlock()
+					if s.escrowJob != nil && s.escrowJob.JobID == jobID {
+						s.checkEscrowConfirmTTLLocked()
+					}
+				})
+				s.logger.Info("local-api: escrow ceremony R claimed — the package waits for the typed-back words (R-1151)",
+					"vmid", vmid, "job_id", jobID, "confirm_ttl_s", int(escrowConfirmTTL.Seconds()))
+				w.Header().Set("Cache-Control", "no-store")
+				writeOK(w, map[string]any{"vmid": vmid, "job_id": jobID, "recovery_code": recovery,
+					"confirm_required": true, "verify_positions": []int{pos[0], pos[1]},
+					"confirm_expires_in_sec": int(escrowConfirmTTL.Seconds())})
+				recovery = ""
+				_ = recovery
+				return
+			}
+			s.logger.Warn("local-api: escrow claim — could not draw the confirm positions; package uploads at the claim (R-978 order)", "err", perr)
+		}
+	}
+	// R-978: store the NEW package on the hub first; only then hand out the code that opens it. Until this
+	// upload succeeds the hub keeps the previous package, and the customer's previous code keeps working.
+	if len(s.escrowPending) > 0 {
+		if s.escrowCeremony.Upload == nil {
+			s.escrowMu.Unlock()
+			writeErr(w, http.StatusServiceUnavailable, "the recovery package cannot be stored on this host (no upload configured) — the previous code still works")
+			return
+		}
+		pkg := s.escrowPending
+		jobID := s.escrowJob.JobID
+		s.escrowClaiming = true
+		s.escrowMu.Unlock()
+
+		start := time.Now()
+		base := s.baseCtx
+		if base == nil {
+			base = context.Background()
+		}
+		uctx, cancel := context.WithTimeout(base, escrowUploadTimeout)
+		err := s.escrowCeremony.Upload(uctx, pkg)
+		cancel()
+
+		s.escrowMu.Lock()
+		s.escrowClaiming = false
+		if err != nil {
+			s.escrowMu.Unlock()
+			s.logger.Warn("local-api: escrow claim — package upload to the hub failed; code kept claimable, previous package stays current (R-978)",
+				"vmid", vmid, "job_id", jobID, "err", err, "duration_ms", time.Since(start).Milliseconds())
+			writeErr(w, http.StatusBadGateway, "the new recovery package could not be stored on the hub — the previous code still works; try again")
+			return
+		}
+		s.escrowPending = nil
+		if s.escrowJob != nil && s.escrowJob.JobID == jobID {
+			s.escrowJob.Uploaded = true
+			s.escrowJob.UpdatedAt = s.nowFn().UTC().Format(time.RFC3339)
+		}
+		s.logger.Info("local-api: escrow claim — package uploaded to the hub; handing out the code (R-978)",
+			"vmid", vmid, "job_id", jobID, "duration_ms", time.Since(start).Milliseconds())
+		if len(s.escrowR) == 0 { // cannot happen (TTL and supersede wait for the claim) — never hand out nothing
+			s.escrowMu.Unlock()
+			s.logger.Error("local-api: escrow claim — code vanished during the upload (invariant broken)", "job_id", jobID)
+			writeErr(w, http.StatusInternalServerError, "the recovery code was lost during the claim — run a new ceremony")
+			return
+		}
+	}
 	recovery := string(s.escrowR)
 	s.wipeEscrowRLocked()
 	s.escrowRClaimed = true
@@ -352,11 +556,103 @@ func (s *Server) handleEscrowCeremonyClaim(w http.ResponseWriter, r *http.Reques
 
 	s.logger.Info("local-api: escrow ceremony R claimed (one-shot; holder zeroed)", "vmid", vmid, "job_id", jobID)
 	w.Header().Set("Cache-Control", "no-store")
-	writeOK(w, map[string]any{"vmid": vmid, "job_id": jobID, "recovery_code": recovery})
+	writeOK(w, map[string]any{"vmid": vmid, "job_id": jobID, "recovery_code": recovery, "confirm_required": false})
 	recovery = "" // drop the reference promptly (GC caveat: best-effort)
 	_ = recovery
 }
 
+// handleEscrowCeremonyConfirm is POST /escrow/ceremony/confirm {"words": [w1, w2]} (R-1151): the two words at the
+// positions the claim named, as the household typed them back. Right → the package goes to the hub now (200); wrong →
+// 422 and the wait goes on; nothing waiting → 404 (or 410 after the window ran out); a failed upload → 502 and the
+// wait goes on (a retry may succeed). The words are compared in constant time, case and spaces aside.
+func (s *Server) handleEscrowCeremonyConfirm(w http.ResponseWriter, r *http.Request, vmid int) {
+	if s.escrowCeremony == nil {
+		writeErr(w, http.StatusServiceUnavailable, "escrow ceremony not configured on this host")
+		return
+	}
+	var req struct {
+		Words []string `json:"words"`
+	}
+	if !decodeBody(w, r, &req) {
+		return
+	}
+	s.escrowMu.Lock()
+	s.checkEscrowConfirmTTLLocked()
+	switch {
+	case s.escrowClaiming:
+		s.escrowMu.Unlock()
+		writeErr(w, http.StatusConflict, "the recovery package is being stored — try again in a moment")
+		return
+	case !s.escrowAwaitConfirm:
+		gone := s.escrowJob != nil && s.escrowJob.Phase == escrowPhaseUnconfirmed
+		s.escrowMu.Unlock()
+		if gone {
+			writeErr(w, http.StatusGone, "the new recovery code was not confirmed in time — the previous code still works; run a new ceremony")
+			return
+		}
+		writeErr(w, http.StatusNotFound, "no recovery code waits for confirmation")
+		return
+	case len(req.Words) != 2:
+		s.escrowMu.Unlock()
+		writeErr(w, http.StatusBadRequest, "two words are needed")
+		return
+	}
+	match := 1
+	for i := 0; i < 2; i++ {
+		typed := []byte(strings.ToLower(strings.TrimSpace(req.Words[i])))
+		match &= subtle.ConstantTimeCompare(typed, s.escrowConfirmWords[i])
+		zeroBytes(typed)
+	}
+	req.Words = nil
+	if match != 1 {
+		s.escrowMu.Unlock()
+		s.logger.Info("local-api: escrow confirm — the typed words do not match; still waiting (R-1151)", "vmid", vmid)
+		writeErr(w, http.StatusUnprocessableEntity, "the words do not match the recovery code")
+		return
+	}
+	if s.escrowCeremony.Upload == nil || len(s.escrowPending) == 0 { // the claim refuses to defer without both
+		s.escrowMu.Unlock()
+		writeErr(w, http.StatusServiceUnavailable, "the recovery package cannot be stored on this host")
+		return
+	}
+	pkg := s.escrowPending
+	jobID := ""
+	if s.escrowJob != nil {
+		jobID = s.escrowJob.JobID
+	}
+	s.escrowClaiming = true
+	s.escrowMu.Unlock()
+
+	start := time.Now()
+	base := s.baseCtx
+	if base == nil {
+		base = context.Background()
+	}
+	uctx, cancel := context.WithTimeout(base, escrowUploadTimeout)
+	err := s.escrowCeremony.Upload(uctx, pkg)
+	cancel()
+
+	s.escrowMu.Lock()
+	s.escrowClaiming = false
+	if err != nil {
+		s.escrowMu.Unlock()
+		s.logger.Warn("local-api: escrow confirm — package upload to the hub failed; still waiting, previous package stays current (R-1151)",
+			"vmid", vmid, "job_id", jobID, "err", err, "duration_ms", time.Since(start).Milliseconds())
+		writeErr(w, http.StatusBadGateway, "the new recovery package could not be stored on the hub — the previous code still works; try again")
+		return
+	}
+	s.escrowPending = nil
+	s.wipeEscrowConfirmLocked()
+	if s.escrowJob != nil && s.escrowJob.JobID == jobID {
+		s.escrowJob.Uploaded = true
+		s.escrowJob.UpdatedAt = s.nowFn().UTC().Format(time.RFC3339)
+	}
+	s.escrowMu.Unlock()
+	s.logger.Info("local-api: escrow confirm — words typed back; the new package is on the hub (R-1151)",
+		"vmid", vmid, "job_id", jobID, "duration_ms", time.Since(start).Milliseconds())
+	writeOK(w, map[string]any{"vmid": vmid, "job_id": jobID, "uploaded": true})
+}
+
 // handleEscrowPreflight is GET /escrow/preflight: the wizard's prerequisite checklist. Each item
 // is {id, ok, detail}. Deliberately NOT checked: the PBS key file itself — /etc/pve/priv is 0700
 // root and the daemon cannot stat it; a missing key fails the ceremony fast with a clear error

internal/localapi/escrow_ceremony_test.go

@@ -2,12 +2,14 @@ package localapi
 
 import (
 	"context"
+	"encoding/base64"
 	"encoding/json"
 	"fmt"
 	"net/http"
 	"os"
 	"path/filepath"
 	"strings"
+	"sync"
 	"sync/atomic"
 	"testing"
 	"time"
@@ -19,25 +21,91 @@ import (
 
 const testR = "canary-alpha-bravo-charlie-delta-echo-foxtrot-golf-hotel-india"
 
-// cannedCeremonyJSON is shaped exactly like the agent's --output=json object (spike §2.6 values).
+// cannedCeremonyJSON is shaped exactly like the agent's --output=json object (spike §2.6 values;
+// version 2, R-978: the package comes back un-uploaded in pending_upload). The package is a MODEL of
+// the R-wrapped blob: it "opens" with exactly the code it was sealed under (fakeEscrowHub.opensWith).
 func cannedCeremonyJSON(r string) string {
-	return fmt.Sprintf(`{"version":1,"recovery_code":%q,"key_fingerprint":"f2:87:68:2a:88:50:16:01","entropy_bits":129,"blob_bytes":383,"identity_blob_bytes":450,"restic_pw_sealed":true,"uploaded":true}`, r)
+	return fmt.Sprintf(`{"version":2,"recovery_code":%q,"key_fingerprint":"f2:87:68:2a:88:50:16:01","entropy_bits":129,"blob_bytes":383,"identity_blob_bytes":450,"restic_pw_sealed":true,"uploaded":false,"pending_upload":%s}`,
+		r, sealedPackage(r))
+}
+
+// sealedPackage is the model hub upload body for a ceremony that minted code r.
+func sealedPackage(r string) string {
+	return fmt.Sprintf(`{"blob_b64":%q,"key_fingerprint":"f2:87:68:2a:88:50:16:01","posture":"zero_knowledge","created_at":"2026-06-10T12:00:00Z"}`,
+		base64.StdEncoding.EncodeToString([]byte("sealed-under:"+r)))
+}
+
+// fakeEscrowHub is the hub's current escrow package for this host (the PUT target). Upload replaces it,
+// as the hub does when the sealed secrets are unchanged (the R-978 case: no retained copy).
+type fakeEscrowHub struct {
+	mu      sync.Mutex
+	current []byte
+	uploads int
+	err     error         // non-nil → the upload fails, nothing stored
+	gate    chan struct{} // non-nil → the upload blocks until closed
+	entered chan struct{} // non-nil → closed when an upload starts
+}
+
+func (f *fakeEscrowHub) upload(ctx context.Context, pkg []byte) error {
+	if f.entered != nil {
+		close(f.entered)
+		f.entered = nil
+	}
+	if f.gate != nil {
+		<-f.gate
+	}
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	if f.err != nil {
+		return f.err
+	}
+	f.current = append([]byte(nil), pkg...)
+	f.uploads++
+	return nil
+}
+
+// opensWith reports whether the hub's CURRENT package opens with code r — the customer's question.
+func (f *fakeEscrowHub) opensWith(r string) bool {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	var pkg struct {
+		BlobB64 string `json:"blob_b64"`
+	}
+	if json.Unmarshal(f.current, &pkg) != nil {
+		return false
+	}
+	raw, err := base64.StdEncoding.DecodeString(pkg.BlobB64)
+	return err == nil && string(raw) == "sealed-under:"+r
+}
+
+func (f *fakeEscrowHub) count() int {
+	f.mu.Lock()
+	defer f.mu.Unlock()
+	return f.uploads
 }
 
 // newEscrowTestServer builds a server with the ceremony configured and every seam faked.
 func newEscrowTestServer(t *testing.T, run ceremonyRunner) *Server {
+	t.Helper()
+	srv, _ := newEscrowTestServerHub(t, run, &fakeEscrowHub{})
+	return srv
+}
+
+// newEscrowTestServerHub is newEscrowTestServer with the hub's escrow record exposed (R-978).
+func newEscrowTestServerHub(t *testing.T, run ceremonyRunner, hub *fakeEscrowHub) (*Server, *fakeEscrowHub) {
 	t.Helper()
 	srv := newTestServerS(t, &fakeGuests{}, &fakeBackups{}, &fakeStore{}, nil)
 	srv.escrowCeremony = &EscrowCeremonyConfig{
 		PBSStorageID:  "felhom-pbs",
 		HubConfigured: true,
 		DRConfigured:  func() bool { return true },
+		Upload:        hub.upload,
 	}
 	srv.ceremonyRun = run
 	srv.escrowSudoCheck = func(context.Context) error { return nil }
 	srv.escrowLookPath = func(string) (string, error) { return "/usr/bin/age", nil }
 	srv.statFile = func(string) bool { return true }
-	return srv
+	return srv, hub
 }
 
 // okRunner returns the canned success output and counts invocations.
@@ -160,8 +228,8 @@ func TestEscrowCeremony_OneShotClaim(t *testing.T) {
 	if !strings.Contains(body, `"phase":"done"`) || !strings.Contains(body, `"claimable":true`) {
 		t.Fatalf("status not done/claimable: %s", body)
 	}
-	if !strings.Contains(body, `"restic_pw_sealed":true`) || !strings.Contains(body, `"uploaded":true`) {
-		t.Fatalf("summary fields missing: %s", body)
+	if !strings.Contains(body, `"restic_pw_sealed":true`) || !strings.Contains(body, `"uploaded":false`) {
+		t.Fatalf("summary fields missing (R-978: not uploaded before the claim): %s", body)
 	}
 	assertNoR(t, "status payload", body)
 
@@ -199,7 +267,8 @@ func TestEscrowCeremony_OneShotClaim(t *testing.T) {
 
 	// Post-claim status: claimed, not claimable, still phase done.
 	st2 := do(t, h, "GET", "/escrow/ceremony/status", "A", "")
-	if !strings.Contains(st2.Body.String(), `"claimed":true`) || strings.Contains(st2.Body.String(), `"claimable":true`) {
+	if !strings.Contains(st2.Body.String(), `"claimed":true`) || strings.Contains(st2.Body.String(), `"claimable":true`) ||
+		!strings.Contains(st2.Body.String(), `"uploaded":true`) {
 		t.Fatalf("post-claim status wrong: %s", st2.Body.String())
 	}
 }
@@ -344,7 +413,7 @@ func TestEscrowCeremony_FailurePaths(t *testing.T) {
 	})
 	t.Run("wrong version", func(t *testing.T) {
 		srv := newEscrowTestServer(t, func(context.Context) ([]byte, []byte, int, error) {
-			return []byte(`{"version":2,"recovery_code":"` + testR + `"}`), nil, 0, nil
+			return []byte(`{"version":1,"recovery_code":"` + testR + `","uploaded":true}`), nil, 0, nil
 		})
 		h := srv.Handler()
 		startAndWait(t, srv, h)

internal/localapi/escrow_r1151_test.go

@@ -0,0 +1,223 @@
+package localapi
+
+import (
+	"encoding/json"
+	"errors"
+	"net/http"
+	"strings"
+	"testing"
+	"time"
+)
+
+// R-1151 — the new recovery code replaces the hub's package only after the household TYPED IT BACK. Before, the
+// claim uploaded it (R-978), so a household that opened the new code and closed the page without writing it down
+// had already replaced the package their old code opens. These tests ask the hub's current package the household's
+// question — „does my written-down code still open it?".
+
+type deferredClaim struct {
+	code      int
+	r         string
+	positions []int
+	required  bool
+	body      string
+}
+
+func claimDeferred(t *testing.T, h http.Handler) deferredClaim {
+	t.Helper()
+	w := do(t, h, "POST", "/escrow/ceremony/claim", "A", `{"defer_upload":true}`)
+	var env struct {
+		Data struct {
+			RecoveryCode    string `json:"recovery_code"`
+			ConfirmRequired bool   `json:"confirm_required"`
+			VerifyPositions []int  `json:"verify_positions"`
+		} `json:"data"`
+	}
+	_ = json.Unmarshal(w.Body.Bytes(), &env)
+	return deferredClaim{w.Code, env.Data.RecoveryCode, env.Data.VerifyPositions, env.Data.ConfirmRequired, w.Body.String()}
+}
+
+// typedBack is what a household that wrote the code down types for the claim's two positions.
+func (c deferredClaim) typedBack(t *testing.T) []string {
+	t.Helper()
+	words := strings.Split(c.r, "-")
+	if len(c.positions) != 2 {
+		t.Fatalf("the claim named %d positions, want 2: %s", len(c.positions), c.body)
+	}
+	return []string{words[c.positions[0]-1], words[c.positions[1]-1]}
+}
+
+func confirm(t *testing.T, h http.Handler, words []string) int {
+	t.Helper()
+	b, _ := json.Marshal(map[string]any{"words": words})
+	return do(t, h, "POST", "/escrow/ceremony/confirm", "A", string(b)).Code
+}
+
+// THE R-1151 CASE. Ceremony 1 is claimed and typed back (the household's code 1). Ceremony 2 is started and its code
+// SHOWN (claimed) — and then the page is closed: the words never come back. Three ways it ends: the confirm window
+// runs out, a new start supersedes it, the agent restarts. In every case code 1 must still open the hub's package.
+//
+// COMPANION RED-PROOF (observed 2026-10-19): ignore the claim's defer_upload (`if false && req.DeferUpload …`, so
+// every claim takes the R-978 order and uploads at once) → all three subtests FAIL with "a code that was shown but
+// never typed back replaced the package: code 1 no longer opens it (uploads=2)". Restored.
+func TestEscrowCeremony_R1151_ShownButNotTypedBackKeepsFirstCode(t *testing.T) {
+	setup := func(t *testing.T) (*Server, *fakeEscrowHub, http.Handler, *time.Time) {
+		srv, hub := newEscrowTestServerHub(t, sequenceRunner(firstR, secondR, "third-code-charlie"), &fakeEscrowHub{})
+		cur := testNow
+		srv.now = func() time.Time { return cur }
+		h := srv.Handler()
+		startAndWait(t, srv, h)
+		c1 := claimDeferred(t, h)
+		if c1.code != http.StatusOK || c1.r != firstR {
+			t.Fatalf("deferred claim of ceremony 1: %d %s", c1.code, c1.body)
+		}
+		if c1.required { // the household types code 1 back
+			if code := confirm(t, h, c1.typedBack(t)); code != http.StatusOK {
+				t.Fatalf("typing code 1 back: %d", code)
+			}
+		}
+		if !hub.opensWith(firstR) || hub.count() != 1 {
+			t.Fatalf("after ceremony 1 the hub must hold code 1's package (uploads=%d)", hub.count())
+		}
+		startAndWait(t, srv, h)
+		if c2 := claimDeferred(t, h); c2.code != http.StatusOK || c2.r != secondR {
+			t.Fatalf("deferred claim of ceremony 2: %d %s", c2.code, c2.body)
+		} // …shown, and the page is closed
+		return srv, hub, h, &cur
+	}
+	check := func(t *testing.T, hub *fakeEscrowHub) {
+		t.Helper()
+		if !hub.opensWith(firstR) {
+			t.Fatalf("a code that was shown but never typed back replaced the package: code 1 no longer opens it (uploads=%d)", hub.count())
+		}
+		if hub.opensWith(secondR) || hub.count() != 1 {
+			t.Fatalf("a package for the never-confirmed code 2 reached the hub (uploads=%d)", hub.count())
+		}
+	}
+
+	t.Run("the confirm window runs out", func(t *testing.T) {
+		_, hub, h, cur := setup(t)
+		*cur = cur.Add(escrowConfirmTTL + time.Minute)
+		check(t, hub)
+		st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String()
+		if !strings.Contains(st, `"phase":"unconfirmed_void"`) || !strings.Contains(st, `"awaiting_confirm":false`) {
+			t.Fatalf("after the window the job must read unconfirmed_void: %s", st)
+		}
+		if code := confirm(t, h, []string{"second", "bravo"}); code != http.StatusGone {
+			t.Fatalf("a confirm after the window: got %d, want 410", code)
+		}
+	})
+	t.Run("a new start supersedes it", func(t *testing.T) {
+		srv, hub, h, _ := setup(t)
+		startAndWait(t, srv, h)
+		if code := confirm(t, h, []string{"second", "bravo"}); code != http.StatusNotFound {
+			t.Fatalf("a confirm for the superseded code: got %d, want 404", code)
+		}
+		check(t, hub)
+	})
+	t.Run("the agent restarts", func(t *testing.T) {
+		_, hub, _, _ := setup(t)
+		srv2, _ := newEscrowTestServerHub(t, sequenceRunner("unused"), hub)
+		if code := confirm(t, srv2.Handler(), []string{"second", "bravo"}); code != http.StatusNotFound {
+			t.Fatalf("a confirm after a restart: got %d, want 404", code)
+		}
+		check(t, hub)
+	})
+}
+
+// The positive control and the typing: wrong words store nothing and the wait goes on; the right words (any case,
+// spaces around) store the package, once.
+//
+// COMPANION RED-PROOF (observed 2026-10-19): skip the word check (`if match != 1 && false`) → this fails with "one
+// wrong word: got 200 (uploads=1), want 422 and nothing stored". Restored.
+func TestEscrowCeremony_R1151_TypedBackStoresThePackage(t *testing.T) {
+	srv, hub := newEscrowTestServerHub(t, sequenceRunner(secondR), &fakeEscrowHub{})
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+	c := claimDeferred(t, h)
+	if c.code != http.StatusOK || !c.required || hub.count() != 0 {
+		t.Fatalf("a deferred claim must hand out the code and store nothing yet: %d required=%v uploads=%d", c.code, c.required, hub.count())
+	}
+	right := c.typedBack(t)
+	if code := confirm(t, h, []string{right[0], "wrong"}); code != http.StatusUnprocessableEntity || hub.count() != 0 {
+		t.Fatalf("one wrong word: got %d (uploads=%d), want 422 and nothing stored", code, hub.count())
+	}
+	if code := confirm(t, h, []string{right[0]}); code != http.StatusBadRequest {
+		t.Fatalf("one word only: got %d, want 400", code)
+	}
+	st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String()
+	if !strings.Contains(st, `"awaiting_confirm":true`) || !strings.Contains(st, `"uploaded":false`) {
+		t.Fatalf("a wrong try must leave the wait open: %s", st)
+	}
+	assertNoRIn(t, "status while awaiting the words", st, secondR)
+	for _, w := range right {
+		assertNoRIn(t, "status while awaiting the words", st, `"`+w+`"`)
+	}
+	if code := confirm(t, h, []string{" " + strings.ToUpper(right[0]), right[1] + " "}); code != http.StatusOK {
+		t.Fatalf("the right words: got %d, want 200", code)
+	}
+	if !hub.opensWith(secondR) || hub.count() != 1 {
+		t.Fatalf("after the words came back the hub must hold code 2's package (uploads=%d)", hub.count())
+	}
+	if code := confirm(t, h, right); code != http.StatusNotFound || hub.count() != 1 {
+		t.Fatalf("a second confirm: got %d (uploads=%d), want 404 and no second upload", code, hub.count())
+	}
+	if st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String(); !strings.Contains(st, `"uploaded":true`) {
+		t.Fatalf("status after the confirm: %s", st)
+	}
+}
+
+// A controller that has not updated (0.331.0 and older) claims with `{}` and never confirms: it keeps the R-978
+// order — the package is stored at the claim — so its household's ceremony still completes.
+func TestEscrowCeremony_R1151_OldControllerKeepsTheClaimOrder(t *testing.T) {
+	for _, body := range []string{"", "{}"} {
+		srv, hub := newEscrowTestServerHub(t, sequenceRunner(firstR), &fakeEscrowHub{})
+		h := srv.Handler()
+		startAndWait(t, srv, h)
+		w := do(t, h, "POST", "/escrow/ceremony/claim", "A", body)
+		if w.Code != http.StatusOK || !strings.Contains(w.Body.String(), firstR) || !strings.Contains(w.Body.String(), `"confirm_required":false`) {
+			t.Fatalf("old-shape claim %q: %d %s", body, w.Code, w.Body.String())
+		}
+		if !hub.opensWith(firstR) || hub.count() != 1 {
+			t.Fatalf("old-shape claim %q must store the package at once (uploads=%d)", body, hub.count())
+		}
+		if code := confirm(t, h, []string{"first", "alpha"}); code != http.StatusNotFound {
+			t.Fatalf("nothing waits after an old-shape claim: got %d, want 404", code)
+		}
+	}
+}
+
+// The hub failing at the confirm keeps the wait open and the previous package current; a retry succeeds.
+func TestEscrowCeremony_R1151_UploadFailureAtConfirmKeepsWaiting(t *testing.T) {
+	hub := &fakeEscrowHub{err: errors.New("hub returned HTTP 502")}
+	srv, _ := newEscrowTestServerHub(t, sequenceRunner(secondR), hub)
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+	c := claimDeferred(t, h)
+	if code := confirm(t, h, c.typedBack(t)); code != http.StatusBadGateway {
+		t.Fatalf("confirm with the hub failing: got %d, want 502", code)
+	}
+	hub.mu.Lock()
+	hub.err = nil
+	hub.mu.Unlock()
+	if code := confirm(t, h, c.typedBack(t)); code != http.StatusOK || !hub.opensWith(secondR) {
+		t.Fatalf("retry after the hub recovered: %d (opens: %v)", code, hub.opensWith(secondR))
+	}
+}
+
+// The two positions are distinct, in range, smaller first — over many draws every position appears.
+func TestPickConfirmPositions(t *testing.T) {
+	seen := map[int]bool{}
+	for i := 0; i < 2000; i++ {
+		p, err := pickConfirmPositions(10)
+		if err != nil {
+			t.Fatal(err)
+		}
+		if p[0] < 1 || p[1] > 10 || p[0] >= p[1] {
+			t.Fatalf("bad positions %v", p)
+		}
+		seen[p[0]], seen[p[1]] = true, true
+	}
+	if len(seen) != 10 {
+		t.Fatalf("only positions %v were ever drawn", seen)
+	}
+}

internal/localapi/escrow_r978_test.go

@@ -0,0 +1,219 @@
+package localapi
+
+import (
+	"context"
+	"errors"
+	"net/http"
+	"strings"
+	"testing"
+	"time"
+)
+
+// R-978 — starting the escrow ceremony again must not make the customer's written-down recovery code
+// useless. The hub overwrites the current package when the sealed secrets are unchanged, so the NEW
+// package may reach the hub only when its code is claimed. These tests ask the customer's question of
+// the hub's current package — „does my code still open it?" — not whether some function ran.
+
+const (
+	firstR  = "first-ceremony-code-alpha"
+	secondR = "second-ceremony-code-bravo"
+)
+
+// sequenceRunner mints the given codes in order, one per ceremony.
+func sequenceRunner(codes ...string) ceremonyRunner {
+	i := 0
+	return func(context.Context) ([]byte, []byte, int, error) {
+		r := codes[i]
+		if i < len(codes)-1 {
+			i++
+		}
+		return []byte(cannedCeremonyJSON(r)), nil, 0, nil
+	}
+}
+
+func claimCode(t *testing.T, h http.Handler) (int, string) {
+	t.Helper()
+	w := do(t, h, "POST", "/escrow/ceremony/claim", "A", "")
+	return w.Code, w.Body.String()
+}
+
+// THE R-978 CASE. Ceremony 1 is claimed (the customer writes code 1 down). Ceremony 2 is started and
+// abandoned — three ways: its code expires unclaimed, a third start supersedes it, the agent restarts.
+// In every case the hub's current package must still open with code 1, and nothing was uploaded for 2.
+//
+// COMPANION RED-PROOF (observed 2026-10-18): restore the pre-R-978 shape — the package uploaded when the
+// ceremony finishes (s.escrowCeremony.Upload(base, out.PendingUpload) in the job goroutine, as the root
+// subprocess did with `--upload`) and nothing left for the claim (finishEscrowCeremony(job, R, nil)) →
+// all three subtests FAIL with "the abandoned second ceremony replaced the package: code 1 no longer
+// opens it (uploads=2)" (uploads=3 for the superseding start). Restored.
+func TestEscrowCeremony_R978_AbandonedSecondCeremonyKeepsFirstCode(t *testing.T) {
+	setup := func(t *testing.T) (*Server, *fakeEscrowHub, http.Handler, *time.Time) {
+		srv, hub := newEscrowTestServerHub(t, sequenceRunner(firstR, secondR, "third-code-charlie"), &fakeEscrowHub{})
+		cur := testNow
+		srv.now = func() time.Time { return cur }
+		h := srv.Handler()
+		startAndWait(t, srv, h)
+		if code, body := claimCode(t, h); code != http.StatusOK || !strings.Contains(body, firstR) {
+			t.Fatalf("claim of ceremony 1: %d %s", code, body)
+		}
+		if !hub.opensWith(firstR) || hub.count() != 1 {
+			t.Fatalf("after the first claim the hub must hold code 1's package (uploads=%d)", hub.count())
+		}
+		startAndWait(t, srv, h) // ceremony 2 — about to be abandoned
+		return srv, hub, h, &cur
+	}
+	check := func(t *testing.T, hub *fakeEscrowHub) {
+		t.Helper()
+		if !hub.opensWith(firstR) {
+			t.Fatalf("the abandoned second ceremony replaced the package: code 1 no longer opens it (uploads=%d)", hub.count())
+		}
+		if hub.opensWith(secondR) || hub.count() != 1 {
+			t.Fatalf("a package for the never-claimed code 2 reached the hub (uploads=%d)", hub.count())
+		}
+	}
+
+	t.Run("code 2 expires unclaimed", func(t *testing.T) {
+		_, hub, h, cur := setup(t)
+		*cur = cur.Add(escrowClaimTTL + time.Minute)
+		if code, _ := claimCode(t, h); code != http.StatusGone {
+			t.Fatalf("claim after TTL: got %d, want 410", code)
+		}
+		check(t, hub)
+	})
+	t.Run("a third start supersedes code 2", func(t *testing.T) {
+		srv, hub, h, _ := setup(t)
+		startAndWait(t, srv, h) // ceremony 3, also left unclaimed
+		check(t, hub)
+	})
+	t.Run("the agent restarts before code 2 is claimed", func(t *testing.T) {
+		_, hub, _, _ := setup(t)
+		// A restart is a new process with an empty slot; only the hub's record survives it.
+		srv2, _ := newEscrowTestServerHub(t, sequenceRunner("unused"), hub)
+		if code, _ := claimCode(t, srv2.Handler()); code != http.StatusNotFound {
+			t.Fatalf("claim after a restart: got %d, want 404", code)
+		}
+		check(t, hub)
+	})
+}
+
+// The positive control: CLAIMING code 2 is what replaces the package — the code handed out opens the
+// package the hub now holds, and the claim answers only after the hub took it.
+func TestEscrowCeremony_R978_ClaimStoresThePackageBeforeHandingOutTheCode(t *testing.T) {
+	srv, hub := newEscrowTestServerHub(t, sequenceRunner(firstR, secondR), &fakeEscrowHub{})
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+	if hub.count() != 0 {
+		t.Fatalf("the package was uploaded before any claim (uploads=%d)", hub.count())
+	}
+	claimCode(t, h)
+	startAndWait(t, srv, h)
+	code, body := claimCode(t, h)
+	if code != http.StatusOK || !strings.Contains(body, secondR) {
+		t.Fatalf("claim of ceremony 2: %d %s", code, body)
+	}
+	if !hub.opensWith(secondR) || hub.count() != 2 {
+		t.Fatalf("after claiming code 2 the hub must hold its package (uploads=%d, opens with 2: %v)", hub.count(), hub.opensWith(secondR))
+	}
+}
+
+// A failed upload hands out NO code, keeps the code claimable and the previous package current; a retry
+// that reaches the hub succeeds.
+func TestEscrowCeremony_R978_UploadFailureKeepsTheCodeClaimable(t *testing.T) {
+	hub := &fakeEscrowHub{err: errors.New("hub returned HTTP 502")}
+	srv, _ := newEscrowTestServerHub(t, sequenceRunner(firstR), hub)
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+
+	code, body := claimCode(t, h)
+	if code != http.StatusBadGateway {
+		t.Fatalf("claim with the hub failing: got %d, want 502 (%s)", code, body)
+	}
+	assertNoRIn(t, "failed-upload claim payload", body, firstR)
+	if st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String(); !strings.Contains(st, `"claimable":true`) || !strings.Contains(st, `"uploaded":false`) {
+		t.Fatalf("after a failed upload the code must stay claimable and un-uploaded: %s", st)
+	}
+
+	hub.mu.Lock()
+	hub.err = nil
+	hub.mu.Unlock()
+	code, body = claimCode(t, h)
+	if code != http.StatusOK || !strings.Contains(body, firstR) || !hub.opensWith(firstR) {
+		t.Fatalf("retry after the hub recovered: %d %s (opens: %v)", code, body, hub.opensWith(firstR))
+	}
+}
+
+// While the claim's upload is in flight, neither the TTL nor a new start may take the code away — else
+// the hub would hold a package whose code nobody was given (the R-978 failure, reached the other way).
+func TestEscrowCeremony_R978_TTLAndRestartWaitForAnInFlightClaim(t *testing.T) {
+	hub := &fakeEscrowHub{gate: make(chan struct{}), entered: make(chan struct{})}
+	entered := hub.entered
+	srv, _ := newEscrowTestServerHub(t, sequenceRunner(firstR, secondR), hub)
+	var mu = make(chan struct{}, 1)
+	mu <- struct{}{}
+	cur := testNow
+	srv.now = func() time.Time { <-mu; defer func() { mu <- struct{}{} }(); return cur }
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+
+	type res struct {
+		code int
+		body string
+	}
+	done := make(chan res, 1)
+	go func() { c, b := claimCode(t, h); done <- res{c, b} }()
+	<-entered
+
+	<-mu
+	cur = cur.Add(escrowClaimTTL + time.Minute) // the TTL passes mid-upload
+	mu <- struct{}{}
+	if st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String(); strings.Contains(st, `"phase":"unclaimed_void"`) {
+		t.Fatalf("the TTL voided the code while its package was uploading: %s", st)
+	}
+	if w := do(t, h, "POST", "/escrow/ceremony", "A", ""); w.Code != http.StatusConflict {
+		t.Fatalf("a new start during the claim's upload: got %d, want 409", w.Code)
+	}
+	if code, _ := claimCode(t, h); code != http.StatusConflict {
+		t.Fatalf("a second claim during the upload: got %d, want 409", code)
+	}
+	close(hub.gate)
+	r := <-done
+	if r.code != http.StatusOK || !strings.Contains(r.body, firstR) || !hub.opensWith(firstR) {
+		t.Fatalf("the in-flight claim must deliver the code whose package it stored: %d %s", r.code, r.body)
+	}
+}
+
+// No upload configured → the claim refuses rather than hand out a code for a package the hub lacks.
+func TestEscrowCeremony_R978_NoUploaderRefusesTheClaim(t *testing.T) {
+	srv, _ := newEscrowTestServerHub(t, sequenceRunner(firstR), &fakeEscrowHub{})
+	srv.escrowCeremony.Upload = nil
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+	code, body := claimCode(t, h)
+	if code != http.StatusServiceUnavailable {
+		t.Fatalf("claim with no uploader: got %d, want 503 (%s)", code, body)
+	}
+	assertNoRIn(t, "no-uploader claim payload", body, firstR)
+}
+
+// An output that uploaded by itself (the pre-R-978 shape) or carries no package fails the ceremony.
+func TestEscrowCeremony_R978_OutputWithoutPendingPackageFails(t *testing.T) {
+	srv, _ := newEscrowTestServerHub(t, func(context.Context) ([]byte, []byte, int, error) {
+		return []byte(`{"version":2,"recovery_code":"` + firstR + `","uploaded":true}`), nil, 0, nil
+	}, &fakeEscrowHub{})
+	h := srv.Handler()
+	startAndWait(t, srv, h)
+	st := do(t, h, "GET", "/escrow/ceremony/status", "A", "").Body.String()
+	if !strings.Contains(st, `"phase":"failed"`) || !strings.Contains(st, "no pending package") {
+		t.Fatalf("want failed: %s", st)
+	}
+	if code, _ := claimCode(t, h); code != http.StatusConflict {
+		t.Fatalf("claim on a failed ceremony: got %d, want 409", code)
+	}
+}
+
+func assertNoRIn(t *testing.T, where, body, r string) {
+	t.Helper()
+	if strings.Contains(body, r) {
+		t.Fatalf("%s carries the recovery code", where)
+	}
+}

internal/localapi/guestbind.go

@@ -1,13 +1,13 @@
 package localapi
 
 import (
-	"time"
 	"context"
 	"fmt"
 	"log/slog"
 	"strconv"
 	"strings"
 	"sync"
+	"time"
 
 	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
 )

internal/localapi/internalplace.go

@@ -0,0 +1,207 @@
+package localapi
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"os"
+	"path"
+	"path/filepath"
+	"regexp"
+	"sync"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/storage"
+)
+
+// The internal data place („Belső tárhely", agent v0.158.0) — felhom.eu architecture/16-storage-physical-identity.md
+// §5, §7. On a one-disk box the root helper configs/felhom-internal-place makes a second thin volume in local-lvm
+// (<vg>/felhom-belso, ext4, label felhom-belso). The installer (a new box) or the signed add_internal_place job (an old
+// box) then writes InternalPlaceRecord to the agent's state dir. From then on the agent treats the volume exactly like
+// an enrolled drive: a Felhom mount unit by fs-UUID at /mnt/belso (EnsureMount), the `enrolled` intent, and — when the
+// controller asks with its own token (POST /disks/guest-attach) — the bind at /mnt/felhom-drives/belso. So it is a
+// HOST mount under the guest's mp8 parent: not in the guest config, not in vzdump, not in PBS (`16` §5).
+//
+// What it is NOT: a drive the household can format, wipe, eject or decommission. Its device is a device-mapper volume
+// on the system disk, so the role gate (roleForMountPath → system) and the format path (ValidateBlockDevice refuses
+// dm) refuse all four — pinned by internalplace_test.go. Its identity is the system disk's (build A), so a Tier-2 copy
+// between it and sys_drive is never a second copy.
+const (
+	InternalPlaceName  = "belso"
+	InternalPlaceWhere = "/mnt/belso"
+	// InternalPlaceFile is the record's name inside the agent's state dir.
+	InternalPlaceFile = "internal-place.json"
+)
+
+var reFSUUID = regexp.MustCompile(`^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}$`)
+
+// InternalPlaceRecord is the state file the maker writes once the volume exists and is formatted.
+type InternalPlaceRecord struct {
+	FSUUID string `json:"fs_uuid"`
+	GiB    int    `json:"gib,omitempty"`
+	MadeBy string `json:"made_by,omitempty"` // "installer" | "signed-job"
+	MadeAt string `json:"made_at,omitempty"`
+}
+
+// ReadInternalPlace reads the record; ok=false when there is none (the common case: a box with no internal place).
+func ReadInternalPlace(file string) (InternalPlaceRecord, bool, error) {
+	var rec InternalPlaceRecord
+	b, err := os.ReadFile(file)
+	if errors.Is(err, os.ErrNotExist) {
+		return rec, false, nil
+	}
+	if err != nil {
+		return rec, false, err
+	}
+	if err := json.Unmarshal(b, &rec); err != nil {
+		return rec, false, fmt.Errorf("internal place record %s: %w", file, err)
+	}
+	if !reFSUUID.MatchString(rec.FSUUID) {
+		return rec, false, fmt.Errorf("internal place record %s: fs_uuid %q is not a filesystem UUID", file, rec.FSUUID)
+	}
+	return rec, true, nil
+}
+
+// WriteInternalPlace writes the record atomically (0600).
+func WriteInternalPlace(file string, rec InternalPlaceRecord) error {
+	if !reFSUUID.MatchString(rec.FSUUID) {
+		return fmt.Errorf("internal place: fs_uuid %q is not a filesystem UUID", rec.FSUUID)
+	}
+	b, err := json.MarshalIndent(rec, "", "  ")
+	if err != nil {
+		return err
+	}
+	tmp := file + ".tmp"
+	if err := os.WriteFile(tmp, append(b, '\n'), 0o600); err != nil {
+		return err
+	}
+	return os.Rename(tmp, file)
+}
+
+// internalPlaceState memoises what EnsureInternalPlace last saw, so the 20-second tick logs a state CHANGE once and
+// does no work on the steady state.
+type internalPlaceState struct {
+	mu      sync.Mutex
+	done    string // the fs-UUID that is mounted and recorded enrolled
+	lastWhy string // the last reason it could not be ensured (logged once per change)
+}
+
+// internalPlaceUUID returns the configured internal place's fs-UUID, or "" when this box has none (or the record is
+// unreadable — logged by EnsureInternalPlace).
+func (s *Server) internalPlaceUUID() string {
+	if s.internalPlaceFile == "" {
+		return ""
+	}
+	rec, ok, err := ReadInternalPlace(s.internalPlaceFile)
+	if err != nil || !ok {
+		return ""
+	}
+	return rec.FSUUID
+}
+
+// isInternalPlace reports whether a host path is this box's internal place: its raw mount (/mnt/belso) or its bind
+// under the guests' shared parent (/mnt/felhom-drives/belso — the path POST /disks/same-disk judges).
+func (s *Server) isInternalPlace(hostPath string) bool {
+	p := path.Clean(hostPath)
+	return (p == InternalPlaceWhere || p == StableParentDir+"/"+InternalPlaceName) && s.internalPlaceUUID() != ""
+}
+
+// EnsureInternalPlace mounts the internal place on the host at /mnt/belso by fs-UUID and records it `enrolled`,
+// when this box has one. Idempotent and cheap on the steady state; safe on the 20-second tick. It never formats,
+// never unmounts, and never mounts anything at /mnt/belso that is not the recorded UUID: a household drive that
+// happens to be named „belso" keeps its mount, and the internal place waits (logged once).
+func (s *Server) EnsureInternalPlace(ctx context.Context) {
+	if s.internalPlaceFile == "" {
+		return
+	}
+	rec, ok, err := ReadInternalPlace(s.internalPlaceFile)
+	if err != nil {
+		s.internalPlaceNote("record unreadable: " + err.Error())
+		return
+	}
+	if !ok {
+		return
+	}
+	uuid := rec.FSUUID
+	st := &s.ipState
+	st.mu.Lock()
+	done := st.done == uuid
+	st.mu.Unlock()
+
+	// What is at /mnt/belso now, if anything?
+	mountedUUID, mounted := "", false
+	if mounts, merr := s.hostReader().Mounts(); merr == nil {
+		for _, m := range mounts {
+			if path.Clean(m.MountPoint) == InternalPlaceWhere {
+				mounted = true
+				if u, okU := s.hostReader().ResolveUUID(m.Device); okU {
+					mountedUUID = u
+				}
+			}
+		}
+	}
+	if mounted && mountedUUID != "" && mountedUUID != uuid {
+		s.internalPlaceNote("another filesystem (uuid " + mountedUUID + ") is mounted at " + InternalPlaceWhere + " — the internal place waits")
+		return
+	}
+	// A Felhom unit for /mnt/belso that names another UUID (a household drive called „belso", unplugged): never
+	// overwrite it — that would silently take the drive's name and mount point.
+	if s.driveTargets != nil {
+		if drives, derr := s.driveTargets.Known(ctx); derr == nil {
+			for _, d := range drives {
+				if path.Clean(d.MountPath) == InternalPlaceWhere && d.UUID != "" && d.UUID != uuid {
+					s.internalPlaceNote("a drive unit for " + InternalPlaceWhere + " names another uuid (" + d.UUID + ") — the internal place waits")
+					return
+				}
+			}
+		}
+	}
+	if mounted && mountedUUID == uuid {
+		if !done {
+			s.recordIntent(ctx, InternalPlaceWhere, "enrolled")
+			st.mu.Lock()
+			st.done, st.lastWhy = uuid, ""
+			st.mu.Unlock()
+			s.logger.Info("internal place: mounted and enrolled", "where", InternalPlaceWhere, "fs_uuid", uuid, "gib", rec.GiB)
+		}
+		return
+	}
+	if s.disks == nil {
+		s.internalPlaceNote("disk management is not configured on this host")
+		return
+	}
+	if s.resolveStorageDevice != nil {
+		if _, rerr := s.resolveStorageDevice("uuid:" + uuid); rerr != nil {
+			s.internalPlaceNote("its volume is not present (uuid " + uuid + ")")
+			return
+		}
+	}
+	start := time.Now()
+	if err := s.disks.EnsureMount(ctx, storage.MountSpec{Name: InternalPlaceName, UUID: uuid, Where: InternalPlaceWhere, FSType: "ext4"}); err != nil {
+		s.internalPlaceNote("mount failed: " + err.Error())
+		return
+	}
+	s.recordIntent(ctx, InternalPlaceWhere, "enrolled")
+	st.mu.Lock()
+	st.done, st.lastWhy = uuid, ""
+	st.mu.Unlock()
+	s.logger.Info("internal place: mounted on the host", "where", InternalPlaceWhere, "fs_uuid", uuid, "gib", rec.GiB,
+		"duration_ms", time.Since(start).Milliseconds())
+}
+
+func (s *Server) internalPlaceNote(why string) {
+	st := &s.ipState
+	st.mu.Lock()
+	changed := st.lastWhy != why
+	st.lastWhy, st.done = why, ""
+	st.mu.Unlock()
+	if changed {
+		s.logger.Warn("internal place: not ensured", "why", why)
+	} else {
+		s.logger.Debug("internal place: still not ensured", "why", why)
+	}
+}
+
+// InternalPlaceFileIn is the record's path inside a state dir.
+func InternalPlaceFileIn(stateDir string) string { return filepath.Join(stateDir, InternalPlaceFile) }

internal/localapi/internalplace_test.go

@@ -0,0 +1,249 @@
+package localapi
+
+import (
+	"context"
+	"errors"
+	"io"
+	"log/slog"
+	"net/http"
+	"os/exec"
+	"path/filepath"
+	"runtime"
+	"sync"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/storage"
+)
+
+// The internal data place (felhom.eu architecture/16 §5, agent v0.158.0): mounted like an enrolled drive, but never
+// at the cost of a household drive, never formatted, ejected or decommissioned, and the same disk as sys_drive.
+
+const ipUUID = "0b6e4c5a-1d2e-4f3a-9b8c-7d6e5f4a3b2c"
+const ipDev = "/dev/mapper/pve-felhom--belso"
+
+// ipHost is a HostReader whose mount table the test changes, and whose by-uuid map it controls.
+type ipHost struct {
+	mu     sync.Mutex
+	mounts []storage.Mount
+	uuids  map[string]string // device → fs-UUID
+}
+
+func (h *ipHost) Mounts() ([]storage.Mount, error) {
+	h.mu.Lock()
+	defer h.mu.Unlock()
+	return append([]storage.Mount(nil), h.mounts...), nil
+}
+func (h *ipHost) ResolveUUID(dev string) (string, bool) { u, ok := h.uuids[dev]; return u, ok }
+func (h *ipHost) DeviceExists(string) bool              { return true }
+func (h *ipHost) Rotational(string) (bool, bool)        { return false, false }
+func (h *ipHost) Removable(string) (bool, bool)         { return false, false }
+func (h *ipHost) BlockSlaves(string) ([]string, bool)   { return nil, false }
+
+func newIPHost(extra ...storage.Mount) *ipHost {
+	return &ipHost{
+		// root on a whole-disk partition (as sysOnSDA): the system disk resolves, so a USB drive is user-data.
+		mounts: append([]storage.Mount{{Device: "/dev/sda1", MountPoint: "/", FSType: "ext4"}}, extra...),
+		uuids:  map[string]string{ipDev: ipUUID, "/dev/sdb1": "99999999-aaaa-bbbb-cccc-dddddddddddd"},
+	}
+}
+
+type ipRig struct {
+	srv    *Server
+	disks  *fakeDiskOps
+	intent *fakeIntent
+}
+
+func newIPRig(t *testing.T, host *ipHost, known []storage.KnownTarget, record bool) ipRig {
+	t.Helper()
+	dir := t.TempDir()
+	file := filepath.Join(dir, InternalPlaceFile)
+	if record {
+		if err := WriteInternalPlace(file, InternalPlaceRecord{FSUUID: ipUUID, GiB: 183, MadeBy: "installer"}); err != nil {
+			t.Fatal(err)
+		}
+	}
+	d, in := &fakeDiskOps{}, newFakeIntent()
+	srv, err := NewServer(Options{
+		ListenAddr: "127.0.0.1:0", Guests: &fakeGuests{}, Backups: &fakeBackups{}, Store: &fakeStore{},
+		Storage: fakeStorage{}, Tokens: staticTokens{"A": 8200}, Disks: d, DiskGate: &fakeGate{},
+		HostReader: host, Intent: in, DriveTargets: fakeKnownTargets{drives: known},
+		InternalPlaceFile: file, Logger: slog.New(slog.NewTextHandler(io.Discard, nil)),
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	srv.baseCtx = context.Background()
+	srv.resolveStorageDevice = func(id string) (string, error) {
+		if id == "uuid:"+ipUUID {
+			return "/dev/dm-7", nil
+		}
+		return "", errors.New("absent")
+	}
+	return ipRig{srv, d, in}
+}
+
+func TestInternalPlace_NoRecord_DoesNothing(t *testing.T) {
+	r := newIPRig(t, newIPHost(), nil, false)
+	r.srv.EnsureInternalPlace(context.Background())
+	if len(r.disks.mountCalls) != 0 {
+		t.Fatalf("no record: no mount may be made, got %+v", r.disks.mountCalls)
+	}
+}
+
+func TestInternalPlace_Record_MountsByUUIDAtMntBelso(t *testing.T) {
+	r := newIPRig(t, newIPHost(), nil, true)
+	r.srv.EnsureInternalPlace(context.Background())
+	if len(r.disks.mountCalls) != 1 {
+		t.Fatalf("want exactly one EnsureMount, got %+v", r.disks.mountCalls)
+	}
+	got := r.disks.mountCalls[0]
+	if got.UUID != ipUUID || got.Where != InternalPlaceWhere || got.FSType != "ext4" || got.Name != InternalPlaceName {
+		t.Fatalf("EnsureMount spec: %+v", got)
+	}
+}
+
+func TestInternalPlace_AlreadyMounted_EnrolledOnceNoRemount(t *testing.T) {
+	host := newIPHost(storage.Mount{Device: ipDev, MountPoint: InternalPlaceWhere, FSType: "ext4"})
+	r := newIPRig(t, host, nil, true)
+	r.srv.EnsureInternalPlace(context.Background())
+	r.srv.EnsureInternalPlace(context.Background())
+	if len(r.disks.mountCalls) != 0 {
+		t.Fatalf("already mounted: no EnsureMount, got %+v", r.disks.mountCalls)
+	}
+	if r.intent.Get("uuid:"+ipUUID) != storage.IntentEnrolled {
+		t.Fatalf("the internal place must be recorded enrolled")
+	}
+}
+
+// A household drive called „belso" keeps its mount point: the internal place never takes it.
+// Positive control: TestInternalPlace_Record_MountsByUUIDAtMntBelso (the same record, /mnt/belso free).
+func TestInternalPlace_NeverTakesAHouseholdDriveNamedBelso(t *testing.T) {
+	mounted := newIPHost(storage.Mount{Device: "/dev/sdb1", MountPoint: InternalPlaceWhere, FSType: "ext4"})
+	r := newIPRig(t, mounted, nil, true)
+	r.srv.EnsureInternalPlace(context.Background())
+	if len(r.disks.mountCalls) != 0 || r.intent.Get("uuid:"+ipUUID) != "" {
+		t.Fatalf("a foreign filesystem at /mnt/belso: nothing may be mounted or recorded, got %+v", r.disks.mountCalls)
+	}
+	unplugged := []storage.KnownTarget{{Name: "belso", UUID: "99999999-aaaa-bbbb-cccc-dddddddddddd", MountPath: InternalPlaceWhere, MountBacked: true}}
+	r2 := newIPRig(t, newIPHost(), unplugged, true)
+	r2.srv.EnsureInternalPlace(context.Background())
+	if len(r2.disks.mountCalls) != 0 {
+		t.Fatalf("an unplugged household drive's unit for /mnt/belso must not be overwritten, got %+v", r2.disks.mountCalls)
+	}
+}
+
+func TestInternalPlace_VolumeAbsent_NoMount(t *testing.T) {
+	r := newIPRig(t, newIPHost(), nil, true)
+	r.srv.resolveStorageDevice = func(string) (string, error) { return "", errors.New("absent") }
+	r.srv.EnsureInternalPlace(context.Background())
+	if len(r.disks.mountCalls) != 0 {
+		t.Fatalf("absent volume: no mount, got %+v", r.disks.mountCalls)
+	}
+}
+
+// The household cannot eject or decommission it: its device is a device-mapper volume on the system disk.
+// Positive control: the same calls for a whole-disk USB drive (/dev/sdb1) are not refused by role.
+func TestInternalPlace_EjectAndDecommissionRefused(t *testing.T) {
+	host := newIPHost(
+		storage.Mount{Device: ipDev, MountPoint: InternalPlaceWhere, FSType: "ext4"},
+		storage.Mount{Device: "/dev/sdb1", MountPoint: "/mnt/hdd_1", FSType: "ext4"},
+	)
+	r := newIPRig(t, host, nil, true)
+	h := r.srv.Handler()
+	for _, route := range []string{"/disks/eject", "/disks/decommission"} {
+		if w := do(t, h, "POST", route, "A", `{"where":"/mnt/belso"}`); w.Code != http.StatusForbidden {
+			t.Fatalf("%s /mnt/belso: want 403, got %d %s", route, w.Code, w.Body.String())
+		}
+	}
+	if w := do(t, h, "POST", "/disks/eject", "A", `{"where":"/mnt/hdd_1"}`); w.Code == http.StatusForbidden {
+		t.Fatalf("positive control: a USB drive's eject must not be refused by role, got %d %s", w.Code, w.Body.String())
+	}
+	for _, u := range r.disks.unmountCalls {
+		if u == InternalPlaceWhere {
+			t.Fatalf("the internal place must never be unmounted, got %+v", r.disks.unmountCalls)
+		}
+	}
+}
+
+// The consequence `16` §5 needs: a copy between the internal place and sys_drive is the SAME disk (build A then never
+// counts it), and the place is reported as kind internal.
+type ipIdentHost struct{ fakeIdentHost }
+
+func (h *ipIdentHost) KernelName(dev string) (string, bool) {
+	if dev == ipDev {
+		return "dm-7", true
+	}
+	return h.fakeIdentHost.KernelName(dev)
+}
+func (h *ipIdentHost) Slaves(n string) []string {
+	if n == "dm-7" {
+		return []string{"dm-3"} // felhom-belso → the thin pool
+	}
+	if n == "dm-3" {
+		return []string{"sda3"} // the pool → the system disk's LVM partition
+	}
+	return nil
+}
+
+func TestInternalPlace_IsTheSystemDiskAndKindInternal(t *testing.T) {
+	ih := &ipIdentHost{fakeIdentHost{
+		virt:   "none",
+		parent: map[string]string{"sda3": "sda"},
+		props:  map[string]map[string]string{"sda": {"ID_BUS": "ata", "ID_MODEL": "SanDisk", "ID_SERIAL_SHORT": "1"}},
+		mounts: []storage.Mount{
+			{Device: "/dev/mapper/pve-root", MountPoint: "/", FSType: "ext4"},
+			{Device: "/dev/mapper/pve-root", MountPoint: "/mnt/felhom-drives", FSType: "ext4"},
+			{Device: ipDev, MountPoint: InternalPlaceWhere, FSType: "ext4"},
+			{Device: ipDev, MountPoint: "/mnt/felhom-drives/belso", FSType: "ext4"},
+		},
+	}}
+	fg := &fakeGuestsCfg{mounts: map[int]map[string]string{8200: {
+		"mp0": "local-lvm:vm-8200-disk-1,mp=/var/lib/felhom,backup=1,size=100G",
+		"mp8": "/mnt/felhom-drives,mp=/mnt/felhom-drives",
+	}}}
+	dir := t.TempDir()
+	file := filepath.Join(dir, InternalPlaceFile)
+	if err := WriteInternalPlace(file, InternalPlaceRecord{FSUUID: ipUUID}); err != nil {
+		t.Fatal(err)
+	}
+	srv, err := NewServer(Options{
+		ListenAddr: "127.0.0.1:0", Guests: fg, Backups: &fakeBackups{}, Store: &fakeStore{},
+		Storage: fakeStorage{targets: []hub.StorageTarget{provenSSD}}, Tokens: staticTokens{"A": 8200},
+		HostReader: fakeHostReader{mounts: ih.mounts}, Identity: &storage.IdentityResolver{Host: ih},
+		InternalPlaceFile: file, Logger: slog.New(slog.NewTextHandler(io.Discard, nil)),
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	srv.baseCtx = context.Background()
+	got, code := sameDisk(t, srv.Handler(), "/mnt/felhom-drives/belso/felhom-data/jellyfin", "/mnt/sys_drive")
+	if code != http.StatusOK || got.Different || !got.Proven || got.Reason != "same_disk" {
+		t.Fatalf("internal place vs sys_drive: want same disk, proven; got %d %+v", code, got)
+	}
+	if got.A.Kind != storage.PlaceKindInternal {
+		t.Fatalf("kind: want %q, got %q", storage.PlaceKindInternal, got.A.Kind)
+	}
+}
+
+func TestInternalPlaceRecord_RefusesANonUUID(t *testing.T) {
+	file := filepath.Join(t.TempDir(), InternalPlaceFile)
+	if err := WriteInternalPlace(file, InternalPlaceRecord{FSUUID: "../../etc"}); err == nil {
+		t.Fatal("a non-UUID must not be written")
+	}
+}
+
+// The root helper's own tests (configs/test_felhom_internal_place.py — the ruled size table, never formatting what
+// holds anything) run with the Go suite.
+func TestInternalPlaceHelperPythonSuite(t *testing.T) {
+	py, err := exec.LookPath("python3")
+	if err != nil {
+		t.Skip("python3 not found")
+	}
+	_, file, _, _ := runtime.Caller(0)
+	script := filepath.Join(filepath.Dir(file), "..", "..", "configs", "test_felhom_internal_place.py")
+	out, err := exec.Command(py, script).CombinedOutput()
+	if err != nil {
+		t.Fatalf("helper suite failed: %v\n%s", err, out)
+	}
+}

internal/localapi/samedisk.go

@@ -0,0 +1,202 @@
+package localapi
+
+import (
+	"context"
+	"net/http"
+	"path"
+	"regexp"
+	"strings"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/storage"
+)
+
+// POST /disks/same-disk (v0.157.0) — felhom.eu architecture/16-storage-physical-identity.md §4.
+//
+// The controller asks whether two of ITS paths sit on the same physical disk; the agent answers from its
+// own /sys + udev read, never from the caller's claim (`16` §10). A guest path is mapped only to a place
+// the agent itself made:
+//   - /mnt/felhom-drives/<name>[/…] → the host mount at exactly /mnt/felhom-drives/<name> (the drive's
+//     felhom-data bind, or a network share's mount). A path whose <name> is not mounted there (an absent
+//     drive: the bare directory on the root disk) is UNMAPPED — never "the system disk".
+//   - /mnt/sys_drive, /var/lib/felhom, /var/lib/docker [/…] → the guest's data volume (mp0, mp=/var/lib/felhom;
+//     or a legacy mp at that exact path) → the PVE storage holding it → that storage's identity.
+//   - anything else → unmapped.
+// Answer: {different, proven, reason, a, b}. Act on `proven` first: different+proven = a counted second
+// copy; same_disk+proven = not a second copy; proven=false = unknown (reason vm|network|usb_bridge|
+// weak_wwn|no_serial|unmapped). Pinned by samedisk_test.go.
+
+type sameDiskRequest struct {
+	VMID int    `json:"vmid,omitempty"`
+	A    string `json:"a"`
+	B    string `json:"b"`
+}
+
+// SamePlace is one side of the answer.
+type SamePlace struct {
+	Path          string   `json:"path"`
+	Mapped        bool     `json:"mapped"`
+	Kind          string   `json:"kind,omitempty"`
+	PhysicalDisks []string `json:"physical_disks,omitempty"`
+	Identity      string   `json:"identity,omitempty"`
+	Reason        string   `json:"reason,omitempty"`
+}
+
+// SameDiskResponse is the wire answer.
+type SameDiskResponse struct {
+	Different bool      `json:"different"`
+	Proven    bool      `json:"proven"`
+	Reason    string    `json:"reason,omitempty"`
+	A         SamePlace `json:"a"`
+	B         SamePlace `json:"b"`
+}
+
+// sysDataPaths are the guest paths that live on the guest's one data volume (golden v3: mp0 at
+// /var/lib/felhom, its docker/ and sys_drive/ folders bound in the guest — 07 §7.5.1).
+var sysDataPaths = []string{"/var/lib/felhom", "/mnt/sys_drive", "/var/lib/docker"}
+
+func (s *Server) handleSameDisk(w http.ResponseWriter, r *http.Request, vmid int) {
+	if s.ident == nil {
+		writeErr(w, http.StatusServiceUnavailable, "disk identity not configured on this host")
+		return
+	}
+	var req sameDiskRequest
+	if !decodeBody(w, r, &req) {
+		return
+	}
+	if !s.scopedFromBody(w, req.VMID, vmid, r.URL.Path) {
+		return
+	}
+	if !path.IsAbs(req.A) || !path.IsAbs(req.B) {
+		writeErr(w, http.StatusBadRequest, "a and b must be absolute guest paths")
+		return
+	}
+	a := s.mapGuestPlace(r.Context(), vmid, req.A)
+	b := s.mapGuestPlace(r.Context(), vmid, req.B)
+	resp := SameDiskResponse{A: a.view, B: b.view}
+	resp.Different, resp.Proven, resp.Reason = storage.Relation(a.id, b.id)
+	s.logger.Info("local-api: same-disk", "vmid", vmid, "a", req.A, "b", req.B,
+		"different", resp.Different, "proven", resp.Proven, "reason", resp.Reason)
+	writeOK(w, resp)
+}
+
+type mappedPlace struct {
+	view SamePlace
+	id   storage.PlaceIdentity
+}
+
+func (s *Server) mapGuestPlace(ctx context.Context, vmid int, guestPath string) mappedPlace {
+	p := path.Clean(guestPath)
+	out := mappedPlace{view: SamePlace{Path: p}}
+	set := func(kind string, id storage.PlaceIdentity) mappedPlace {
+		if !id.Computed() {
+			return out
+		}
+		out.id = id
+		out.view.Mapped, out.view.Kind = true, kind
+		out.view.PhysicalDisks, out.view.Identity, out.view.Reason = id.PhysicalDisks, id.Identity, id.Reason
+		return out
+	}
+	if strings.HasPrefix(p, StableParentDir+"/") {
+		name := strings.SplitN(strings.TrimPrefix(p, StableParentDir+"/"), "/", 2)[0]
+		if id, kind, ok := s.ownMountIdentity(StableParentDir + "/" + name); ok {
+			return set(kind, id)
+		}
+		return out
+	}
+	for _, root := range sysDataPaths {
+		if p != root && !strings.HasPrefix(p, root+"/") {
+			continue
+		}
+		id := s.guestDataVolumeIdentity(ctx, vmid, root)
+		return set(storage.PlaceKindSystem, id)
+	}
+	return out
+}
+
+// ownMountIdentity judges a HOST path only when it is ITSELF a mountpoint (an enrolled drive's raw mount
+// or its felhom-data bind, a network share). A path that is not its own mount — an absent drive, whose
+// directory now sits on the root disk — is not computed, never root's identity (the R-118 trap).
+func (s *Server) ownMountIdentity(hostPath string) (storage.PlaceIdentity, string, bool) {
+	if s.ident == nil || hostPath == "" {
+		return storage.PlaceIdentity{}, "", false
+	}
+	mounts, err := s.hostReader().Mounts()
+	if err != nil {
+		return storage.PlaceIdentity{}, "", false
+	}
+	m, ok := storage.MountFor(mounts, hostPath)
+	if !ok || path.Clean(m.MountPoint) != path.Clean(hostPath) {
+		return storage.PlaceIdentity{}, "", false
+	}
+	id := s.ident.HostPath(hostPath)
+	if !id.Computed() {
+		return id, "", false
+	}
+	kind := storage.PlaceKindDisk
+	switch {
+	case s.isInternalPlace(hostPath): // v0.158.0 `16` §5: the one-disk box's data place, on the system disk
+		kind = storage.PlaceKindInternal
+	case id.Reason == storage.ReasonNetwork:
+		kind = storage.PlaceKindNetwork
+	case strings.HasPrefix(m.Device, "/dev/") && !strings.HasPrefix(m.Device, "/dev/mapper/") && isPartitionPath(m.Device):
+		kind = storage.PlaceKindPartition
+	}
+	return id, kind, true
+}
+
+var rePartitionName = regexp.MustCompile(`^(nvme[0-9]+n[0-9]+p[0-9]+|(sd|vd|hd)[a-z]+[0-9]+)$`)
+
+func isPartitionPath(dev string) bool { return rePartitionName.MatchString(path.Base(dev)) }
+
+// guestDataVolumeIdentity finds the guest mountpoint that holds root (mp=/var/lib/felhom on golden v3, or a
+// legacy mp at that exact path) and returns the identity of the PVE storage that holds its volume.
+func (s *Server) guestDataVolumeIdentity(ctx context.Context, vmid int, root string) storage.PlaceIdentity {
+	if s.guests == nil || s.storage == nil {
+		return storage.PlaceIdentity{}
+	}
+	cfg, err := s.guests.GuestConfig(ctx, vmid)
+	if err != nil {
+		return storage.PlaceIdentity{}
+	}
+	storeID, fallback := "", ""
+	for _, spec := range cfg.MountPoints() {
+		st, mp, _ := parseMount(spec)
+		if strings.HasPrefix(st, "/") {
+			continue // a host bind (mp8, mp9), not a volume
+		}
+		switch mp {
+		case root:
+			storeID = st // a legacy mp at exactly this path wins
+		case "/var/lib/felhom":
+			fallback = st // golden v3: the one data volume
+		}
+	}
+	if storeID == "" {
+		storeID = fallback
+	}
+	if storeID == "" {
+		return storage.PlaceIdentity{}
+	}
+	targets, err := s.storage.Observe(ctx)
+	if err != nil {
+		return storage.PlaceIdentity{}
+	}
+	for _, t := range targets {
+		if t.Name == storeID {
+			return identityOfTarget(t)
+		}
+	}
+	return storage.PlaceIdentity{}
+}
+
+func identityOfTarget(t hub.StorageTarget) storage.PlaceIdentity {
+	if t.Identity == "" {
+		return storage.PlaceIdentity{}
+	}
+	disks := t.PhysicalDisks
+	if disks == nil {
+		disks = []string{}
+	}
+	return storage.PlaceIdentity{PhysicalDisks: disks, Identity: t.Identity, Reason: t.IdentityReason}
+}

internal/localapi/samedisk_test.go

@@ -0,0 +1,146 @@
+package localapi
+
+import (
+	"context"
+	"encoding/json"
+	"io"
+	"log/slog"
+	"net/http"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/storage"
+)
+
+// POST /disks/same-disk (`16` §4) — the controller's Tier-2 question, answered from the agent's own read.
+
+type fakeIdentHost struct {
+	mounts []storage.Mount
+	parent map[string]string
+	props  map[string]map[string]string
+	virt   string
+}
+
+func (f *fakeIdentHost) Mounts() ([]storage.Mount, error) { return f.mounts, nil }
+func (f *fakeIdentHost) KernelName(dev string) (string, bool) {
+	if len(dev) > 5 && dev[:5] == "/dev/" {
+		return dev[5:], true
+	}
+	return "", false
+}
+func (f *fakeIdentHost) Parent(n string) (string, bool)        { p, ok := f.parent[n]; return p, ok }
+func (f *fakeIdentHost) Slaves(string) []string                { return nil }
+func (f *fakeIdentHost) LoopBackingFile(string) (string, bool) { return "", false }
+func (f *fakeIdentHost) UdevProps(n string) map[string]string  { return f.props[n] }
+func (f *fakeIdentHost) Virt() string                          { return f.virt }
+
+// sameDiskRig: guest 8200 (token A) with mp0 on local-lvm (the system SSD, proven) + the given mounts.
+func sameDiskRig(t *testing.T, host *fakeIdentHost, localLVM hub.StorageTarget) http.Handler {
+	t.Helper()
+	fg := &fakeGuestsCfg{mounts: map[int]map[string]string{8200: {
+		"mp0": "local-lvm:vm-8200-disk-1,mp=/var/lib/felhom,backup=1,size=70G",
+		"mp8": "/mnt/felhom-drives,mp=/mnt/felhom-drives",
+	}}}
+	srv, err := NewServer(Options{
+		ListenAddr: "127.0.0.1:0", Guests: fg, Backups: &fakeBackups{}, Store: &fakeStore{},
+		Storage: fakeStorage{targets: []hub.StorageTarget{localLVM}}, Tokens: staticTokens{"A": 8200},
+		HostReader: fakeHostReader{mounts: host.mounts},
+		Identity:   &storage.IdentityResolver{Host: host},
+		Logger:     slog.New(slog.NewTextHandler(io.Discard, nil)),
+	})
+	if err != nil {
+		t.Fatal(err)
+	}
+	srv.baseCtx = context.Background()
+	return srv.Handler()
+}
+
+var provenSSD = hub.StorageTarget{Name: "local-lvm", Type: hub.StorageTypeLVMThin, Kind: "system",
+	PhysicalDisks: []string{"serial:ata:SanDisk:1"}, Identity: "proven"}
+
+func baseHost() *fakeIdentHost {
+	return &fakeIdentHost{
+		virt:   "none",
+		parent: map[string]string{"sdc1": "sdc", "sdc2": "sdc"},
+		props: map[string]map[string]string{
+			"sdb": {"ID_BUS": "ata", "ID_MODEL": "TOSHIBA", "ID_SERIAL_SHORT": "65NOP3HDT"},
+			"sdc": {"ID_BUS": "usb", "ID_MODEL": "Portable", "ID_SERIAL_SHORT": "NAAB7C21"},
+		},
+		mounts: []storage.Mount{
+			{Device: "/dev/mapper/pve-root", MountPoint: "/", FSType: "ext4"},
+			{Device: "/dev/mapper/pve-root", MountPoint: "/mnt/felhom-drives", FSType: "ext4"},
+			{Device: "/dev/sdb", MountPoint: "/mnt/felhom-drives/hdd_1", FSType: "ext4"},
+			{Device: "/dev/sdc1", MountPoint: "/mnt/felhom-drives/a", FSType: "ext4"},
+			{Device: "/dev/sdc2", MountPoint: "/mnt/felhom-drives/b", FSType: "ext4"},
+			{Device: "//nas/films", MountPoint: "/mnt/felhom-drives/nas", FSType: "cifs"},
+		},
+	}
+}
+
+func sameDisk(t *testing.T, h http.Handler, a, b string) (SameDiskResponse, int) {
+	t.Helper()
+	body, _ := json.Marshal(map[string]string{"a": a, "b": b})
+	w := do(t, h, "POST", "/disks/same-disk", "A", string(body))
+	var env struct {
+		Data SameDiskResponse `json:"data"`
+	}
+	_ = json.Unmarshal(w.Body.Bytes(), &env)
+	return env.Data, w.Code
+}
+
+func TestSameDisk_DriveVsSystemData_DifferentProven(t *testing.T) {
+	h := sameDiskRig(t, baseHost(), provenSSD)
+	got, code := sameDisk(t, h, "/mnt/felhom-drives/hdd_1/felhom-data/paperless", "/mnt/sys_drive")
+	if code != http.StatusOK || !got.Different || !got.Proven || got.Reason != "" {
+		t.Fatalf("drive vs sys_drive: want different+proven, got %d %+v", code, got)
+	}
+	if got.B.Kind != "system" || got.A.Kind != "disk" {
+		t.Fatalf("kinds: got a=%q b=%q", got.A.Kind, got.B.Kind)
+	}
+}
+
+func TestSameDisk_TwoPartitionsOfOneDisk_SameProven(t *testing.T) {
+	h := sameDiskRig(t, baseHost(), provenSSD)
+	got, _ := sameDisk(t, h, "/mnt/felhom-drives/a", "/mnt/felhom-drives/b/x")
+	if got.Different || !got.Proven || got.Reason != "same_disk" || got.A.Kind != "partition" {
+		t.Fatalf("two partitions: want same+proven, got %+v", got)
+	}
+}
+
+func TestSameDisk_InsideAVM_NotProven(t *testing.T) {
+	host := baseHost()
+	host.virt = "kvm"
+	h := sameDiskRig(t, host, hub.StorageTarget{Name: "local-lvm", Identity: "unknown", IdentityReason: "vm",
+		PhysicalDisks: []string{"serial:scsi:QEMU_HARDDISK:drive-scsi0"}})
+	got, _ := sameDisk(t, h, "/mnt/felhom-drives/hdd_1", "/mnt/sys_drive")
+	if got.Proven || got.Reason != "vm" {
+		t.Fatalf("inside a VM: want not proven (vm), got %+v", got)
+	}
+}
+
+func TestSameDisk_NetworkAndUnmapped(t *testing.T) {
+	h := sameDiskRig(t, baseHost(), provenSSD)
+	if got, _ := sameDisk(t, h, "/mnt/felhom-drives/nas", "/mnt/sys_drive"); got.Proven || got.Reason != "network" {
+		t.Fatalf("a NAS: want not proven (network), got %+v", got)
+	}
+	// An ABSENT drive: its directory sits on the root disk — must be unmapped, never "the system disk".
+	if got, _ := sameDisk(t, h, "/mnt/felhom-drives/gone", "/mnt/sys_drive"); got.Proven || got.Reason != "unmapped" || got.A.Mapped {
+		t.Fatalf("an absent drive: want unmapped, got %+v", got)
+	}
+	if got, _ := sameDisk(t, h, "/opt/elsewhere", "/mnt/sys_drive"); got.Proven || got.Reason != "unmapped" {
+		t.Fatalf("an unknown guest path: want unmapped, got %+v", got)
+	}
+}
+
+func TestSameDisk_ScopedAndValidated(t *testing.T) {
+	h := sameDiskRig(t, baseHost(), provenSSD)
+	if w := do(t, h, "POST", "/disks/same-disk", "A", `{"vmid":9300,"a":"/mnt/sys_drive","b":"/mnt/sys_drive"}`); w.Code != http.StatusForbidden {
+		t.Fatalf("another guest's vmid: got %d want 403", w.Code)
+	}
+	if w := do(t, h, "POST", "/disks/same-disk", "A", `{"a":"relative","b":"/mnt/sys_drive"}`); w.Code != http.StatusBadRequest {
+		t.Fatalf("a relative path: got %d want 400", w.Code)
+	}
+	if w := do(t, h, "POST", "/disks/same-disk", "", `{"a":"/x","b":"/y"}`); w.Code != http.StatusUnauthorized {
+		t.Fatalf("no token: got %d want 401", w.Code)
+	}
+}

internal/localapi/server.go

@@ -90,6 +90,10 @@ type BackupStore interface {
 type LastKnownBackupStore interface {
 	RecordBackupSuccess(target string, b hub.Backup) error
 	LastKnownSuccess(target string, vmid int) (time.Time, bool)
+	// R-973: the cadence cap's restart-proof history and its refusals.
+	RecentBackupSuccesses(target string, vmid int, since time.Time) []time.Time
+	RecordBackupRefusal(target string, vmid int, at time.Time) error
+	LastBackupRefusal() (target string, vmid int, at time.Time, ok bool)
 }
 
 // StorageView yields the host's observed storage targets (for mapping a mount's storage id →
@@ -208,6 +212,9 @@ type Options struct {
 	// Intent records drive enroll/eject intent for the self-heal watchdog (slice 10 P3). OPTIONAL —
 	// when nil, no intent is recorded (self-heal runs ungated).
 	Intent IntentRecorder
+	// InternalPlaceFile (v0.158.0, felhom.eu `16` §5) is the path of the internal data place record
+	// (InternalPlaceRecord). OPTIONAL — "" means this host has no internal place and EnsureInternalPlace is a no-op.
+	InternalPlaceFile string
 	// GuestBinds persists which user-data drives (by durable-id) are enrolled into each guest, so the
 	// startup re-assert (ReassertGuestBinds) can restore a bind that a re-provision dropped (F9).
 	// OPTIONAL — when nil, guest binds are not recorded and the startup re-assert is a no-op.
@@ -220,6 +227,10 @@ type Options struct {
 	// ROLE (it backs SystemDisks for the eject role-gate + the /disks role hints). OPTIONAL — when nil
 	// it defaults to the production *storage.ProcHostReader. Injectable so the role-gate is testable.
 	HostReader storage.HostReader
+	// Identity computes physical-disk identity (felhom.eu architecture/16-storage-physical-identity.md §3)
+	// for the /disks rows and POST /disks/same-disk. OPTIONAL — when nil, /disks carries no identity and
+	// same-disk answers 503 "not configured" (the controller then keeps its old st_dev rule).
+	Identity *storage.IdentityResolver
 	// HostMetrics serves GET /host/metrics (slice 9) — host-wide health (cpu%/mem/load/uptime/
 	// cpu-temp) + per-storage capacity, host-wide and token-authed (one-customer-per-host). When
 	// nil the endpoint reports "not configured" (host still reports/reconciles).
@@ -295,6 +306,11 @@ type Server struct {
 	logger             *slog.Logger
 	now                func() time.Time
 
+	// R-973 cadence cap: this process's successful runs per tier+guest, and the newest refusal.
+	capMu      sync.Mutex
+	capRuns    map[backupJobKey][]time.Time
+	capRefusal capRefusal
+
 	disks           DiskOps           // slice 8C (optional)
 	diskGate        StorageGate       // slice 8C (optional)
 	guestList       GuestLister       // slice 8C (optional)
@@ -312,11 +328,14 @@ type Server struct {
 	// identity blob from the hub, unseal it with the customer's recovery code, return ONLY the
 	// offsite repository password. OPTIONAL — nil (no hub client configured) makes
 	// POST /escrow/recover-offsite-password answer 503 rather than pretending.
-	escrowRecovery EscrowRecoverer
-	intent         IntentRecorder      // slice 10 P3 (optional)
-	guestBinds     *GuestBindStore     // F9 startup bind re-assert record (optional)
-	formatJobs     *FormatJobStore     // F20-BUG3 detached-format job record (optional)
-	staleLock      StaleLockController // F2-b startup stale-lock recovery (optional)
+	escrowRecovery    EscrowRecoverer
+	intent            IntentRecorder                      // slice 10 P3 (optional)
+	internalPlaceFile string                              // v0.158.0 `16` §5 (optional)
+	ipState           internalPlaceState                  // EnsureInternalPlace's memo
+	guestBinds        *GuestBindStore                     // F9 startup bind re-assert record (optional)
+	formatJobs        *FormatJobStore                     // F20-BUG3 detached-format job record (optional)
+	staleLock         StaleLockController                 // F2-b startup stale-lock recovery (optional)
+	partialSweep      func(ctx context.Context, vmid int) // R-1006 start-up sweep of dead vzdump leftovers (optional)
 	// guestPower (F-REBOOT) is per-guest start-attempt state for the guest-power watchdog.
 	// Guarded by guestPowerMu in guestpower.go; in-memory on purpose (see guestPowerState).
 	guestPower map[int]guestPowerState
@@ -324,7 +343,8 @@ type Server struct {
 	// guestPowerSweeps counts completed guest-power sweeps, for the liveness observable. Touched only
 	// from GuestPowerTick, which the ticker calls serially.
 	guestPowerSweeps int
-	host             storage.HostReader // role classification source (optional; defaults to ProcHostReader)
+	host             storage.HostReader        // role classification source (optional; defaults to ProcHostReader)
+	ident            *storage.IdentityResolver // `16` §3 physical-disk identity (optional)
 
 	hostMetrics  HostMetricsProvider // slice 9 (optional)
 	hostID       string              // slice 10B: for the data-bearing-format pending-op hint
@@ -415,9 +435,18 @@ type Server struct {
 	escrowMu       sync.Mutex
 	escrowJob      *escrowCeremonyJob
 	escrowR        []byte
+	escrowPending  []byte // R-978: the ceremony's hub upload body, sent by the claim before R is handed out
+	escrowClaiming bool   // R-978: a claim is uploading escrowPending (TTL and a new start wait)
 	escrowRClaimed bool
 	escrowRExpiry  time.Time
-	escrowDone     <-chan struct{} // closes when the detached job finishes (tests wait on it)
+	// R-1151: after a claim that asked to wait for the typed-back words (defer_upload), the package stays in
+	// escrowPending until POST /escrow/ceremony/confirm brings the two words at escrowConfirmPos back. Only those two
+	// words are kept (never R), zeroed on confirm, supersede and expiry.
+	escrowAwaitConfirm  bool
+	escrowConfirmWords  [2][]byte
+	escrowConfirmPos    [2]int
+	escrowConfirmExpiry time.Time
+	escrowDone          <-chan struct{} // closes when the detached job finishes (tests wait on it)
 	// ceremonyRun executes the fixed-argv sudo self-invocation (tests inject canned JSON).
 	ceremonyRun ceremonyRunner
 	privileged  PrivilegedRunner
@@ -449,42 +478,44 @@ func NewServer(o Options) (*Server, error) {
 		cadence = defaultBackupCadence
 	}
 	s := &Server{
-		addr:            o.ListenAddr,
-		cert:            o.Cert,
-		guests:          o.Guests,
-		backups:         o.Backups,
-		store:           o.Store,
-		storage:         o.Storage,
-		driveTargets:    o.DriveTargets,
-		smart:           o.Smart,
-		tokens:          o.Tokens,
-		cadence:         cadence,
-		logger:          o.Logger,
-		now:             func() time.Time { return time.Now().UTC() },
-		disks:           o.Disks,
-		diskGate:        o.DiskGate,
-		guestList:       o.Guests2,
-		guestAttach:     o.GuestAttach,
-		mem:             o.Memory,
-		netStorage:      o.NetStorage,
-		privileged:      o.Privileged,
-		configPath:      o.ConfigPath,
-		stateDir:        o.StateDir,
-		netMountRoot:    storage.NetworkMountRoot,
-		smbCredsDir:     o.SmbCredsDir,
-		escrowStagePath: o.EscrowStagePath,
-		escrowRecovery:  o.EscrowRecovery,
-		intent:          o.Intent,
-		guestBinds:      o.GuestBinds,
-		formatJobs:      o.FormatJobs,
-		staleLock:       o.StaleLock,
-		host:            o.HostReader,
-		hostMetrics:     o.HostMetrics,
-		hostID:          o.HostID,
-		agentVersion:    o.AgentVersion,
-		logRing:         o.LogRing,
-		jobs:            map[backupJobKey]*backupJob{},
-		swapInFlight:    map[int]bool{},
+		addr:              o.ListenAddr,
+		cert:              o.Cert,
+		guests:            o.Guests,
+		backups:           o.Backups,
+		store:             o.Store,
+		storage:           o.Storage,
+		driveTargets:      o.DriveTargets,
+		smart:             o.Smart,
+		tokens:            o.Tokens,
+		cadence:           cadence,
+		logger:            o.Logger,
+		now:               func() time.Time { return time.Now().UTC() },
+		disks:             o.Disks,
+		diskGate:          o.DiskGate,
+		guestList:         o.Guests2,
+		guestAttach:       o.GuestAttach,
+		mem:               o.Memory,
+		netStorage:        o.NetStorage,
+		privileged:        o.Privileged,
+		configPath:        o.ConfigPath,
+		stateDir:          o.StateDir,
+		netMountRoot:      storage.NetworkMountRoot,
+		smbCredsDir:       o.SmbCredsDir,
+		escrowStagePath:   o.EscrowStagePath,
+		escrowRecovery:    o.EscrowRecovery,
+		intent:            o.Intent,
+		internalPlaceFile: o.InternalPlaceFile,
+		guestBinds:        o.GuestBinds,
+		formatJobs:        o.FormatJobs,
+		staleLock:         o.StaleLock,
+		host:              o.HostReader,
+		ident:             o.Identity,
+		hostMetrics:       o.HostMetrics,
+		hostID:            o.HostID,
+		agentVersion:      o.AgentVersion,
+		logRing:           o.LogRing,
+		jobs:              map[backupJobKey]*backupJob{},
+		swapInFlight:      map[int]bool{},
 	}
 	// R-82 tier resolution. Options.BackupTiers is authoritative when supplied; otherwise ONE tier
 	// is synthesized from Backups + BackupCadence — the pre-R-82 shape, so every existing caller
@@ -545,6 +576,7 @@ func (s *Server) Handler() http.Handler {
 	// Disk management (slice 8C) — self-scoped; format routes through the data-bearing classifier+gate.
 	mux.HandleFunc("GET /disks", s.withGuest(s.handleDisks))
 	mux.HandleFunc("GET /disks/candidates", s.withGuest(s.handleDiskCandidates))
+	mux.HandleFunc("POST /disks/same-disk", s.withGuest(s.handleSameDisk)) // `16` §4 (v0.157.0)
 	mux.HandleFunc("POST /disks/assign", s.withGuest(s.handleDiskAssign))
 	mux.HandleFunc("POST /disks/eject", s.withGuest(s.handleDiskEject))
 	mux.HandleFunc("POST /disks/decommission", s.withGuest(s.handleDiskDecommission))
@@ -589,6 +621,8 @@ func (s *Server) Handler() http.Handler {
 	mux.HandleFunc("POST /escrow/ceremony", s.withGuest(s.handleEscrowCeremonyStart))
 	mux.HandleFunc("GET /escrow/ceremony/status", s.withGuest(s.handleEscrowCeremonyStatus))
 	mux.HandleFunc("POST /escrow/ceremony/claim", s.withGuest(s.handleEscrowCeremonyClaim))
+	// R-1151: the household typed two words of the new code back — only now does its package reach the hub.
+	mux.HandleFunc("POST /escrow/ceremony/confirm", s.withGuest(s.handleEscrowCeremonyConfirm))
 
 	// v0.83.0 observability: the agent's always-DEBUG capture ring, for the controller's
 	// Debug page agent tab (same auth/self-scoping wrap as every sibling route).
@@ -864,6 +898,12 @@ func (s *Server) handleBackup(w http.ResponseWriter, r *http.Request, vmid int)
 			"a backup is already in flight on target "+busyTarget+" (job "+busyJob+") — only one backup runs at a time per guest")
 		return
 	}
+	// R-973: the cadence cap (backup_cap.go). Checked under jobsMu, after the in-flight answers, so two
+	// requests cannot both pass it (only one run per guest is ever in flight).
+	if s.capRefuses(w, key, tier) {
+		s.jobsMu.Unlock()
+		return
+	}
 	// Job ids must be unique PER TIER, and by construction rather than by clock luck: two tiers
 	// started inside the same nanosecond (the weekly both-due night, or any injected clock) would
 	// otherwise collide and hand the second caller the first tier's id. The PRIMARY keeps the
@@ -927,6 +967,9 @@ func (s *Server) handleBackup(w http.ResponseWriter, r *http.Request, vmid int)
 			s.logger.Info("local-api: backup job complete", "vmid", vmid, "target", tier.TargetID, "job", jobID, "archive", b.Archive)
 		}
 		s.store.RecordBackup(b)
+		if b.Success {
+			s.recordCapRun(key, b) // R-973
+		}
 		if b.Success && s.lastKnown != nil {
 			if err := s.lastKnown.RecordBackupSuccess(tier.TargetID, b); err != nil {
 				// Not fatal: the backup exists. Only the fallback after a restart loses this copy.
@@ -1300,6 +1343,20 @@ type BackupStatusResponse struct {
 	// presence. Served on the UNTARGETED request only; additive, so an older controller reads the
 	// response exactly as before.
 	Tiers []TierBackupState `json:"tiers,omitempty"`
+	// HostBusy (R-921, agent Unreleased 2026-10-19) says whether the host-wide heavy-operation gate is held RIGHT NOW
+	// (another guest's backup, the restore test, a night OS step, the stale-lock sweep, a power-off) — a POST /backup
+	// would then be refused with 409. The controller reads it BEFORE it stops the household's apps, so a refused tier
+	// costs no stop. Served on every request; ABSENT (nil) = an older agent, or the gate is not wired — the controller
+	// then keeps its old order (stop, ask, resume at once on 409). A read, never a reservation: the gate can be taken
+	// between this answer and the POST, and that race keeps the old order. Pinned by TestBackupStatus_R921_HostBusy.
+	HostBusy *HostBusyState `json:"host_busy,omitempty"`
+}
+
+// HostBusyState is GET /backup/status's host_busy (R-921). What names the holder ("backup:local", "restore-test",
+// "os-pve-step" …), empty when Busy is false.
+type HostBusyState struct {
+	Busy bool   `json:"busy"`
+	What string `json:"what,omitempty"`
 }
 
 func (s *Server) handleBackupStatus(w http.ResponseWriter, r *http.Request, vmid int) {
@@ -1319,6 +1376,10 @@ func (s *Server) handleBackupStatus(w http.ResponseWriter, r *http.Request, vmid
 		resp.JobID = job.JobID
 		resp.Error = job.Error
 	}
+	if s.inFlight != nil {
+		what := s.inFlight.Busy()
+		resp.HostBusy = &HostBusyState{Busy: what != "", What: what}
+	}
 	writeOK(w, resp)
 }
 

internal/localapi/server_test.go

@@ -13,6 +13,7 @@ import (
 	"testing"
 	"time"
 
+	"gitea.dooplex.hu/admin/felhom-agent/internal/backup"
 	"gitea.dooplex.hu/admin/felhom-agent/internal/hub"
 	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
 )
@@ -514,3 +515,47 @@ func TestBackupStatus_FiltersToThisGuest(t *testing.T) {
 		t.Fatalf("status not scoped to this guest / not latest: %s", w.Body.String())
 	}
 }
+
+// R-921: GET /backup/status says when the host-wide heavy-operation gate is held, so the controller can skip the app
+// stop for a tier the POST would refuse. Free → busy false; held → busy true with the holder's name; the gate not wired
+// → no field at all (the controller's „older agent" reading, never „free").
+//
+// COMPANION RED-PROOF (observed 2026-10-19): with the `resp.HostBusy = …` assignment removed, the held case fails with
+//
+//	server_test.go: free gate: host_busy = <nil>, want busy=false
+//
+// Restored.
+func TestBackupStatus_R921_HostBusy(t *testing.T) {
+	read := func(h http.Handler) *HostBusyState {
+		w := do(t, h, "GET", "/backup/status", "A", "")
+		var resp struct {
+			Data BackupStatusResponse `json:"data"`
+		}
+		if err := json.Unmarshal(w.Body.Bytes(), &resp); err != nil {
+			t.Fatalf("status body: %v (%s)", err, w.Body.String())
+		}
+		return resp.Data.HostBusy
+	}
+	srv := newTestServerS(t, &fakeGuests{}, &fakeBackups{}, &fakeStore{}, nil)
+	if got := read(srv.Handler()); got != nil {
+		t.Fatalf("gate not wired: host_busy = %+v, want absent", got)
+	}
+	gate := &backup.InFlight{}
+	srv.inFlight = gate
+	h := srv.Handler()
+	if got := read(h); got == nil || got.Busy || got.What != "" {
+		t.Fatalf("free gate: host_busy = %+v, want busy=false", got)
+	}
+	release, _, ok := gate.TryAcquire("os-pve-step")
+	if !ok {
+		t.Fatal("could not take the gate")
+	}
+	if got := read(h); got == nil || !got.Busy || got.What != "os-pve-step" {
+		t.Fatalf("held gate: host_busy = %+v, want busy=true what=\"os-pve-step\"", got)
+	}
+	// The read never takes the gate: a POST after the holder lets go is accepted.
+	release()
+	if got := read(h); got == nil || got.Busy {
+		t.Fatalf("released gate: host_busy = %+v, want busy=false", got)
+	}
+}

internal/localapi/stalelock.go

@@ -2,9 +2,14 @@ package localapi
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"log/slog"
+	"os"
+	"path/filepath"
 	"strconv"
+	"strings"
+	"sync"
 
 	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
 )
@@ -142,6 +147,10 @@ type staleLockController struct {
 	runner proxmox.Runner
 	pool   string       // ownership registry: only members of this PVE pool are ever scanned (A1)
 	logger *slog.Logger // the one success-path scan-summary line; nil = silent
+	// procDir is where taskAlive reads /proc/<pid>/stat (R-1008); "" = /proc. Tests point it at a fixture.
+	procDir  string
+	deadMu   sync.Mutex
+	deadSeen map[string]bool // UPIDs of dead tasks already logged (once per task)
 }
 
 // staleLockAPI is the subset of *proxmox.Client the controller uses (kept narrow for clarity/testing).
@@ -152,6 +161,7 @@ type staleLockAPI interface {
 	ListSnapshots(ctx context.Context, vmid int) ([]proxmox.Snapshot, error)
 	ListRunningTasks(ctx context.Context) ([]proxmox.TaskStatus, error)
 	DeleteSnapshot(ctx context.Context, vmid int, snapname string) (string, error)
+	DeleteSnapshotForce(ctx context.Context, vmid int, snapname string) (string, error)
 	Start(ctx context.Context, vmid int) (string, error)
 	WaitTask(ctx context.Context, upid string, opts proxmox.WaitOptions) (proxmox.TaskStatus, error)
 }
@@ -209,19 +219,100 @@ func (c *staleLockController) Lock(ctx context.Context, vmid int) (string, bool,
 	return cfg.Lock(), cfg.OnBoot(), nil
 }
 
+// BackupRunning is THE answer to „is a vzdump in flight for vmid" for all four callers (startup stale-lock
+// recovery, the guest-power watchdog, the household power-off guard, the controller supervisor): every one of
+// them reaches this method through the StaleLockController interface. R-1008: Proxmox keeps a task in
+// `tasks?source=active` after a power cut — nobody is left to end it (measured on demo-hp 2026-10-14: still
+// listed RUNNING ~30–60 s after the boot) — so a listed task whose worker process is gone is dead, not in flight.
 func (c *staleLockController) BackupRunning(ctx context.Context, vmid int) (bool, error) {
 	tasks, err := c.px.ListRunningTasks(ctx)
 	if err != nil {
 		return false, err
 	}
+	proc := c.procDir
+	if proc == "" {
+		proc = "/proc"
+	}
+	running, dead := vzdumpInFlight(tasks, vmid, proc)
+	for _, t := range dead {
+		c.logDeadOnce(t, vmid)
+	}
+	return running, nil
+}
+
+// vzdumpInFlight decides, from the node's active-task list, whether a vzdump for vmid is in flight: a matching
+// task is in flight while its worker process lives (taskAlive). A task whose worker is gone is dead (returned in dead,
+// for the log). FAIL SAFE on doubt — the old answer: a UPID that does not parse, or a /proc entry that cannot be read,
+// counts as in flight. Pure over procDir; pinned by stalelock_inflight_test.go.
+func vzdumpInFlight(tasks []proxmox.TaskStatus, vmid int, procDir string) (bool, []proxmox.TaskStatus) {
 	id := strconv.Itoa(vmid)
+	var dead []proxmox.TaskStatus
 	for _, t := range tasks {
 		// A single-guest vzdump task carries the vmid in its ID field.
-		if t.Type == "vzdump" && t.ID == id {
+		if t.Type != "vzdump" || t.ID != id {
+			continue
+		}
+		alive, known := taskAlive(procDir, t.UPID)
+		if !known || alive {
 			return true, nil
 		}
+		dead = append(dead, t)
 	}
-	return false, nil
+	return false, dead
+}
+
+// taskAlive is Proxmox's own liveness rule for a task (PVE::ProcFSTools::check_process_running): the UPID names the
+// worker's pid and its start time in clock ticks since boot (pstart); the task lives while /proc/<pid>/stat exists with
+// that same start time (field 22). No wall clock is involved — a clock step after the boot (an NTP correction after a
+// power cut) cannot make a live task look dead (the 2026-10-14 security review of the first, btime-based version).
+// known=false: the UPID does not parse, or /proc/<pid>/stat exists but cannot be read or parsed.
+func taskAlive(procDir, upid string) (alive, known bool) {
+	u, err := proxmox.ParseUPID(upid)
+	if err != nil || u.PID == 0 {
+		return false, false
+	}
+	raw, err := os.ReadFile(filepath.Join(procDir, strconv.FormatUint(u.PID, 10), "stat"))
+	if err != nil {
+		if os.IsNotExist(err) {
+			// „no such process" is only an answer if this agent can see other users' processes at all (a /proc mounted
+			// hidepid=, or a unit with ProtectProc=, hides root's workers — every task would then look dead). PID 1 is
+			// root's; not seeing it means not knowing.
+			if _, err1 := os.Stat(filepath.Join(procDir, "1", "stat")); err1 != nil {
+				return false, false
+			}
+			return false, true // no such process in this boot: the worker is gone
+		}
+		return false, false
+	}
+	st := string(raw)
+	i := strings.LastIndexByte(st, ')') // the comm field may hold spaces and parentheses
+	if i < 0 {
+		return false, false
+	}
+	f := strings.Fields(st[i+1:]) // f[0] is field 3 (state), so field 22 (starttime) is f[19]
+	if len(f) < 20 {
+		return false, false
+	}
+	start, err := strconv.ParseUint(f[19], 10, 64)
+	if err != nil {
+		return false, false
+	}
+	// Proxmox's rule exactly: the start tick matches (a pstart of 0 is not checked) and the process is not a zombie.
+	return (u.PStart == 0 || start == u.PStart) && f[0] != "Z", true
+}
+
+func (c *staleLockController) logDeadOnce(t proxmox.TaskStatus, vmid int) {
+	c.deadMu.Lock()
+	defer c.deadMu.Unlock()
+	if c.deadSeen == nil {
+		c.deadSeen = map[string]bool{}
+	}
+	if c.deadSeen[t.UPID] || c.logger == nil {
+		return
+	}
+	c.deadSeen[t.UPID] = true
+	c.logger.Warn("backup-in-flight: a vzdump task is listed as active but its worker process is gone — it is dead, not running",
+		"vmid", vmid, "upid", t.UPID)
 }
 
 func (c *staleLockController) HasVzdumpSnapshot(ctx context.Context, vmid int) (bool, error) {
@@ -245,8 +336,34 @@ func (c *staleLockController) Unlock(ctx context.Context, vmid int) error {
 	return nil
 }
 
+// DeleteVzdumpSnapshot removes the dangling `vzdump` snapshot. R-1010: when the delete fails ONLY because one disk's
+// snapshot volume is already gone („Failed to find logical volume pve/snap_vm-9201-disk-0_vzdump" — a backup that ran
+// out of space, 2026-10-14), Proxmox stops at that volume and keeps the entry and the other volumes' snapshots, and a
+// plain retry fails the same way every sweep, forever. Then it retries ONCE with Proxmox's force (the hand fix that
+// day was `pct delsnapshot 9201 vzdump --force`): the remaining volumes' snapshots are removed and the entry goes. Any
+// other error is returned as it is — never a force on an error that might mean a volume still holds data (a full
+// LVM metadata area, a busy volume). Pinned by TestStaleLock_R1010_*.
 func (c *staleLockController) DeleteVzdumpSnapshot(ctx context.Context, vmid int) error {
-	upid, err := c.px.DeleteSnapshot(ctx, vmid, vzdumpSnapshotName)
+	err := c.deleteSnap(ctx, vmid, false)
+	if err == nil || !snapshotVolumeAlreadyGone(err) {
+		return err
+	}
+	if c.logger != nil {
+		c.logger.Warn("stale-lock: a disk's vzdump snapshot volume is already gone — retrying the delete once with force (R-1010)",
+			"vmid", vmid, "err", err)
+	}
+	if ferr := c.deleteSnap(ctx, vmid, true); ferr != nil {
+		return fmt.Errorf("forced snapshot delete after %q also failed: %w", err.Error(), ferr)
+	}
+	return nil
+}
+
+func (c *staleLockController) deleteSnap(ctx context.Context, vmid int, force bool) error {
+	del := c.px.DeleteSnapshot
+	if force {
+		del = c.px.DeleteSnapshotForce
+	}
+	upid, err := del(ctx, vmid, vzdumpSnapshotName)
 	if err != nil {
 		return err
 	}
@@ -257,6 +374,23 @@ func (c *staleLockController) DeleteVzdumpSnapshot(ctx context.Context, vmid int
 	return err
 }
 
+// snapshotVolumeAlreadyGone: the error is Proxmox's „Failed to find logical volume …_vzdump" from a snapshot delete
+// (in the task's exit status, its log tail, or the API error text). Narrow on purpose: the force is safe only when the
+// volume it cannot remove does not exist.
+func snapshotVolumeAlreadyGone(err error) bool {
+	text := err.Error()
+	var te *proxmox.TaskError
+	if errors.As(err, &te) {
+		text += "\n" + te.ExitStatus + "\n" + strings.Join(te.LogTail, "\n")
+	}
+	for _, l := range strings.Split(text, "\n") {
+		if strings.Contains(strings.ToLower(l), "failed to find logical volume") && strings.Contains(l, "_"+vzdumpSnapshotName) {
+			return true
+		}
+	}
+	return false
+}
+
 func (c *staleLockController) Start(ctx context.Context, vmid int) error {
 	upid, err := c.px.Start(ctx, vmid)
 	if err != nil {

internal/localapi/stalelock_inflight_test.go

@@ -0,0 +1,266 @@
+package localapi
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"log/slog"
+	"os"
+	"path/filepath"
+	"regexp"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1008: after a power cut Proxmox keeps the dead vzdump task in `tasks?source=active` (measured on demo-hp
+// 2026-10-14: still RUNNING ~30–60 s after the boot). A task whose worker process is gone is dead, not in flight.
+// Liveness is Proxmox's own rule (pid + pstart vs /proc/<pid>/stat field 22), never the wall clock: the first version
+// compared start time with btime and a clock step after the boot could make a live task look dead (security review
+// 2026-10-14). These tests pin the helper, the production controller, three callers end to end; the fourth (the
+// controller supervisor) by the structural pin at the bottom.
+
+// A worker: pid 0x1E6EEC = 1994476, pstart 0x1B4A0D = 1788429 (UPID hex fields).
+const (
+	testPID    = 1994476
+	testPStart = 1788429
+)
+
+func vz(vmid string) proxmox.TaskStatus {
+	return proxmox.TaskStatus{UPID: "UPID:demo-hp:001E6EEC:001B4A0D:6ACF4071:vzdump:" + vmid + ":felhom-agent@pve!agent:",
+		Type: "vzdump", ID: vmid, Status: "RUNNING", StartTime: 1791967345}
+}
+
+// procFixture writes <dir>/<pid>/stat with the given start ticks (comm with a space and a parenthesis, as real ones can).
+func procFixture(t *testing.T, pid int, start int) string {
+	return procFixtureState(t, pid, start, "S")
+}
+
+func procFixtureState(t *testing.T, pid int, start int, state string) string {
+	t.Helper()
+	d := t.TempDir()
+	// PID 1 is always visible on a real host; procFixtureHidden leaves it out
+	if err := os.MkdirAll(filepath.Join(d, "1"), 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if err := os.WriteFile(filepath.Join(d, "1", "stat"), []byte("1 (systemd) S 0 1 1 0 -1 0 0 0 0 0 0 0 0 0 20 0 1 0 2 0 0\n"), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	if pid > 0 {
+		p := filepath.Join(d, itoaT(pid))
+		if err := os.MkdirAll(p, 0o755); err != nil {
+			t.Fatal(err)
+		}
+		f := []string{itoaT(pid), "(task UPID:x (vzdump))", state}
+		for i := 4; i <= 21; i++ {
+			f = append(f, "0")
+		}
+		f = append(f, itoaT(start), "0", "0")
+		if err := os.WriteFile(filepath.Join(p, "stat"), []byte(strings.Join(f, " ")+"\n"), 0o644); err != nil {
+			t.Fatal(err)
+		}
+	}
+	return d
+}
+
+// RED-PROOF: make taskAlive return (true, true) for a missing /proc entry → the dead task runs → fails.
+func TestVzdumpInFlight_WorkerGoneIsDead(t *testing.T) {
+	running, dead := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, procFixture(t, 0, 0))
+	if running || len(dead) != 1 {
+		t.Fatalf("a task whose worker is gone must be dead: running=%v dead=%d", running, len(dead))
+	}
+}
+
+// RED-PROOF: drop the `start == u.PStart` comparison (alive whenever the pid exists) → a reused pid runs → fails.
+func TestVzdumpInFlight_ReusedPidIsDead(t *testing.T) {
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, procFixture(t, testPID, 42)); running {
+		t.Fatal("the pid exists but started at another tick: another process, the task is dead")
+	}
+}
+
+// RED-PROOF: drop the `f[0] != "Z"` test → a zombie worker counts as running → fails.
+func TestVzdumpInFlight_ZombieIsDead(t *testing.T) {
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, procFixtureState(t, testPID, testPStart, "Z")); running {
+		t.Fatal("a zombie worker is not running (Proxmox's own rule)")
+	}
+}
+
+func TestVzdumpInFlight_LiveWorkerRuns(t *testing.T) {
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, procFixture(t, testPID, testPStart)); !running {
+		t.Fatal("the worker lives (same pid, same start tick): in flight")
+	}
+}
+
+// The reason for the rewrite: the wall clock plays no part. A live task whose list start time lies „before the boot"
+// (a clock step after the boot) is still running.
+func TestVzdumpInFlight_ClockStepCannotKillALiveTask(t *testing.T) {
+	tk := vz("9201")
+	tk.StartTime = 1 // absurdly early by the wall clock
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{tk}, 9201, procFixture(t, testPID, testPStart)); !running {
+		t.Fatal("a live worker must count as running whatever its wall-clock start says")
+	}
+}
+
+// RED-PROOF: make an unparsable UPID or unreadable stat „known dead" → fails.
+func TestVzdumpInFlight_DoubtFailsSafe(t *testing.T) {
+	bad := vz("9201")
+	bad.UPID = "garbage"
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{bad}, 9201, procFixture(t, 0, 0)); !running {
+		t.Fatal("an unparsable UPID must count as running (fail safe)")
+	}
+	d := procFixture(t, testPID, testPStart)
+	if err := os.WriteFile(filepath.Join(d, itoaT(testPID), "stat"), []byte("garbage"), 0o644); err != nil {
+		t.Fatal(err)
+	}
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, d); !running {
+		t.Fatal("an unparsable stat must count as running (fail safe)")
+	}
+}
+
+// RED-PROOF: drop the PID-1 visibility check → a hidden /proc makes the live task „gone" → fails.
+func TestVzdumpInFlight_HiddenProcFailsSafe(t *testing.T) {
+	d := t.TempDir() // nothing visible: not even PID 1 (hidepid / ProtectProc)
+	if running, _ := vzdumpInFlight([]proxmox.TaskStatus{vz("9201")}, 9201, d); !running {
+		t.Fatal("a /proc that hides other users' processes must count as running (fail safe)")
+	}
+}
+
+func TestVzdumpInFlight_OtherGuestOrTypeIgnored(t *testing.T) {
+	tasks := []proxmox.TaskStatus{vz("9202"), {UPID: vz("9201").UPID, Type: "vzsnapshot", ID: "9201"}}
+	if running, dead := vzdumpInFlight(tasks, 9201, procFixture(t, testPID, testPStart)); running || len(dead) != 0 {
+		t.Fatalf("other guests' and other types' tasks are not 9201's backup: running=%v dead=%d", running, len(dead))
+	}
+}
+
+func inflightCtrl(api *fakeStaleLockAPI, runner *recordingRunner, logger *slog.Logger, proc string) *staleLockController {
+	c := NewStaleLockController(api, runner, "felhom", logger).(*staleLockController)
+	c.procDir = proc
+	return c
+}
+
+// The production controller answers through the helper, and logs a dead task ONCE.
+// RED-PROOF: make BackupRunning ignore the helper (any listed task runs) → fails.
+func TestStaleLockController_BackupRunningIgnoresDeadTask(t *testing.T) {
+	var buf bytes.Buffer
+	api := &fakeStaleLockAPI{tasks: []proxmox.TaskStatus{vz("9201")}}
+	c := inflightCtrl(api, &recordingRunner{}, slog.New(slog.NewTextHandler(&buf, nil)), procFixture(t, 0, 0))
+	for i := 0; i < 3; i++ {
+		if running, err := c.BackupRunning(context.Background(), 9201); err != nil || running {
+			t.Fatalf("dead task reported running=%v err=%v", running, err)
+		}
+	}
+	if n := strings.Count(buf.String(), "backup-in-flight: a vzdump task is listed as active but its worker process is gone"); n != 1 {
+		t.Fatalf("the dead task must be logged exactly once, got %d", n)
+	}
+	c.procDir = procFixture(t, testPID, testPStart)
+	if running, _ := c.BackupRunning(context.Background(), 9201); !running {
+		t.Fatal("a live worker must be running")
+	}
+}
+
+func fixtureFor(t *testing.T, alive bool) string {
+	if alive {
+		return procFixture(t, testPID, testPStart)
+	}
+	return procFixture(t, 0, 0)
+}
+
+func poolOf(vmids ...int) proxmox.PoolInfo {
+	var p proxmox.PoolInfo
+	raw := `{"members":[`
+	for i, v := range vmids {
+		if i > 0 {
+			raw += ","
+		}
+		raw += `{"vmid":` + itoaT(v) + `,"type":"lxc"}`
+	}
+	raw += `]}`
+	_ = json.Unmarshal([]byte(raw), &p)
+	return p
+}
+
+// Caller 1 — the startup stale-lock recovery clears the lock a power cut left (the night of 2026-10-14 it said
+// „genuinely in-flight — leaving the lock").
+func TestCaller_StartupStaleLockClearsAfterPowerCut(t *testing.T) {
+	for _, tc := range []struct {
+		alive   bool
+		cleared bool
+	}{{false, true}, {true, false}} {
+		api := &fakeStaleLockAPI{tasks: []proxmox.TaskStatus{vz("9201")},
+			cfg: map[int]proxmox.GuestConfig{9201: staleCfg("backup", true)}}
+		runner := &recordingRunner{}
+		s := &Server{staleLock: inflightCtrl(api, runner, nil, fixtureFor(t, tc.alive)), logger: slog.New(slog.NewTextHandler(discardW{}, nil))}
+		s.recoverOneStaleLock(context.Background(), proxmox.Guest{VMID: 9201, Status: "running"})
+		if runner.unlocked("9201") != tc.cleared {
+			t.Fatalf("worker alive=%v: unlocked=%v, want %v", tc.alive, runner.unlocked("9201"), tc.cleared)
+		}
+	}
+}
+
+// Caller 2 — the guest-power watchdog starts a stopped guest despite a dead pre-boot vzdump task.
+func TestCaller_GuestPowerStartsGuestAfterPowerCut(t *testing.T) {
+	for _, tc := range []struct {
+		alive   bool
+		started bool
+	}{{false, true}, {true, false}} {
+		api := &fakeStaleLockAPI{lxc: []proxmox.Guest{{VMID: 9201, Status: "stopped"}}, pool: poolOf(9201),
+			tasks: []proxmox.TaskStatus{vz("9201")}, cfg: map[int]proxmox.GuestConfig{9201: staleCfg("", true)}}
+		s := gpServer(t, inflightCtrl(api, &recordingRunner{}, nil, fixtureFor(t, tc.alive)), nil)
+		s.GuestPowerTick(context.Background())
+		if (len(api.started) == 1) != tc.started {
+			t.Fatalf("worker alive=%v: started=%v, want %v", tc.alive, api.started, tc.started)
+		}
+	}
+}
+
+// Caller 3 — the household power-off guard does not refuse „backup running" for a dead task.
+func TestCaller_PowerOffJobBlockerIgnoresDeadTask(t *testing.T) {
+	for _, tc := range []struct {
+		alive bool
+		want  string
+	}{{false, ""}, {true, PowerOffBackup}} {
+		rig := newPowerRig(t, 0)
+		rig.srv.staleLock = inflightCtrl(&fakeStaleLockAPI{tasks: []proxmox.TaskStatus{vz("9201")}}, &recordingRunner{}, nil, fixtureFor(t, tc.alive))
+		if code, _ := rig.srv.jobBlocker(context.Background(), 9201); code != tc.want {
+			t.Fatalf("worker alive=%v: jobBlocker=%q, want %q", tc.alive, code, tc.want)
+		}
+	}
+}
+
+// Caller 4 (the supervisor), the R-1006 cleanup (VzdumpInFlight) and the rest — structural pin: every non-test BackupRunning call in this package goes through
+// s.staleLock (whose one production implementation is staleLockController, which uses vzdumpInFlight), and no
+// other production type implements BackupRunning. If someone adds a second path, this fails.
+func TestBackupRunning_AllCallersUseTheOneHelper(t *testing.T) {
+	files, _ := filepath.Glob("*.go")
+	call := regexp.MustCompile(`([A-Za-z_.]+)\.BackupRunning\(`)
+	impl := regexp.MustCompile(`func \([a-z]+ \*?([A-Za-z]+)\) BackupRunning\(`)
+	callers := 0
+	for _, f := range files {
+		if strings.HasSuffix(f, "_test.go") {
+			continue
+		}
+		b, err := os.ReadFile(f)
+		if err != nil {
+			t.Fatal(err)
+		}
+		for _, m := range call.FindAllStringSubmatch(string(b), -1) {
+			if m[1] != "s.staleLock" {
+				t.Errorf("%s: BackupRunning called on %q, not s.staleLock", f, m[1])
+			}
+			callers++
+		}
+		for _, m := range impl.FindAllStringSubmatch(string(b), -1) {
+			if m[1] != "staleLockController" {
+				t.Errorf("%s: a second production BackupRunning on %s", f, m[1])
+			}
+		}
+	}
+	if callers != 5 {
+		t.Errorf("expected the five known callers (stale-lock, guest-power, power-off, supervisor, the R-1006 cleanup), found %d", callers)
+	}
+	b, _ := os.ReadFile("stalelock.go")
+	if !strings.Contains(string(b), "vzdumpInFlight(tasks, vmid, proc)") {
+		t.Error("staleLockController.BackupRunning no longer answers through vzdumpInFlight")
+	}
+}

internal/localapi/stalelock_pool_test.go

@@ -30,6 +30,9 @@ type fakeStaleLockAPI struct {
 
 	delsnap []int
 	started []int
+	// R-1010: scripted delete answers, consumed in order (plain and forced share the list); calls records "plain"/"force".
+	delErrs  []error
+	delCalls []string
 }
 
 func (f *fakeStaleLockAPI) ListLXC(context.Context) ([]proxmox.Guest, error) { return f.lxc, nil }
@@ -55,7 +58,22 @@ func (f *fakeStaleLockAPI) ListRunningTasks(context.Context) ([]proxmox.TaskStat
 
 func (f *fakeStaleLockAPI) DeleteSnapshot(_ context.Context, vmid int, snapname string) (string, error) {
 	f.delsnap = append(f.delsnap, vmid)
-	return "", nil // synchronous — no WaitTask
+	return "", f.nextDelErr("plain") // synchronous — no WaitTask
+}
+
+func (f *fakeStaleLockAPI) DeleteSnapshotForce(_ context.Context, vmid int, snapname string) (string, error) {
+	f.delsnap = append(f.delsnap, vmid)
+	return "", f.nextDelErr("force")
+}
+
+func (f *fakeStaleLockAPI) nextDelErr(kind string) error {
+	f.delCalls = append(f.delCalls, kind)
+	if len(f.delErrs) == 0 {
+		return nil
+	}
+	e := f.delErrs[0]
+	f.delErrs = f.delErrs[1:]
+	return e
 }
 
 func (f *fakeStaleLockAPI) Start(_ context.Context, vmid int) (string, error) {

internal/localapi/stalelock_r1010_test.go

@@ -0,0 +1,59 @@
+package localapi
+
+import (
+	"context"
+	"errors"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1010: a half-removed vzdump snapshot. The error text is the one demo-hp's task printed on 2026-10-14
+// (`audits/power-cut-fixes-2026-10-14/C4/hand-fix.txt`).
+const r1010MissingLV = `lvremove snapshot 'pve/snap_vm-9201-disk-0_vzdump' error:   Failed to find logical volume "pve/snap_vm-9201-disk-0_vzdump"`
+
+func r1010TaskErr(exit string) error {
+	return &proxmox.TaskError{UPID: "UPID:demo-hp:0001:0002:0003:vzdelsnapshot:9201:root@pam:", ExitStatus: exit,
+		LogTail: []string{exit}}
+}
+
+// COMPANION RED-PROOF (observed 2026-10-19): with DeleteVzdumpSnapshot back to the single plain delete, the first case
+// fails with
+//
+//	stalelock_r1010_test.go: missing volume: calls [plain], want [plain force]; err = proxmox: task … failed: exitstatus "lvremove snapshot …"
+//
+// Restored.
+func TestStaleLock_R1010_ForcesOnceOnlyWhenTheVolumeIsGone(t *testing.T) {
+	cases := []struct {
+		name      string
+		errs      []error
+		wantCalls []string
+		wantErr   string
+	}{
+		{"missing volume", []error{r1010TaskErr(r1010MissingLV)}, []string{"plain", "force"}, ""},
+		{"missing volume, named only in the log tail", []error{&proxmox.TaskError{UPID: "x", ExitStatus: "command 'lvremove' failed: exit code 5",
+			LogTail: []string{r1010MissingLV}}}, []string{"plain", "force"}, ""},
+		{"the plain delete works", nil, []string{"plain"}, ""},
+		{"another error: no force", []error{r1010TaskErr("lvremove snapshot 'pve/snap_vm-9201-disk-0_vzdump' error: Failed to update pool pve/data.")},
+			[]string{"plain"}, "Failed to update pool"},
+		{"a missing volume of ANOTHER snapshot: no force", []error{r1010TaskErr(`Failed to find logical volume "pve/snap_vm-9201-disk-0_mine"`)},
+			[]string{"plain"}, "_mine"},
+		{"the forced delete fails too: returned, never a second force", []error{r1010TaskErr(r1010MissingLV), errors.New("still broken")},
+			[]string{"plain", "force"}, "still broken"},
+	}
+	for _, c := range cases {
+		api := &fakeStaleLockAPI{delErrs: c.errs}
+		ctrl := NewStaleLockController(api, &recordingRunner{}, "felhom", nil)
+		err := ctrl.DeleteVzdumpSnapshot(context.Background(), 9201)
+		if strings.Join(api.delCalls, " ") != strings.Join(c.wantCalls, " ") {
+			t.Errorf("%s: calls %v, want %v; err = %v", c.name, api.delCalls, c.wantCalls, err)
+		}
+		switch {
+		case c.wantErr == "" && err != nil:
+			t.Errorf("%s: err = %v, want nil", c.name, err)
+		case c.wantErr != "" && (err == nil || !strings.Contains(err.Error(), c.wantErr)):
+			t.Errorf("%s: err = %v, want one naming %q", c.name, err, c.wantErr)
+		}
+	}
+}

internal/localapi/vzpartial_sweep.go

@@ -0,0 +1,33 @@
+package localapi
+
+import "context"
+
+// R-1006: the half-written vzdump cleanup's two needs from the Server, and its start-up sweep. The cleaner itself is
+// internal/vzpartial; the root helper is configs/felhom-vzdump-partial-clean.
+
+// VzdumpInFlight is the one in-flight answer (staleLockController.BackupRunning, R-1008) for the cleanup. No stale-lock
+// controller wired → true: fail safe, nothing is removed.
+func (s *Server) VzdumpInFlight(ctx context.Context, vmid int) (bool, error) {
+	if s.staleLock == nil {
+		return true, nil
+	}
+	return s.staleLock.BackupRunning(ctx, vmid)
+}
+
+// SetPartialSweep sets the start-up sweep: called with every owned guest (the felhom pool, ownership proven).
+func (s *Server) SetPartialSweep(fn func(ctx context.Context, vmid int)) { s.partialSweep = fn }
+
+// SweepVzdumpPartials runs the sweep once for every owned guest. No-op when unwired.
+func (s *Server) SweepVzdumpPartials(ctx context.Context) {
+	if s.partialSweep == nil || s.staleLock == nil {
+		return
+	}
+	guests, err := s.staleLock.Guests(ctx)
+	if err != nil {
+		s.logger.Warn("vzdump-partial: guest list unavailable — skipping the start-up sweep", "err", err)
+		return
+	}
+	for _, g := range guests {
+		s.partialSweep(ctx, g.VMID)
+	}
+}

internal/mgmtplane/mgmtplane.go

@@ -6,12 +6,12 @@
 // the report stanza the hub surfaces:
 //
 //   - /run/sshd present?            — OpenSSH's SHARED privsep dir; its absence is the KEXINIT-reset
-//                                     lockout (SPIKE-felhom-sshd-2026-07-05 §8).
+//     lockout (SPIKE-felhom-sshd-2026-07-05 §8).
 //   - stock sshd listener answers?  — a TCP connect to the sshd port (22 for G1; H1 passes the
-//                                     discovered felhom-sshd port later).
+//     discovered felhom-sshd port later).
 //   - did the watchdog auto-heal?   — the watchdog writes an RFC3339 marker to /run when it heals;
-//                                     its presence (+ timestamp) tells the hub a clobber recurred, so
-//                                     the operator learns of a recurring cause BEFORE a lockout.
+//     its presence (+ timestamp) tells the hub a clobber recurred, so
+//     the operator learns of a recurring cause BEFORE a lockout.
 //
 // Read-only and dependency-light: os.Stat + os.ReadFile + a short net.Dial, no exec, no sudo. Safe to
 // run every heartbeat.
@@ -44,9 +44,9 @@ const (
 type Reporter struct {
 	privsepDir string
 	marker     string
-	sshdAddr   string                                  // host:port dialed for the reachability probe
-	statDir    func(string) bool                       // dir-exists check (os.Stat wrapper; test seam)
-	readMarker func(string) (string, bool)             // marker read → (timestamp, present)
+	sshdAddr   string                      // host:port dialed for the reachability probe
+	statDir    func(string) bool           // dir-exists check (os.Stat wrapper; test seam)
+	readMarker func(string) (string, bool) // marker read → (timestamp, present)
 	dialSSHD   func(ctx context.Context, addr string) bool
 }
 

internal/osupdate/leg.go

@@ -370,7 +370,8 @@ func HealthVerdict(before, after *Health) (bool, string) {
 
 // HostHealthVerdict is THE host health rule (`11` §8.2; pinned by TestHostHealthVerdict): the Proxmox daemons and the
 // agent are active, the customer guest still runs, the guest's own rule still passes against the start of the pass,
-// and the tunnel is `running` (R-841).
+// and the tunnel is `running` (R-841) — or `absent` both now and before the update: a box with no tunnel set up
+// (R-1070; pinned by TestHostHealthVerdict_R1070_NoTunnelSetUp).
 func HostHealthVerdict(before, after *Health, tunnel string) (bool, string) {
 	if after == nil {
 		return false, "no health reading"
@@ -398,12 +399,28 @@ func HostHealthVerdict(before, after *Health, tunnel string) (bool, string) {
 	if ok, why := HealthVerdict(gb, after.Guest); !ok {
 		return false, "guest: " + why
 	}
-	if tunnel != hub.TunnelRunning {
+	switch {
+	case tunnel == hub.TunnelRunning:
+	case tunnel == hub.TunnelAbsent && noTunnelBefore(before):
+		// R-1070: no cloudflared container before the update and none after — the box has no tunnel to check.
+	case tunnel == hub.TunnelAbsent:
+		return false, "the tunnel is absent (no cloudflared container now, and the reading from before the update does not show it was absent then)"
+	default:
 		return false, "the tunnel is " + tunnel
 	}
 	return true, ""
 }
 
+// noTunnelBefore: the reading from the start of the pass shows the guest's Docker answered and listed NO cloudflared
+// container. A missing or failed reading is not „no tunnel" (an empty list from a failed `docker ps` must not pass).
+func noTunnelBefore(before *Health) bool {
+	if before == nil || before.Guest == nil || !before.Guest.DockerOK {
+		return false
+	}
+	_, had := before.Guest.Containers["cloudflared"]
+	return !had
+}
+
 // EngineOf is the engine version `docker version` prints for a docker-ce package version: "5:29.8.2-1~debian.13~trixie"
 // → "29.8.2" (no epoch, no Debian revision).
 func EngineOf(pkgVersion string) string {

internal/osupdate/leg_test.go

@@ -568,3 +568,42 @@ func TestDocker_OOMCheckReachesTheHubUnchanged(t *testing.T) {
 		t.Fatalf("no docker report posted: %s", calls(w))
 	}
 }
+
+// R-1070: a box with no tunnel set up (no cloudflared container before the update and none after) passes the host
+// rule; a box WITH a tunnel that is down, gone, or unreadable still fails it.
+//
+// COMPANION RED-PROOF (observed 2026-10-19): with the „not applicable" branch switched off (only `running` passes, the
+// pre-fix rule) the first case fails with
+//
+//	leg_test.go: no tunnel before, none after: got false (the tunnel is absent (no cloudflared container now, …)), want true
+//
+// and with the pre-fix case-sensitive match TestClassifyTunnel's Docker 29 row reads `unknown` („could not ask the
+// guest"), the state Tester4 failed on.
+// Restored.
+func TestHostHealthVerdict_R1070_NoTunnelSetUp(t *testing.T) {
+	withTunnel := hostOK()
+	withTunnel.Guest.Containers["cloudflared"] = Container{State: "running", Health: "healthy"}
+	dockerFailed := hostOK()
+	dockerFailed.Guest = &Health{DockerOK: false, Containers: map[string]Container{}}
+	cases := []struct {
+		name   string
+		before *Health
+		tunnel string
+		want   bool
+		why    string
+	}{
+		{"no tunnel before, none after", hostOK(), hub.TunnelAbsent, true, ""},
+		{"a tunnel before, gone after", withTunnel, hub.TunnelAbsent, false, "cloudflared was running"},
+		{"a tunnel before, down after", withTunnel, hub.TunnelNotRunning, false, "cloudflared was running"},
+		{"no tunnel before, a stopped one after", hostOK(), hub.TunnelNotRunning, false, "not_running"},
+		{"no tunnel before, the probe could not ask after", hostOK(), hub.TunnelUnknown, false, "unknown"},
+		{"no reading from before", nil, hub.TunnelAbsent, false, "absent"},
+		{"docker failed before (an empty list is not proof of none)", dockerFailed, hub.TunnelAbsent, false, "absent"},
+	}
+	for _, c := range cases {
+		got, why := HostHealthVerdict(c.before, hostOK(), c.tunnel)
+		if got != c.want || !strings.Contains(why, c.why) {
+			t.Errorf("%s: got %v (%s), want %v (…%s…)", c.name, got, why, c.want, c.why)
+		}
+	}
+}

internal/proxmox/mutate.go

@@ -180,6 +180,18 @@ func (c *Client) DeleteSnapshot(ctx context.Context, vmid int, snapname string)
 	return c.dataString(ctx, http.MethodDelete, path, nil)
 }
 
+// DeleteSnapshotForce is DeleteSnapshot with Proxmox's own `force=1` („remove from the config file, even if removing
+// disk snapshots fails"): the API's flag, so the same token privilege (VM.Snapshot) and no new root command. Used by
+// the stale-lock janitor ONLY after a delete failed because one disk's snapshot volume is already gone (R-1010). The
+// flag rides the query string, where pveproxy reads a DELETE's parameters. Pinned by TestDeleteSnapshotForce_R1010.
+func (c *Client) DeleteSnapshotForce(ctx context.Context, vmid int, snapname string) (string, error) {
+	if vmid == 0 || snapname == "" {
+		return "", fmt.Errorf("proxmox: DeleteSnapshotForce needs vmid and snapname")
+	}
+	path := fmt.Sprintf("/nodes/%s/lxc/%d/snapshot/%s?force=1", c.node, vmid, url.PathEscape(snapname))
+	return c.dataString(ctx, http.MethodDelete, path, nil)
+}
+
 // SetConfig applies config changes via PUT /nodes/{node}/lxc/{vmid}/config
 // (e.g. memory, cores, net0, mpN with a backup flag). PVE may apply this
 // synchronously (no UPID) — the returned string is empty in that case, and "" is

internal/proxmox/scratch_destroy_test.go

@@ -3,6 +3,7 @@ package proxmox
 import (
 	"context"
 	"io"
+	"net/http"
 	"strings"
 	"testing"
 )
@@ -99,3 +100,24 @@ func TestDestroyScratchLXC_RefusesUnconfiguredBand(t *testing.T) {
 		}
 	}
 }
+
+// R-1010: the forced snapshot delete is the API's own flag on the same DELETE (no new root command).
+func TestDeleteSnapshotForce_R1010(t *testing.T) {
+	var got *http.Request
+	d := &mockDoer{fn: func(r *http.Request) (*http.Response, error) {
+		got = r
+		return jsonResp(200, `{"data":"UPID:demo-felhom:0001:0002:0003:vzdelsnapshot:9201:u@pve!t:"}`), nil
+	}}
+	upid, err := newTestClient(d).DeleteSnapshotForce(context.Background(), 9201, "vzdump")
+	if err != nil || upid == "" {
+		t.Fatalf("DeleteSnapshotForce: upid %q err %v", upid, err)
+	}
+	if got.Method != http.MethodDelete || got.URL.Path != "/api2/json/nodes/demo-felhom/lxc/9201/snapshot/vzdump" ||
+		got.URL.Query().Get("force") != "1" {
+		t.Fatalf("request = %s %s, want DELETE …/lxc/9201/snapshot/vzdump?force=1", got.Method, got.URL.String())
+	}
+	d.fn = func(r *http.Request) (*http.Response, error) { got = r; return jsonResp(200, `{"data":null}`), nil }
+	if _, err := newTestClient(d).DeleteSnapshot(context.Background(), 9201, "vzdump"); err != nil || got.URL.RawQuery != "" {
+		t.Fatalf("the plain delete must carry no force: %s (err %v)", got.URL.String(), err)
+	}
+}

internal/reconcile/bringup.go

@@ -59,7 +59,7 @@ const DefaultPool = "felhom"
 //
 // SysDataGrowGB itself is NOT removed — see its field comment: the host installer still passes
 // `-sysdata-grow`, and its GiB are FOLDED INTO the single volume's grow rather than dropped.
-const DefaultSysDataMount = "" 
+const DefaultSysDataMount = ""
 
 // Structural host-bind mountpoints every provisioned guest carries (GL-5; verdict of
 // SPIKE-dr-bindmount-source-2026-07-07): the permanent drives parent bind (mp8,

internal/reconcile/classify.go

@@ -64,6 +64,11 @@ const (
 	// The config bundle (agent v0.143.0, R-840, `11` §5.4.2): the box's ROOT-OWNED files (sudoers, wrappers, units).
 	// Destructive-class (signed, operational key) like agent_update; the root wrapper re-verifies the signature itself.
 	ClassAgentConfigUpdate OpClass = "agent_config_update"
+
+	// The internal data place on an OLD one-disk box (agent v0.158.0, felhom.eu `16` §9): thin space on the system
+	// disk, made by the root helper felhom-internal-place. Destructive-class (signed, operational key): it is an
+	// operator act, never a household button, and no agent-internal provenance may start it.
+	ClassAddInternalPlace OpClass = "add_internal_place"
 )
 
 // Disposition is the classifier verdict.
@@ -132,7 +137,7 @@ func Classify(class OpClass, prov Provenance) Disposition {
 		return Destructive
 	case ClassKeyRotation:
 		return Destructive
-	case ClassAgentUpdate, ClassOSDockerStep, ClassOSPVEStep, ClassOSKernelStep, ClassAgentConfigUpdate:
+	case ClassAgentUpdate, ClassOSDockerStep, ClassOSPVEStep, ClassOSKernelStep, ClassAgentConfigUpdate, ClassAddInternalPlace:
 		// Never benign — no agent-internal provenance can make replacing the agent binary
 		// unsigned-safe (a compromised process must not be able to self-bless an update).
 		return Destructive

internal/signedjobs/decommission.go

@@ -39,9 +39,21 @@ type decommissionParams struct {
 // none is wanted, because a drive can legitimately be decommissioned while physically ABSENT.
 type DecommissionExecutor struct {
 	intent IntentDecommissioner
+	binds  GuestBindPruner // optional (v0.159.1): drop the drive from every guest's bind record, as the self-serve path does
 	logger *slog.Logger
 }
 
+// GuestBindPruner is the per-guest enrolled-bind record (*localapi.GuestBindStore). The self-serve decommission prunes
+// the drive from it; the signed one did not, so a forgotten drive (R-1124) left a record the 20-second guest-bind
+// re-assert warned about forever ("skipping non-enrolled drive"), measured on demo-hp 2026-10-17.
+type GuestBindPruner interface {
+	Guests() map[int][]string
+	Remove(vmid int, durableID string) error
+}
+
+// SetGuestBinds wires the bind record (nil-safe: no record, nothing pruned).
+func (d *DecommissionExecutor) SetGuestBinds(b GuestBindPruner) { d.binds = b }
+
 // NewDecommissionExecutor wires the executor to the agent's intent store.
 func NewDecommissionExecutor(intent IntentDecommissioner, logger *slog.Logger) *DecommissionExecutor {
 	if logger == nil {
@@ -75,6 +87,20 @@ func (d *DecommissionExecutor) Execute(_ context.Context, op string, params json
 	if err := d.intent.SetDecommissioned(p.DurableID); err != nil {
 		return fmt.Errorf("decommission: record intent for %q: %w", p.DurableID, err)
 	}
+	if d.binds != nil {
+		for vmid, ids := range d.binds.Guests() {
+			for _, id := range ids {
+				if id != p.DurableID {
+					continue
+				}
+				if err := d.binds.Remove(vmid, id); err != nil {
+					d.logger.Warn("decommission: guest-bind record not pruned", "vmid", vmid, "durable_id", id, "err", err)
+				} else {
+					d.logger.Info("decommission: drive removed from the guest's bind record", "vmid", vmid, "durable_id", id)
+				}
+			}
+		}
+	}
 	d.logger.Warn("decommission: operator-signed decommission complete (drive will never auto-mount)", "durable_id", p.DurableID)
 	return nil
 }

internal/signedjobs/decommission_test.go

@@ -89,3 +89,37 @@ func TestExecutorChain_DispatchesToOwner(t *testing.T) {
 		t.Errorf("want ErrNoExecutor for an unowned op, got %v", err)
 	}
 }
+
+type fakeBinds struct{ m map[int][]string }
+
+func (f *fakeBinds) Guests() map[int][]string { return f.m }
+func (f *fakeBinds) Remove(vmid int, id string) error {
+	var kept []string
+	for _, x := range f.m[vmid] {
+		if x != id {
+			kept = append(kept, x)
+		}
+	}
+	f.m[vmid] = kept
+	return nil
+}
+
+// v0.159.1: the signed decommission drops the drive from EVERY guest's bind record, as the self-serve path does — and
+// only that drive. RED-PROOF (felhom.eu audits/agent-allowlist-rotation-2026-10-17/P1/redproof-bindprune.txt): with
+// the prune loop removed, the record still holds the drive and this test fails.
+func TestDecommissionExecutor_PrunesTheGuestBindRecord(t *testing.T) {
+	fd := &fakeDecommissioner{}
+	fb := &fakeBinds{m: map[int][]string{9201: {"uuid:keep", "uuid:gone"}, 9202: {"uuid:gone"}}}
+	ex := NewDecommissionExecutor(fd, nil)
+	ex.SetGuestBinds(fb)
+	params, _ := json.Marshal(decommissionParams{DurableID: "uuid:gone"})
+	if err := ex.Execute(context.Background(), opDecommission, params); err != nil {
+		t.Fatal(err)
+	}
+	if got := fb.m[9201]; len(got) != 1 || got[0] != "uuid:keep" {
+		t.Fatalf("9201 record = %v, want [uuid:keep]", got)
+	}
+	if got := fb.m[9202]; len(got) != 0 {
+		t.Fatalf("9202 record = %v, want empty", got)
+	}
+}

internal/signedjobs/internalplace.go

@@ -0,0 +1,91 @@
+package signedjobs
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"log/slog"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/localapi"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// opAddInternalPlace is the op class this executor serves (mirrors reconcile.ClassAddInternalPlace).
+const opAddInternalPlace = "add_internal_place"
+
+// InternalPlaceHelper is the root helper's path (configs/felhom-internal-place, from the config bundle).
+const InternalPlaceHelper = "/usr/local/sbin/felhom-internal-place"
+
+var reHelperResult = regexp.MustCompile(`^(created|existing|formatted) uuid=([0-9a-fA-F-]{36}) gib=([0-9]+)$`)
+
+// InternalPlaceExecutor adds the internal data place to an OLD one-disk box (felhom.eu architecture/16 §9: „one-disk
+// boxes get no internal place by themselves; adding one later is a signed operator job, not a household button").
+// It runs the SAME root helper the installer runs — `felhom-internal-place create 0` (0 GiB planned: the guest's
+// volumes already exist and are counted) — writes the record the agent mounts from, and asks the local API to mount
+// it now. Every safety rule (the ruled size, never format what holds anything) lives in the helper, as root.
+//
+// The op carries no parameters: the size is the rule's, never the signer's. Destructive-class by default (the
+// classifier's fail-safe; ClassAddInternalPlace documents it): it takes thin space on the system disk, so only an
+// operator signature may start it.
+type InternalPlaceExecutor struct {
+	runner     proxmox.Runner
+	recordFile string
+	ensure     func(context.Context) // late-bound: the local API server is built after the jobs runner
+	logger     *slog.Logger
+}
+
+// NewInternalPlaceExecutor wires the executor. ensure may be nil at construction and set later through SetEnsure.
+func NewInternalPlaceExecutor(runner proxmox.Runner, recordFile string, logger *slog.Logger) *InternalPlaceExecutor {
+	if logger == nil {
+		logger = slog.Default()
+	}
+	return &InternalPlaceExecutor{runner: runner, recordFile: recordFile, logger: logger}
+}
+
+// SetEnsure late-binds the mount step (localapi.Server.EnsureInternalPlace).
+func (e *InternalPlaceExecutor) SetEnsure(f func(context.Context)) { e.ensure = f }
+
+// Execute implements signedjobs.Executor for add_internal_place.
+func (e *InternalPlaceExecutor) Execute(ctx context.Context, op string, params json.RawMessage) error {
+	if op != opAddInternalPlace {
+		return ErrNoExecutor
+	}
+	if p := bytes.TrimSpace(params); len(p) > 0 && !bytes.Equal(p, []byte("{}")) && !bytes.Equal(p, []byte("null")) {
+		return fmt.Errorf("add_internal_place: takes no parameters (the size is the ruled rule's), got %s", p)
+	}
+	if e.runner == nil || e.recordFile == "" {
+		return fmt.Errorf("add_internal_place: not wired on this host (no privileged runner or state dir)")
+	}
+	start := time.Now()
+	e.logger.Warn("add_internal_place: executing operator-signed job (the internal data place, `16` §9)")
+	out, stderr, err := e.runner.Run(ctx, InternalPlaceHelper, "create", "0")
+	line := strings.TrimSpace(string(out))
+	if err != nil {
+		return fmt.Errorf("add_internal_place: helper refused: %v: %s", err, strings.TrimSpace(string(stderr)))
+	}
+	if strings.HasPrefix(line, "skipped ") {
+		// Not an error the operator can retry away: the pool has no room by the ruled rule. Fail the job with the
+		// helper's own numbers so the hub shows why.
+		return fmt.Errorf("add_internal_place: %s", line)
+	}
+	m := reHelperResult.FindStringSubmatch(line)
+	if m == nil {
+		return fmt.Errorf("add_internal_place: unexpected helper output %q", line)
+	}
+	gib, _ := strconv.Atoi(m[3])
+	rec := localapi.InternalPlaceRecord{FSUUID: m[2], GiB: gib, MadeBy: "signed-job", MadeAt: time.Now().UTC().Format(time.RFC3339)}
+	if err := localapi.WriteInternalPlace(e.recordFile, rec); err != nil {
+		return fmt.Errorf("add_internal_place: writing the record: %w", err)
+	}
+	if e.ensure != nil {
+		e.ensure(ctx)
+	}
+	e.logger.Warn("add_internal_place: complete", "result", m[1], "fs_uuid", m[2], "gib", gib,
+		"duration_ms", time.Since(start).Milliseconds())
+	return nil
+}

internal/signedjobs/internalplace_test.go

@@ -0,0 +1,94 @@
+package signedjobs
+
+import (
+	"context"
+	"errors"
+	"io"
+	"log/slog"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/localapi"
+	"gitea.dooplex.hu/admin/felhom-agent/internal/reconcile"
+)
+
+// add_internal_place (felhom.eu architecture/16 §9, agent v0.158.0): the operator's job for an OLD one-disk box.
+
+type ipRunner struct {
+	out, stderr string
+	err         error
+	calls       [][]string
+}
+
+func (r *ipRunner) Run(_ context.Context, name string, args ...string) ([]byte, []byte, error) {
+	r.calls = append(r.calls, append([]string{name}, args...))
+	return []byte(r.out), []byte(r.stderr), r.err
+}
+func (r *ipRunner) RunStdin(ctx context.Context, _ io.Reader, name string, args ...string) ([]byte, []byte, error) {
+	return r.Run(ctx, name, args...)
+}
+
+func ipExec(t *testing.T, r *ipRunner) (*InternalPlaceExecutor, string, *int) {
+	t.Helper()
+	file := filepath.Join(t.TempDir(), localapi.InternalPlaceFile)
+	e := NewInternalPlaceExecutor(r, file, slog.New(slog.NewTextHandler(io.Discard, nil)))
+	n := 0
+	e.SetEnsure(func(context.Context) { n++ })
+	return e, file, &n
+}
+
+func TestAddInternalPlace_RunsTheHelperWritesTheRecordAndMounts(t *testing.T) {
+	r := &ipRunner{out: "created uuid=0b6e4c5a-1d2e-4f3a-9b8c-7d6e5f4a3b2c gib=33\n"}
+	e, file, ensured := ipExec(t, r)
+	if err := e.Execute(context.Background(), "add_internal_place", []byte(`{}`)); err != nil {
+		t.Fatal(err)
+	}
+	if len(r.calls) != 1 || strings.Join(r.calls[0], " ") != InternalPlaceHelper+" create 0" {
+		t.Fatalf("the helper must be called exactly as `create 0`, got %v", r.calls)
+	}
+	rec, ok, err := localapi.ReadInternalPlace(file)
+	if err != nil || !ok || rec.FSUUID != "0b6e4c5a-1d2e-4f3a-9b8c-7d6e5f4a3b2c" || rec.GiB != 33 || rec.MadeBy != "signed-job" {
+		t.Fatalf("record: %+v ok=%v err=%v", rec, ok, err)
+	}
+	if *ensured != 1 {
+		t.Fatalf("the mount must be asked for once, got %d", *ensured)
+	}
+}
+
+func TestAddInternalPlace_NoRoomOrRefusalOrGarbage_WritesNothing(t *testing.T) {
+	for name, r := range map[string]*ipRunner{
+		"skipped": {out: "skipped only 7 GiB free for it (pool 53, reserve 16, promised 102, planned 0; needs 20)"},
+		"refused": {err: errors.New("exit status 1"), stderr: "felhom-internal-place: REFUSED: holds something"},
+		"garbage": {out: "created uuid=../../etc gib=1"},
+	} {
+		e, file, ensured := ipExec(t, r)
+		if err := e.Execute(context.Background(), "add_internal_place", nil); err == nil {
+			t.Fatalf("%s: want an error", name)
+		}
+		if _, statErr := os.Stat(file); !os.IsNotExist(statErr) || *ensured != 0 {
+			t.Fatalf("%s: no record and no mount may follow, stat=%v ensured=%d", name, statErr, *ensured)
+		}
+	}
+}
+
+func TestAddInternalPlace_TakesNoParametersAndIgnoresOtherOps(t *testing.T) {
+	r := &ipRunner{out: "created uuid=0b6e4c5a-1d2e-4f3a-9b8c-7d6e5f4a3b2c gib=33"}
+	e, _, _ := ipExec(t, r)
+	if err := e.Execute(context.Background(), "add_internal_place", []byte(`{"gib":900}`)); err == nil || len(r.calls) != 0 {
+		t.Fatalf("a size from the signer must be refused before the helper runs, err=%v calls=%v", err, r.calls)
+	}
+	if err := e.Execute(context.Background(), "decommission", []byte(`{}`)); !errors.Is(err, ErrNoExecutor) {
+		t.Fatalf("another op: want ErrNoExecutor, got %v", err)
+	}
+}
+
+func TestAddInternalPlace_IsSignatureOnly(t *testing.T) {
+	if got := reconcile.Classify(reconcile.ClassAddInternalPlace, reconcile.Provenance{}); got != reconcile.Destructive {
+		t.Fatalf("add_internal_place must need an operator signature, classified %v", got)
+	}
+	if string(reconcile.ClassAddInternalPlace) != opAddInternalPlace {
+		t.Fatalf("op string drift: %q vs %q", reconcile.ClassAddInternalPlace, opAddInternalPlace)
+	}
+}

internal/signedjobs/runner.go

@@ -25,8 +25,8 @@ import (
 // operator's signing CLI (cmd/felhom-opsign) produces it; this runner consumes it. The hub never
 // forges one (it holds no signing key) and never opens it.
 type Envelope struct {
-	OpBlobB64  string `json:"op_blob_b64"`  // base64 of the canonical OpBlob JSON
-	SigArmored string `json:"sig_armored"`  // armored SSHSIG over the op-blob (namespace felhom-op-v1)
+	OpBlobB64  string `json:"op_blob_b64"` // base64 of the canonical OpBlob JSON
+	SigArmored string `json:"sig_armored"` // armored SSHSIG over the op-blob (namespace felhom-op-v1)
 }
 
 // JobSource fetches + clears jobs. Satisfied by *hub.Client.

internal/storage/candidates.go

@@ -13,14 +13,16 @@ import (
 // (guarded, Impl-1) and/or mounts + binds. Blank disks are "initialize"-only; disks already carrying a
 // mountable FS are ALSO "attach" candidates.
 type CandidateDisk struct {
-	Device      string `json:"device"`       // whole disk, e.g. /dev/sdd
-	SizeBytes   int64  `json:"size_bytes"`   // 0 when unreadable
+	Device      string `json:"device"`     // whole disk, e.g. /dev/sdd
+	SizeBytes   int64  `json:"size_bytes"` // 0 when unreadable
 	Model       string `json:"model,omitempty"`
-	FSType      string `json:"fstype,omitempty"`      // first filesystem found ("" = blank)
-	DataBearing bool   `json:"data_bearing"`          // has any FS / partition (→ wipe-confirm downstream)
-	Mountable   bool   `json:"mountable"`             // carries an ext4/xfs FS the agent can mount as-is
+	FSType      string `json:"fstype,omitempty"`       // first filesystem found ("" = blank)
+	DataBearing bool   `json:"data_bearing"`           // has any FS / partition (→ wipe-confirm downstream)
+	Mountable   bool   `json:"mountable"`              // carries an ext4/xfs FS the agent can mount as-is
 	MountSource string `json:"mount_source,omitempty"` // the node to mount for attach (e.g. /dev/sdd1)
 	DurableID   string `json:"durable_id,omitempty"`   // "uuid:<fs-uuid>" if it has an FS; "" for a blank disk
+	// Partitions (v0.158.0, felhom.eu `16` §6): each partition of this disk as its own place — see partitions.go.
+	Partitions []CandidatePartition `json:"partitions,omitempty"`
 }
 
 // mountableFSTypes are the filesystems the agent can mount as-is (EnsureMount) → an "attach" candidate.
@@ -83,6 +85,8 @@ func (h *SudoHostOps) buildCandidate(dev, name string) CandidateDisk {
 			if len(nodes) > 1 {
 				c.DataBearing = true
 			}
+			c.Partitions = candidatePartitions(lout, name, readSysPartSize, readSysPartNumber, h.host.ResolveUUID,
+				felhomOwnedMounts(dev, nodes, h.mountTable))
 		}
 	}
 	return c

internal/storage/claim_test.go

@@ -14,10 +14,10 @@ func TestClassifyClaim(t *testing.T) {
 			nodes: []claimNode{{name: "sdd", fstype: "", mountpoint: ""}, {name: "sdd1", fstype: "ntfs", mountpoint: ""}}}
 	}
 	cases := []struct {
-		name       string
-		mutate     func(f *claimFacts)
+		name          string
+		mutate        func(f *claimFacts)
 		wantUnclaimed bool
-		reasonHas  string
+		reasonHas     string
 	}{
 		{"clean unclaimed disk (ntfs, unmounted)", func(f *claimFacts) {}, true, "unclaimed"},
 		{"blank disk", func(f *claimFacts) { f.nodes = []claimNode{{name: "sdd"}} }, true, "unclaimed"},

internal/storage/durable_device.go

@@ -4,6 +4,7 @@ import (
 	"fmt"
 	"os"
 	"path/filepath"
+	"regexp"
 	"sort"
 	"strings"
 )
@@ -24,8 +25,13 @@ var devDiskRoot = "/dev/disk"
 const (
 	durableByID   = "byid:"
 	durableByUUID = "byuuid:"
+	// partFSUUIDSep joins a partition's by-id link and its fs-UUID in one wipe id (v0.158.0, `16` §6).
+	partFSUUIDSep = "+fsuuid:"
 )
 
+// rePartLink matches a by-id link of a PARTITION (udev appends -part<N> to the disk's link).
+var rePartLink = regexp.MustCompile(`-part[1-9][0-9]{0,2}$`)
+
 // byIDPriority ranks /dev/disk/by-id link prefixes most-stable-first: wwn (hardware world-wide
 // name) > nvme-eui (NVMe EUI) > nvme-/ata-/scsi-/usb- (model+serial). dm-/lvm-/md- names are
 // excluded (they are mapper constructs, not the physical disk we want to bind a wipe to).
@@ -41,6 +47,14 @@ func DeviceDurableID(device string) (string, error) {
 	}
 	// 1. best by-id link pointing at this device.
 	if name := bestByIDLink(target); name != "" {
+		// v0.158.0 (felhom.eu `16` §6): a PARTITION's link already carries the disk id + the part number
+		// (wwn-…-part2). When the part holds a filesystem, its fs-UUID is bound too, so a part that was reformatted —
+		// or another disk's part2 that took the link — derives a different id and the wipe refuses.
+		if rePartLink.MatchString(name) {
+			if uuid := byUUIDFor(target); uuid != "" {
+				return durableByID + name + partFSUUIDSep + uuid, nil
+			}
+		}
 		return durableByID + name, nil
 	}
 	// 2. fallback: a filesystem UUID symlink.
@@ -56,10 +70,29 @@ func ResolveDurableDevice(durableID string) (string, error) {
 	switch {
 	case strings.HasPrefix(durableID, durableByID):
 		name := strings.TrimPrefix(durableID, durableByID)
+		fsUUID := ""
+		if i := strings.Index(name, partFSUUIDSep); i >= 0 {
+			name, fsUUID = name[:i], name[i+len(partFSUUIDSep):]
+			if !rePartLink.MatchString(name) || !safeLinkName(fsUUID) {
+				return "", fmt.Errorf("storage: unsafe durable id %q", durableID)
+			}
+		}
 		if !safeLinkName(name) {
 			return "", fmt.Errorf("storage: unsafe durable id %q", durableID)
 		}
-		return filepath.EvalSymlinks(filepath.Join(devDiskRoot, "by-id", name))
+		dev, err := filepath.EvalSymlinks(filepath.Join(devDiskRoot, "by-id", name))
+		if err != nil || fsUUID == "" {
+			return dev, err
+		}
+		// Both halves must name the SAME device now (`16` §6): the part by its disk id + number, and its filesystem.
+		byFS, ferr := filepath.EvalSymlinks(filepath.Join(devDiskRoot, "by-uuid", fsUUID))
+		if ferr != nil {
+			return "", fmt.Errorf("storage: durable id %q: its filesystem %s no longer resolves: %w", durableID, fsUUID, ferr)
+		}
+		if byFS != dev {
+			return "", fmt.Errorf("storage: durable id %q: the part (%s) and its filesystem (%s) are different devices now", durableID, dev, byFS)
+		}
+		return dev, nil
 	case strings.HasPrefix(durableID, durableByUUID):
 		uuid := strings.TrimPrefix(durableID, durableByUUID)
 		if !safeLinkName(uuid) {

internal/storage/forget_retired.go

@@ -0,0 +1,85 @@
+package storage
+
+import (
+	"context"
+	"os"
+	"path/filepath"
+	"strings"
+)
+
+// ForgetRetiredDrives (R-1124) removes the host mount unit of every DECOMMISSIONED drive that is physically gone, so a
+// later host boot no longer waits for it and the agent stops listing it (RegistryKnownTargets and
+// ReassertEnrolledMounts both enumerate the units — no unit, no entry, no „absent by UUID" line every 20 s).
+//
+// Only a drive that is BOTH decommissioned (the operator's or the customer's permanent removal) AND absent (its
+// /dev/disk/by-uuid link does not resolve) AND not mounted is forgotten. A decommissioned drive that is still plugged in
+// keeps its unit and its raw host mount — that is what makes the one-click re-enroll (H3) work, and decommission stays a
+// logical retire (`03` §storage). The intent record stays `decommissioned`: a forgotten drive that comes back is a new
+// candidate, never auto-mounted. The root half (`felhom-priv-apply unit-forget`) re-checks every one of these conditions
+// and refuses otherwise. Called at agent start, on the 20-second storage tick, and right after a decommission. Best
+// effort per unit; returns how many units it removed. Pinned by TestForgetRetiredDrives_*.
+func (h *SudoHostOps) ForgetRetiredDrives(ctx context.Context, intent IntentReader) int {
+	if intent == nil {
+		return 0 // no intent store → we cannot know a drive was retired; forget nothing
+	}
+	entries, err := os.ReadDir(h.unitDir)
+	if err != nil {
+		h.logger.Warn("storage: forget retired drives — cannot read unit dir", "dir", h.unitDir, "err", err)
+		return 0
+	}
+	mounted := map[string]bool{}
+	if ms, merr := h.host.Mounts(); merr == nil {
+		for _, m := range ms {
+			mounted[m.MountPoint] = true
+		}
+	} else {
+		h.logger.Warn("storage: forget retired drives — mount table unreadable; forgetting nothing", "err", merr)
+		return 0
+	}
+	removed := 0
+	for _, e := range entries {
+		name := e.Name()
+		if !strings.HasSuffix(name, ".mount") {
+			continue
+		}
+		data, rerr := os.ReadFile(filepath.Join(h.unitDir, name))
+		if rerr != nil {
+			continue
+		}
+		spec, ok := parseFelhomMountUnit(string(data))
+		if !ok {
+			continue // not one of our by-UUID drive units (network units are refused by the parser)
+		}
+		durableID := "uuid:" + spec.UUID
+		if intent.Get(durableID) != IntentDecommissioned {
+			continue
+		}
+		if mounted[spec.Where] {
+			h.logger.Debug("storage: decommissioned drive still mounted — keeping its unit", "where", spec.Where)
+			continue
+		}
+		if _, derr := ResolveStorageDevice(durableID); derr == nil {
+			h.logger.Debug("storage: decommissioned drive is present — keeping its unit (re-enroll stays one click)", "where", spec.Where)
+			continue
+		}
+		if h.unitEnabled(name) {
+			if err := h.run(ctx, h.bins.Systemctl, "disable", "--", name); err != nil {
+				h.logger.Warn("storage: forget retired drive — disable failed", "where", spec.Where, "unit", name, "err", err)
+				continue
+			}
+		}
+		if err := h.run(ctx, h.bins.PrivApply, "unit-forget", name); err != nil {
+			h.logger.Warn("storage: forget retired drive — unit removal refused or failed", "where", spec.Where, "unit", name, "err", err)
+			continue
+		}
+		removed++
+		h.logger.Info("storage: forgot a decommissioned drive that is gone — its mount unit is removed",
+			"name", spec.Name, "where", spec.Where, "uuid", spec.UUID, "unit", name)
+	}
+	if removed > 0 {
+		if err := h.run(ctx, h.bins.Systemctl, "daemon-reload"); err != nil {
+			h.logger.Warn("storage: daemon-reload after forgetting retired drives failed", "err", err)
+		}
+	}
+	return removed
+}

internal/storage/forget_retired_test.go

@@ -0,0 +1,155 @@
+package storage
+
+import (
+	"context"
+	"io"
+	"os"
+	"path/filepath"
+	"reflect"
+	"testing"
+)
+
+// forgetRunner records every privileged call and plays the root half of `felhom-priv-apply unit-forget`: the unit file
+// disappears from the unit dir, so the CONSEQUENCE (the drive is no longer listed) is observable, not just the calls.
+type forgetRunner struct {
+	unitDir string
+	calls   [][]string
+}
+
+func (r *forgetRunner) RunStdin(ctx context.Context, _ io.Reader, name string, args ...string) ([]byte, []byte, error) {
+	return r.Run(ctx, name, args...)
+}
+
+func (r *forgetRunner) Run(_ context.Context, name string, args ...string) ([]byte, []byte, error) {
+	r.calls = append(r.calls, append([]string{name}, args...))
+	if len(args) == 2 && args[0] == "unit-forget" {
+		_ = os.Remove(filepath.Join(r.unitDir, args[1]))
+	}
+	return nil, nil, nil
+}
+
+type forgetCase struct {
+	intent   DriveIntent
+	present  bool // the by-uuid link resolves
+	mounted  bool
+	enabled  bool // the wants-symlink exists
+	mountErr bool
+}
+
+const forgetUUID = "4ebfd36d-87d3-4b63-839e-a5a710fc799a"
+
+func runForget(t *testing.T, c forgetCase, intent IntentReader) (int, *forgetRunner, *RegistryKnownTargets, string) {
+	t.Helper()
+	unitDir := t.TempDir()
+	unit := "mnt-media_a.mount"
+	writeUnit(t, unitDir, unit, renderMountUnit(MountSpec{Name: forgetUUID, UUID: forgetUUID, Where: "/mnt/media_a", FSType: "ext4"}))
+	if c.enabled {
+		if err := os.MkdirAll(filepath.Join(unitDir, "multi-user.target.wants"), 0o755); err != nil {
+			t.Fatal(err)
+		}
+		if err := os.Symlink(filepath.Join(unitDir, unit), filepath.Join(unitDir, "multi-user.target.wants", unit)); err != nil {
+			t.Fatal(err)
+		}
+	}
+	devRoot := t.TempDir()
+	if err := os.MkdirAll(filepath.Join(devRoot, "by-uuid"), 0o755); err != nil {
+		t.Fatal(err)
+	}
+	if c.present {
+		dev := filepath.Join(devRoot, "sdz1")
+		if err := os.WriteFile(dev, nil, 0o600); err != nil {
+			t.Fatal(err)
+		}
+		if err := os.Symlink(dev, filepath.Join(devRoot, "by-uuid", forgetUUID)); err != nil {
+			t.Fatal(err)
+		}
+	}
+	withDevDiskRoot(t, devRoot)
+	hr := &fakeHostReader{}
+	if c.mounted {
+		hr.mounts = []Mount{{MountPoint: "/mnt/media_a", FSType: "ext4"}}
+	}
+	if c.mountErr {
+		hr.mountsErr = os.ErrPermission
+	}
+	rr := &forgetRunner{unitDir: unitDir}
+	ops := NewSudoHostOps(SudoHostOpsConfig{Runner: rr, Bins: Binaries{}.withDefaults(), UnitDir: unitDir,
+		StageDir: t.TempDir(), Host: hr, Logger: quietLogger()})
+	if intent == nil {
+		intent = &fakeIntent{m: map[string]DriveIntent{"uuid:" + forgetUUID: c.intent}}
+	}
+	n := ops.ForgetRetiredDrives(context.Background(), intent)
+	return n, rr, NewRegistryKnownTargets(unitDir, intent, quietLogger()), unitDir
+}
+
+// The R-1124 case: decommissioned, then unplugged. The unit is disabled and removed through the root checker, systemd
+// is reloaded, and the CONSEQUENCE — the agent no longer lists the drive — holds. RED-PROOF (P1/redproof-forget.txt):
+// with ForgetRetiredDrives' body replaced by `return 0` this test fails (the drive is still listed).
+func TestForgetRetiredDrives_DecommissionedAndGone_UnitRemovedAndNotListed(t *testing.T) {
+	n, rr, known, _ := runForget(t, forgetCase{intent: IntentDecommissioned, enabled: true}, nil)
+	if n != 1 {
+		t.Fatalf("removed = %d, want 1 (calls %v)", n, rr.calls)
+	}
+	pa := Binaries{}.withDefaults()
+	want := [][]string{
+		{pa.Systemctl, "disable", "--", "mnt-media_a.mount"},
+		{pa.PrivApply, "unit-forget", "mnt-media_a.mount"},
+		{pa.Systemctl, "daemon-reload"},
+	}
+	if !reflect.DeepEqual(rr.calls, want) {
+		t.Fatalf("calls = %v\nwant    %v", rr.calls, want)
+	}
+	got, err := known.Known(context.Background())
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(got) != 0 {
+		t.Fatalf("the forgotten drive is still listed: %+v", got)
+	}
+}
+
+// Already disabled (the Tester 1 shape after a manual disable): no disable call, the removal still runs.
+func TestForgetRetiredDrives_AlreadyDisabled_NoDisableCall(t *testing.T) {
+	n, rr, _, _ := runForget(t, forgetCase{intent: IntentDecommissioned}, nil)
+	pa := Binaries{}.withDefaults()
+	want := [][]string{{pa.PrivApply, "unit-forget", "mnt-media_a.mount"}, {pa.Systemctl, "daemon-reload"}}
+	if n != 1 || !reflect.DeepEqual(rr.calls, want) {
+		t.Fatalf("n=%d calls=%v want %v", n, rr.calls, want)
+	}
+}
+
+// Everything that must KEEP its unit: a decommissioned drive still plugged in (re-enroll is one click — H3), one still
+// mounted, an enrolled or ejected drive that is merely absent (it comes back), a drive with no recorded intent, and an
+// unreadable mount table. No privileged call at all, and the drive stays listed where it was listed.
+// RED-PROOF: with the present-drive check removed, the "present" case calls unit-forget and fails.
+func TestForgetRetiredDrives_KeepsEveryOtherUnit(t *testing.T) {
+	for name, c := range map[string]forgetCase{
+		"decommissioned but present": {intent: IntentDecommissioned, present: true, enabled: true},
+		"decommissioned but mounted": {intent: IntentDecommissioned, mounted: true},
+		"enrolled and absent":        {intent: IntentEnrolled},
+		"ejected and absent":         {intent: IntentEjected},
+		"no intent":                  {intent: IntentNew},
+		"mount table unreadable":     {intent: IntentDecommissioned, mountErr: true},
+	} {
+		t.Run(name, func(t *testing.T) {
+			n, rr, _, unitDir := runForget(t, c, nil)
+			if n != 0 || len(rr.calls) != 0 {
+				t.Fatalf("n=%d calls=%v — this unit must be kept", n, rr.calls)
+			}
+			if _, err := os.Stat(filepath.Join(unitDir, "mnt-media_a.mount")); err != nil {
+				t.Fatalf("unit file gone: %v", err)
+			}
+		})
+	}
+}
+
+// No intent store on the host (open failure): nothing is known to be retired, so nothing is forgotten.
+func TestForgetRetiredDrives_NilIntent_ForgetsNothing(t *testing.T) {
+	unitDir := t.TempDir()
+	writeUnit(t, unitDir, "mnt-media_a.mount", renderMountUnit(MountSpec{Name: "x", UUID: forgetUUID, Where: "/mnt/media_a"}))
+	rr := &forgetRunner{unitDir: unitDir}
+	ops := NewSudoHostOps(SudoHostOpsConfig{Runner: rr, UnitDir: unitDir, StageDir: t.TempDir(), Host: &fakeHostReader{}, Logger: quietLogger()})
+	if n := ops.ForgetRetiredDrives(context.Background(), nil); n != 0 || len(rr.calls) != 0 {
+		t.Fatalf("n=%d calls=%v", n, rr.calls)
+	}
+}

internal/storage/hostops.go

@@ -103,12 +103,12 @@ type MountSpec struct {
 // Binaries holds the absolute paths of the allow-listed binaries (overridable from config
 // so the sudoers entries and the agent agree on exact paths).
 type Binaries struct {
-	Systemctl   string
-	Install     string
+	Systemctl string
+	Install   string
 	// PrivApply is the root content checker (R-861, agent v0.146.0): a unit file reaches /etc/systemd/system only
 	// through `felhom-priv-apply unit <name>`, which reads the staged copy from /var/lib/felhom-agent/units/ and
 	// refuses anything the renderers here never produce (a bind over /etc, a Where outside /mnt, …).
-	PrivApply string
+	PrivApply   string
 	Smartctl    string
 	Lvs         string
 	Blkid       string // device signature probe (8C data-bearing detection)
@@ -419,8 +419,9 @@ func (h *SudoHostOps) InspectDevice(ctx context.Context, device string) (DeviceP
 	// any TYPE/PTTYPE/USAGE line is positive data-bearing evidence. Its exit code is NOT relied
 	// on (blkid exits 2 on a blank device) — output presence is what matters. A broken/empty
 	// blkid simply adds no positive evidence; lsblk (below) is the read-success authority.
-	bout, _, _ := h.runner.Run(ctx, h.bins.Blkid, "-p", "-o", "export", device)
-	for k, v := range parseBlkidExport(bout) {
+	bout, _, berr := h.runner.Run(ctx, h.bins.Blkid, "-p", "-o", "export", device)
+	blk := parseBlkidExport(bout)
+	for k, v := range blk {
 		switch k {
 		case "TYPE":
 			probe.HasFilesystem = true
@@ -439,11 +440,11 @@ func (h *SudoHostOps) InspectDevice(ctx context.Context, device string) (DeviceP
 	// lsblk -J is the READ-SUCCESS authority + the partition/mount view. It exits 0 on any valid
 	// device (blank or not), so a clean parse means the agent reliably read the device. If lsblk
 	// fails, Probed stays false → DataBearing()=true (fail-safe — never call a device blank on a
-	// failed read).
+	// failed read). An exit 0 whose output does not parse into a device is NOT a read either (R-1093
+	// review): it stays unprobed, never "blank".
 	lout, _, lerr := h.runner.Run(ctx, h.bins.Lsblk, "-J", "-o", "NAME,FSTYPE,PTTYPE,MOUNTPOINT", device)
-	if lerr == nil {
+	if hasChildren, mounted, fstype, pttype, parsed := parseLsblkDevice(lout); lerr == nil && parsed {
 		probe.Probed = true
-		hasChildren, mounted, fstype, pttype := parseLsblkDevice(lout)
 		if hasChildren {
 			probe.HasPartitions = true
 		}
@@ -456,13 +457,33 @@ func (h *SudoHostOps) InspectDevice(ctx context.Context, device string) (DeviceP
 				probe.FSType = fstype
 			}
 		}
-		if pttype != "" {
+		if pttype != "" && !inheritedPartitionTable(blk, berr, pttype) {
 			probe.HasPartitionTable = true
 		}
 	}
 	return probe, nil
 }
 
+// inheritedPartitionTable reports whether lsblk's PTTYPE for the probed node is only the PARENT disk's
+// table (R-1093). lsblk prints the disk's table type on every one of its partitions, so an EMPTY
+// partition read as "device has a partition table" and asked for the typed erase confirmation.
+//
+// The partition's OWN signatures come from `blkid -p` on the partition itself (above): a nested table
+// inside it is its PTTYPE, a filesystem is its TYPE. So lsblk's value is set aside ONLY when blkid
+// positively proved two things — that it completed its probe of the device (exit 0: PART_ENTRY_* alone
+// comes from the PARENT's table, so output without a clean exit does not prove the partition's contents
+// were read), and that the device is a partition of a table of exactly that type (PART_ENTRY_NUMBER +
+// PART_ENTRY_SCHEME == lsblk's PTTYPE). Anything else keeps the old, safe reading: a blkid that failed,
+// exited non-zero or printed nothing, a whole disk, or a scheme that differs from lsblk's (then lsblk sees
+// a table blkid did not attribute to the parent). Pinned by TestInspect_R1093_* (hostops_disk_test.go).
+func inheritedPartitionTable(blkid map[string]string, blkidErr error, lsblkPTType string) bool {
+	if blkidErr != nil || blkid["PART_ENTRY_NUMBER"] == "" {
+		return false
+	}
+	scheme := blkid["PART_ENTRY_SCHEME"]
+	return scheme != "" && strings.EqualFold(scheme, lsblkPTType)
+}
+
 // Format runs mkfs.<fstype> on a validated device. The caller is responsible for authorization
 // (8C: only after classifying the device non-data-bearing, or via a slice-10 operator signature).
 func (h *SudoHostOps) Format(ctx context.Context, device, fstype string) error {
@@ -541,13 +562,14 @@ type lsblkDevice struct {
 }
 
 // parseLsblkDevice parses `lsblk -J -o NAME,FSTYPE,PTTYPE,MOUNTPOINT <device>` for the top device:
-// whether it has child partitions, is mounted (itself or any child), and its fstype/pttype.
-func parseLsblkDevice(out []byte) (hasChildren, mounted bool, fstype, pttype string) {
+// whether it has child partitions, is mounted (itself or any child), and its fstype/pttype. parsed=false
+// when the output is not a device listing — the caller must then treat the device as unprobed.
+func parseLsblkDevice(out []byte) (hasChildren, mounted bool, fstype, pttype string, parsed bool) {
 	var doc struct {
 		BlockDevices []lsblkDevice `json:"blockdevices"`
 	}
 	if json.Unmarshal(out, &doc) != nil || len(doc.BlockDevices) == 0 {
-		return false, false, "", ""
+		return false, false, "", "", false
 	}
 	d := doc.BlockDevices[0]
 	fstype, pttype = d.FSType, d.PTType
@@ -558,7 +580,7 @@ func parseLsblkDevice(out []byte) (hasChildren, mounted bool, fstype, pttype str
 			mounted = true
 		}
 	}
-	return hasChildren, mounted, fstype, pttype
+	return hasChildren, mounted, fstype, pttype, true
 }
 
 // NoopHostOps is the safe fallback when the privileged surface is unavailable or declined

internal/storage/hostops_disk_test.go

@@ -222,3 +222,114 @@ func TestInspect_ReadsFilesystemUUID(t *testing.T) {
 		t.Fatalf("FSUUID = %q, want the blkid UUID", p.FSUUID)
 	}
 }
+
+// ---- R-1093: an EMPTY partition is blank; lsblk's PTTYPE on a partition is the PARENT disk's ----------
+
+// blkidEmptyGPTPart is `blkid -p -o export /dev/sdb1` for an empty GPT partition: only the partition's
+// own entry in the parent table (PART_ENTRY_*), no TYPE, no PTTYPE — nothing inside the partition.
+const blkidEmptyGPTPart = "DEVNAME=/dev/sdb1\nPART_ENTRY_SCHEME=gpt\nPART_ENTRY_UUID=3b0c1a52-55a4-4c4e-9d0b-7a1f0e0f7c11\n" +
+	"PART_ENTRY_TYPE=0fc63daf-8483-4772-8e79-3d69d8477de4\nPART_ENTRY_NUMBER=1\nPART_ENTRY_OFFSET=2048\n" +
+	"PART_ENTRY_SIZE=2097152\nPART_ENTRY_DISK=8:16\n"
+
+// lsblkPartInheritsGPT is `lsblk -J -o NAME,FSTYPE,PTTYPE,MOUNTPOINT /dev/sdb1`: lsblk prints the DISK's
+// table type on the partition — the input that made the agent read an empty partition as data-bearing.
+const lsblkPartInheritsGPT = `{"blockdevices":[{"name":"sdb1","fstype":null,"pttype":"gpt","mountpoint":null}]}`
+
+// TestInspect_R1093_PartitionVerdicts pins the CONSEQUENCE (DataBearing + Reason, what decides whether
+// the typed erase confirmation is asked) for a partition, in both directions.
+//
+// COMPANION RED-PROOFS (observed 2026-10-18): (1) restore the pre-fix `if pttype != "" {` (drop the
+// inheritedPartitionTable call) → BOTH empty-partition rows (GPT and dos) FAIL with
+// `DataBearing=true reason="device has a partition table", want blank`. (2) drop the `blkidErr != nil`
+// check → the "exit 2" and "I/O error" rows FAIL with `DataBearing=false reason="device is blank"`.
+// (3) drop `&& parsed` → both lsblk-exit-0 rows FAIL with `Probed=true … reason="device is blank"`
+// (the unparsable-output hole predates R-1093; closed here). Restored.
+func TestInspect_R1093_PartitionVerdicts(t *testing.T) {
+	cases := []struct {
+		name       string
+		dev        string
+		blkid      string
+		blkidErr   error
+		lsblk      string
+		lsblkErr   error
+		wantBear   bool
+		wantReason string
+		unprobed   bool // the verdict must be "could not be reliably inspected"
+	}{
+		{name: "empty GPT partition is blank (the R-1093 case)", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, lsblk: lsblkPartInheritsGPT,
+			wantBear: false, wantReason: "device is blank"},
+		{name: "empty MBR (dos) partition is blank", dev: "/dev/sdb2",
+			blkid:    "DEVNAME=/dev/sdb2\nPART_ENTRY_SCHEME=dos\nPART_ENTRY_TYPE=0x83\nPART_ENTRY_NUMBER=2\nPART_ENTRY_DISK=8:16\n",
+			lsblk:    `{"blockdevices":[{"name":"sdb2","fstype":null,"pttype":"dos","mountpoint":null}]}`,
+			wantBear: false, wantReason: "device is blank"},
+		{name: "partition with a filesystem stays data-bearing", dev: "/dev/sdb1",
+			blkid:    "DEVNAME=/dev/sdb1\nUUID=8c1e\nTYPE=ext4\nUSAGE=filesystem\n" + blkidEmptyGPTPart[len("DEVNAME=/dev/sdb1\n"):],
+			lsblk:    `{"blockdevices":[{"name":"sdb1","fstype":"ext4","pttype":"gpt","mountpoint":null}]}`,
+			wantBear: true, wantReason: "device has a ext4 filesystem"},
+		{name: "partition holding its OWN (nested) table stays data-bearing", dev: "/dev/sdb1",
+			blkid:    "DEVNAME=/dev/sdb1\nPTUUID=0b8d4358\nPTTYPE=dos\n" + blkidEmptyGPTPart[len("DEVNAME=/dev/sdb1\n"):],
+			lsblk:    lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "blkid printed nothing: lsblk's PTTYPE stands (fail-safe)", dev: "/dev/sdb1",
+			blkid: "", blkidErr: errors.New("sudo: a password is required"), lsblk: lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		// Fail-closed paths (the 2026-10-18 review): PART_ENTRY_* is read from the PARENT's table, so it
+		// proves nothing about the partition's contents unless blkid also completed (exit 0).
+		{name: "blkid exit 2 with only PART_ENTRY output: lsblk stands", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, blkidErr: errors.New("exit status 2"), lsblk: lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "blkid I/O error after PART_ENTRY output: lsblk stands", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, blkidErr: errors.New("exit status 1"), lsblk: lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "blkid binary missing: lsblk stands", dev: "/dev/sdb1",
+			blkid: "", blkidErr: errors.New(`exec: "/usr/sbin/blkid": file does not exist`), lsblk: lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "lsblk exit 0 with unparsable output: unprobed", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, lsblk: `lsblk: something odd`,
+			wantBear: true, unprobed: true},
+		{name: "lsblk exit 0 with no device listed: unprobed", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, lsblk: `{"blockdevices":[]}`,
+			wantBear: true, unprobed: true},
+		{name: "lsblk failed: unprobed", dev: "/dev/sdb1",
+			blkid: blkidEmptyGPTPart, lsblk: "", lsblkErr: errors.New("exit status 32"),
+			wantBear: true, unprobed: true},
+		{name: "scheme differs from lsblk's PTTYPE: lsblk stands", dev: "/dev/sdb1",
+			blkid:    strings.Replace(blkidEmptyGPTPart, "PART_ENTRY_SCHEME=gpt", "PART_ENTRY_SCHEME=dos", 1),
+			lsblk:    lsblkPartInheritsGPT,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "whole disk with a GPT stays data-bearing", dev: "/dev/sdb",
+			blkid:    "DEVNAME=/dev/sdb\nPTUUID=a07f88ff\nPTTYPE=gpt\n",
+			lsblk:    `{"blockdevices":[{"name":"sdb","fstype":null,"pttype":"gpt","mountpoint":null}]}`,
+			wantBear: true, wantReason: "device has a partition table"},
+		{name: "whole disk: blkid silent, lsblk GPT stands", dev: "/dev/sdb",
+			blkid: "", lsblk: `{"blockdevices":[{"name":"sdb","fstype":null,"pttype":"gpt","mountpoint":null}]}`,
+			wantBear: true, wantReason: "device has a partition table"},
+	}
+	for _, c := range cases {
+		t.Run(c.name, func(t *testing.T) {
+			r := &scriptedRunner{
+				outputs: map[string][]byte{"blkid": []byte(c.blkid), "lsblk": []byte(c.lsblk)},
+				errs:    map[string]error{"blkid": c.blkidErr, "lsblk": c.lsblkErr},
+			}
+			p, err := newSudo(r).InspectDevice(context.Background(), c.dev)
+			if err != nil {
+				t.Fatal(err)
+			}
+			if c.unprobed {
+				if p.Probed || !p.DataBearing() || p.Reason() != "device could not be reliably inspected" {
+					t.Fatalf("Probed=%v DataBearing=%v reason=%q, want unprobed + data-bearing (probe %+v)",
+						p.Probed, p.DataBearing(), p.Reason(), p)
+				}
+				return
+			}
+			if !p.Probed {
+				t.Fatal("expected a clean probe (lsblk read cleanly)")
+			}
+			if p.DataBearing() != c.wantBear || p.Reason() != c.wantReason {
+				t.Fatalf("DataBearing=%v reason=%q, want DataBearing=%v reason=%q (probe %+v)",
+					p.DataBearing(), p.Reason(), c.wantBear, c.wantReason, p)
+			}
+		})
+	}
+}

internal/storage/identity.go

@@ -0,0 +1,445 @@
+package storage
+
+import (
+	"bufio"
+	"context"
+	"os"
+	"os/exec"
+	"path/filepath"
+	"sort"
+	"strings"
+	"sync"
+	"time"
+)
+
+// Physical-disk identity of a storage place — felhom.eu architecture/16-storage-physical-identity.md §3.
+//
+// A storage place (a PVE storage, an enrolled drive, the guest's data volume, a network share) is
+// reduced to the PHYSICAL whole disks under it — partition → parent disk, device-mapper / LVM thin →
+// sysfs slaves (recursive), loop → its backing file → the filesystem holding that file — and each disk
+// is named by a key the agent trusts: `eui:` for NVMe, `serial:<bus>:<model>:<serial>` from udev,
+// `wwn:` only when no serial exists and the WWN carries a vendor (OUI ≠ 000000).
+//
+// The verdict is AGENT-COMPUTED evidence (`16` §10): it is read from /sys, /run/udev/data and
+// /proc/mounts only, never from a caller's claim, so a compromised controller cannot make two places
+// "different disks". `proven` means every disk under the place has a key the agent trusts; anything
+// else is `unknown` with ONE reason, chosen by precedence (vm > network > usb_bridge > weak_wwn >
+// no_serial). Nothing is cached: every call reads the current topology, so a disk swapped in the same
+// bay is re-judged on the next read (the "right before each Tier-2 run" rule is the caller calling).
+//
+// Pinned by identity_test.go (fixture /sys + udev trees, one per §3 case) — the spike that proved the
+// walk on real devices: felhom.eu documentation/audits/storage-identity-2026-10-15/.
+
+const (
+	IdentityProven  = "proven"
+	IdentityUnknown = "unknown"
+
+	ReasonVM        = "vm"
+	ReasonNetwork   = "network"
+	ReasonUSBBridge = "usb_bridge"
+	ReasonWeakWWN   = "weak_wwn"
+	ReasonNoSerial  = "no_serial"
+
+	PlaceKindSystem    = "system"
+	PlaceKindInternal  = "internal"
+	PlaceKindDisk      = "disk"
+	PlaceKindPartition = "partition"
+	PlaceKindNetwork   = "network"
+)
+
+// reasonRank orders the unknown reasons: the first one a place hits, by this precedence, is reported.
+var reasonRank = map[string]int{ReasonVM: 0, ReasonNetwork: 1, ReasonUSBBridge: 2, ReasonWeakWWN: 3, ReasonNoSerial: 4}
+
+// PlaceIdentity is what a storage place reports.
+type PlaceIdentity struct {
+	PhysicalDisks []string // disk keys, sorted, deduplicated
+	Identity      string   // proven | unknown | "" (not computed — e.g. the device is absent)
+	Reason        string   // set only when Identity == unknown
+}
+
+// Computed reports whether a verdict was reached at all.
+func (p PlaceIdentity) Computed() bool { return p.Identity != "" }
+
+// IdentityHost is the read surface the resolver needs. SysfsIdentityHost is production; tests use the
+// same struct pointed at fixture trees.
+type IdentityHost interface {
+	Mounts() ([]Mount, error)
+	// KernelName maps a /dev path (symlinks allowed: /dev/mapper/x, /dev/pve/y) to its kernel block name.
+	KernelName(dev string) (string, bool)
+	// Parent returns the parent disk's kernel name when name is a partition.
+	Parent(name string) (string, bool)
+	// Slaves lists /sys/class/block/<name>/slaves.
+	Slaves(name string) []string
+	// LoopBackingFile returns a loop device's backing file.
+	LoopBackingFile(name string) (string, bool)
+	// UdevProps returns the udev E: properties of a block device.
+	UdevProps(name string) map[string]string
+	// Virt returns the VM technology ("none" on bare metal).
+	Virt() string
+}
+
+// SysfsIdentityHost reads the real host. Every path is a field so tests can point it at fixtures.
+type SysfsIdentityHost struct {
+	SysClass   string                       // default /sys/class/block
+	UdevData   string                       // default /run/udev/data
+	ProcMounts string                       // default /proc/mounts
+	DevResolve func(string) (string, error) // default filepath.EvalSymlinks
+	VirtFn     func() string                // default `systemd-detect-virt --vm`, read once
+
+	virtOnce sync.Once
+	virt     string
+}
+
+// NewSysfsIdentityHost returns the production reader.
+func NewSysfsIdentityHost() *SysfsIdentityHost { return &SysfsIdentityHost{} }
+
+func (h *SysfsIdentityHost) sysClass() string {
+	if h.SysClass != "" {
+		return h.SysClass
+	}
+	return "/sys/class/block"
+}
+
+func (h *SysfsIdentityHost) Mounts() ([]Mount, error) {
+	return (&ProcHostReader{ProcMounts: h.ProcMounts}).Mounts()
+}
+
+func (h *SysfsIdentityHost) KernelName(dev string) (string, bool) {
+	if !strings.HasPrefix(dev, "/dev/") {
+		return "", false
+	}
+	resolve := h.DevResolve
+	if resolve == nil {
+		resolve = filepath.EvalSymlinks
+	}
+	real := dev
+	if r, err := resolve(dev); err == nil {
+		real = r
+	}
+	name := filepath.Base(real)
+	if _, err := os.Stat(filepath.Join(h.sysClass(), name)); err != nil {
+		return "", false
+	}
+	return name, true
+}
+
+func (h *SysfsIdentityHost) Parent(name string) (string, bool) {
+	p := filepath.Join(h.sysClass(), name)
+	if _, err := os.Stat(filepath.Join(p, "partition")); err != nil {
+		return "", false
+	}
+	real, err := filepath.EvalSymlinks(p)
+	if err != nil {
+		return "", false
+	}
+	return filepath.Base(filepath.Dir(real)), true
+}
+
+func (h *SysfsIdentityHost) Slaves(name string) []string {
+	entries, err := os.ReadDir(filepath.Join(h.sysClass(), name, "slaves"))
+	if err != nil {
+		return nil
+	}
+	out := make([]string, 0, len(entries))
+	for _, e := range entries {
+		out = append(out, e.Name())
+	}
+	sort.Strings(out)
+	return out
+}
+
+func (h *SysfsIdentityHost) LoopBackingFile(name string) (string, bool) {
+	if !strings.HasPrefix(name, "loop") {
+		return "", false
+	}
+	b, err := os.ReadFile(filepath.Join(h.sysClass(), name, "loop", "backing_file"))
+	if err != nil {
+		return "", false
+	}
+	f := strings.TrimSpace(string(b))
+	return f, f != ""
+}
+
+func (h *SysfsIdentityHost) UdevProps(name string) map[string]string {
+	out := map[string]string{}
+	devB, err := os.ReadFile(filepath.Join(h.sysClass(), name, "dev"))
+	if err != nil {
+		return out
+	}
+	dir := h.UdevData
+	if dir == "" {
+		dir = "/run/udev/data"
+	}
+	f, err := os.Open(filepath.Join(dir, "b"+strings.TrimSpace(string(devB))))
+	if err != nil {
+		return out
+	}
+	defer f.Close()
+	sc := bufio.NewScanner(f)
+	for sc.Scan() {
+		line := sc.Text()
+		if !strings.HasPrefix(line, "E:") {
+			continue
+		}
+		k, v, ok := strings.Cut(line[2:], "=")
+		if ok {
+			out[k] = v
+		}
+	}
+	return out
+}
+
+func (h *SysfsIdentityHost) Virt() string {
+	h.virtOnce.Do(func() {
+		if h.VirtFn != nil {
+			h.virt = h.VirtFn()
+			return
+		}
+		ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+		defer cancel()
+		// --vm: a container is not a reason (the agent runs on the host); exit 1 + "none" on metal.
+		out, _ := exec.CommandContext(ctx, "systemd-detect-virt", "--vm").Output()
+		h.virt = strings.TrimSpace(string(out))
+		if h.virt == "" {
+			h.virt = "unknown"
+		}
+	})
+	return h.virt
+}
+
+// IdentityResolver turns devices and host paths into PlaceIdentity.
+type IdentityResolver struct {
+	Host IdentityHost
+}
+
+// NewIdentityResolver returns a resolver over the real host.
+func NewIdentityResolver() *IdentityResolver {
+	return &IdentityResolver{Host: NewSysfsIdentityHost()}
+}
+
+var networkFSTypes = map[string]bool{
+	"nfs": true, "nfs4": true, "cifs": true, "smb3": true, "fuse.sshfs": true, "9p": true, "autofs": true, "ceph": true, "glusterfs": true,
+}
+
+// NetworkPlace is the fixed verdict for a network share or a remote store.
+func NetworkPlace() PlaceIdentity {
+	return PlaceIdentity{PhysicalDisks: []string{}, Identity: IdentityUnknown, Reason: ReasonNetwork}
+}
+
+// MountFor returns the mount that holds path (longest mountpoint prefix; the LAST entry wins on an equal
+// mountpoint, which is the one stacked on top — an automount's real filesystem over its autofs trigger).
+func MountFor(mounts []Mount, path string) (Mount, bool) {
+	p := filepath.Clean(path)
+	var best Mount
+	found := false
+	for _, m := range mounts {
+		mp := filepath.Clean(m.MountPoint)
+		if p != mp && !strings.HasPrefix(p, strings.TrimSuffix(mp, "/")+"/") {
+			continue
+		}
+		if !found || len(mp) >= len(filepath.Clean(best.MountPoint)) {
+			best, found = m, true
+		}
+	}
+	return best, found
+}
+
+// HostPath resolves the filesystem holding a HOST path.
+func (r *IdentityResolver) HostPath(path string) PlaceIdentity {
+	return r.hostPath(path, 0)
+}
+
+func (r *IdentityResolver) hostPath(path string, depth int) PlaceIdentity {
+	if r == nil || r.Host == nil || depth > 4 {
+		return PlaceIdentity{}
+	}
+	mounts, err := r.Host.Mounts()
+	if err != nil {
+		return PlaceIdentity{}
+	}
+	m, ok := MountFor(mounts, path)
+	if !ok {
+		return PlaceIdentity{}
+	}
+	if networkFSTypes[m.FSType] || !strings.HasPrefix(m.Device, "/dev/") {
+		return NetworkPlace()
+	}
+	return r.device(m.Device, depth)
+}
+
+// Device resolves a block device (a /dev path, symlinks allowed).
+func (r *IdentityResolver) Device(dev string) PlaceIdentity {
+	return r.device(dev, 0)
+}
+
+func (r *IdentityResolver) device(dev string, depth int) PlaceIdentity {
+	if r == nil || r.Host == nil || dev == "" {
+		return PlaceIdentity{}
+	}
+	name, ok := r.Host.KernelName(dev)
+	if !ok {
+		return PlaceIdentity{}
+	}
+	keys := map[string]bool{}
+	reason := ""
+	if !r.walk(name, depth, map[string]bool{}, keys, &reason) {
+		return PlaceIdentity{}
+	}
+	out := PlaceIdentity{PhysicalDisks: sortedKeys(keys), Identity: IdentityProven}
+	if reason != "" {
+		out.Identity, out.Reason = IdentityUnknown, reason
+	}
+	return out
+}
+
+// walk adds the keys of every physical disk under name; ok=false when a branch cannot be grounded.
+func (r *IdentityResolver) walk(name string, depth int, seen, keys map[string]bool, reason *string) bool {
+	if seen[name] {
+		return true // a diamond (a thin pool's tmeta and tdata on one partition): already counted
+	}
+	if len(seen) > 64 {
+		return false
+	}
+	seen[name] = true
+	if parent, ok := r.Host.Parent(name); ok {
+		return r.walk(parent, depth, seen, keys, reason)
+	}
+	if file, ok := r.Host.LoopBackingFile(name); ok {
+		p := r.hostPath(file, depth+1)
+		if !p.Computed() {
+			return false
+		}
+		for _, k := range p.PhysicalDisks {
+			keys[k] = true
+		}
+		noteReason(reason, p.Reason)
+		return true
+	}
+	if slaves := r.Host.Slaves(name); len(slaves) > 0 {
+		for _, s := range slaves {
+			if !r.walk(s, depth, seen, keys, reason) {
+				return false
+			}
+		}
+		return true
+	}
+	if strings.HasPrefix(name, "dm-") || strings.HasPrefix(name, "md") || strings.HasPrefix(name, "loop") {
+		return false // a virtual device with nothing beneath it cannot be grounded
+	}
+	key, why := r.diskKey(name)
+	keys[key] = true
+	noteReason(reason, why)
+	return true
+}
+
+func noteReason(cur *string, why string) {
+	if why == "" {
+		return
+	}
+	if *cur == "" || reasonRank[why] < reasonRank[*cur] {
+		*cur = why
+	}
+}
+
+// diskKey names one whole disk and says why it is not trusted ("" when it is).
+func (r *IdentityResolver) diskKey(name string) (key, reason string) {
+	p := r.Host.UdevProps(name)
+	serial := strings.TrimSpace(p["ID_SERIAL_SHORT"])
+	model := strings.TrimSpace(p["ID_MODEL"])
+	bus := strings.TrimSpace(p["ID_BUS"])
+	wwn := strings.ToLower(strings.TrimSpace(p["ID_WWN"]))
+	if bus == "" && strings.HasPrefix(name, "nvme") {
+		bus = "nvme"
+	}
+	usb := bus == "usb" || strings.Contains(p["DEVPATH"], "/usb")
+	switch {
+	case strings.HasPrefix(name, "nvme") && (strings.HasPrefix(wwn, "eui.") || strings.HasPrefix(wwn, "nvme.")):
+		key = "eui:" + strings.TrimPrefix(strings.TrimPrefix(wwn, "eui."), "nvme.")
+	case serial != "":
+		key = "serial:" + bus + ":" + model + ":" + serial
+		if usb && FakeBridgeSerial(serial) {
+			reason = ReasonUSBBridge
+		}
+	case wwn != "" && WWNHasVendor(wwn):
+		key = "wwn:" + wwn
+	case wwn != "":
+		key, reason = "wwn:"+wwn, ReasonWeakWWN
+	default:
+		key, reason = "dev:"+name, ReasonNoSerial
+	}
+	if v := r.Host.Virt(); v != "none" {
+		reason = ReasonVM // inside a VM a virtual disk says nothing about the host disk (spike cases 5–6)
+	}
+	return key, reason
+}
+
+// WWNHasVendor reports whether an NAA-5/6 WWN carries a registered vendor (OUI ≠ 000000). Other NAA
+// formats are taken as they are. Measured: the N100's AirDisk SSD reports 0x5000000123456f1f.
+func WWNHasVendor(wwn string) bool {
+	h := strings.TrimPrefix(strings.ToLower(wwn), "0x")
+	if len(h) < 7 {
+		return false
+	}
+	if h[0] == '5' || h[0] == '6' {
+		return h[1:7] != "000000"
+	}
+	return true
+}
+
+// FakeBridgeSerial reports a serial that a USB bridge reports for any disk: too short, one repeated
+// character, or a known placeholder. UNPROVEN on a real enclosure (R-1077) — the rule is the design's.
+func FakeBridgeSerial(s string) bool {
+	s = strings.ToUpper(strings.TrimSpace(s))
+	if len(s) < 6 {
+		return true
+	}
+	if strings.Count(s, s[:1]) == len(s) {
+		return true
+	}
+	switch s {
+	case "123456789", "1234567890", "123456789ABC", "0123456789ABCDEF", "1234567890ABCDEF", "DEADBEEF":
+		return true
+	}
+	return false
+}
+
+func sortedKeys(m map[string]bool) []string {
+	out := make([]string, 0, len(m))
+	for k := range m {
+		out = append(out, k)
+	}
+	sort.Strings(out)
+	return out
+}
+
+// Relation compares two places (`16` §4). A SHARED disk key is a positive fact, so two places on one disk
+// are "same, proven" even when that disk is unknown (inside a VM, two parts of one virtual disk). Otherwise
+// proven is true only when both verdicts are proven; different then means "no disk in common". When proven
+// is false, different only says the known keys do not overlap — callers must act on proven first.
+// reason: "" (proven, different), "same_disk", "unmapped" (a side has no verdict), or the unknown reason.
+func Relation(a, b PlaceIdentity) (different, proven bool, reason string) {
+	if !a.Computed() || !b.Computed() {
+		return false, false, "unmapped"
+	}
+	set := map[string]bool{}
+	for _, k := range a.PhysicalDisks {
+		set[k] = true
+	}
+	shared := false
+	for _, k := range b.PhysicalDisks {
+		if set[k] {
+			shared = true
+		}
+	}
+	if shared {
+		return false, true, "same_disk"
+	}
+	if a.Identity == IdentityProven && b.Identity == IdentityProven {
+		return true, true, ""
+	}
+	why := a.Reason
+	if why == "" || (b.Reason != "" && reasonRank[b.Reason] < reasonRank[why]) {
+		why = b.Reason
+	}
+	return true, false, why
+}

internal/storage/identity_test.go

@@ -0,0 +1,257 @@
+package storage
+
+import (
+	"os"
+	"path/filepath"
+	"reflect"
+	"strconv"
+	"strings"
+	"testing"
+)
+
+// Fixture /sys/class/block + /run/udev/data + /proc/mounts trees, one per `16` §3 case. The tree mimics the
+// kernel: /sys/devices/.../<disk>/<part> real dirs (a partition carries a "partition" file), /sys/class/block
+// symlinks, a "dev" file per node, udev "E:" lines keyed by major:minor. The cases are the ones the spike
+// measured on real devices (felhom.eu documentation/audits/storage-identity-2026-10-15/).
+
+type fx struct {
+	t      *testing.T
+	root   string
+	minor  int
+	mounts []string
+	links  map[string]string // /dev path -> kernel name
+	virt   string
+}
+
+func newFx(t *testing.T) *fx {
+	t.Helper()
+	r := t.TempDir()
+	for _, d := range []string{"sys/class/block", "sys/devices", "udev"} {
+		if err := os.MkdirAll(filepath.Join(r, d), 0o755); err != nil {
+			t.Fatal(err)
+		}
+	}
+	return &fx{t: t, root: r, links: map[string]string{}, virt: "none"}
+}
+
+func (f *fx) write(p, s string) {
+	f.t.Helper()
+	if err := os.MkdirAll(filepath.Dir(p), 0o755); err != nil {
+		f.t.Fatal(err)
+	}
+	if err := os.WriteFile(p, []byte(s), 0o644); err != nil {
+		f.t.Fatal(err)
+	}
+}
+
+// node makes a block node. parent "" → a top-level device. udev lines are "KEY=VAL".
+func (f *fx) node(name, parent string, udev ...string) {
+	f.t.Helper()
+	real := filepath.Join(f.root, "sys/devices", name)
+	if parent != "" {
+		pr, _ := filepath.EvalSymlinks(filepath.Join(f.root, "sys/class/block", parent))
+		real = filepath.Join(pr, name)
+		f.write(filepath.Join(real, "partition"), "1\n")
+	}
+	f.minor++
+	dev := "8:" + strconv.Itoa(f.minor)
+	f.write(filepath.Join(real, "dev"), dev+"\n")
+	if err := os.Symlink(real, filepath.Join(f.root, "sys/class/block", name)); err != nil {
+		f.t.Fatal(err)
+	}
+	var b strings.Builder
+	for _, u := range udev {
+		b.WriteString("E:" + u + "\n")
+	}
+	f.write(filepath.Join(f.root, "udev", "b"+dev), b.String())
+	f.links["/dev/"+name] = name
+}
+
+func (f *fx) slaves(name string, slaves ...string) {
+	f.t.Helper()
+	real, _ := filepath.EvalSymlinks(filepath.Join(f.root, "sys/class/block", name))
+	for _, s := range slaves {
+		if err := os.MkdirAll(filepath.Join(real, "slaves", s), 0o755); err != nil {
+			f.t.Fatal(err)
+		}
+	}
+}
+
+func (f *fx) loop(name, file string) {
+	real, _ := filepath.EvalSymlinks(filepath.Join(f.root, "sys/class/block", name))
+	f.write(filepath.Join(real, "loop", "backing_file"), file+"\n")
+}
+
+func (f *fx) mount(dev, mp, fstype string) {
+	f.mounts = append(f.mounts, dev+" "+mp+" "+fstype+" rw 0 0")
+}
+
+func (f *fx) resolver() *IdentityResolver {
+	f.write(filepath.Join(f.root, "mounts"), strings.Join(f.mounts, "\n")+"\n")
+	return &IdentityResolver{Host: &SysfsIdentityHost{
+		SysClass:   filepath.Join(f.root, "sys/class/block"),
+		UdevData:   filepath.Join(f.root, "udev"),
+		ProcMounts: filepath.Join(f.root, "mounts"),
+		DevResolve: func(p string) (string, error) {
+			if n, ok := f.links[p]; ok {
+				return "/dev/" + n, nil
+			}
+			return p, nil
+		},
+		VirtFn: func() string { return f.virt },
+	}}
+}
+
+var sanDisk = []string{"ID_BUS=ata", "ID_MODEL=SanDisk_X600", "ID_SERIAL_SHORT=182195804614", "ID_WWN=0x5001b448b60a9d6e"}
+
+// oneDiskLVM builds the PVE default layout on one SATA SSD: sda3 → pve-root, pve-data tmeta/tdata → tpool →
+// a guest volume (the shape of all three test boxes).
+func oneDiskLVM(f *fx) {
+	f.node("sda", "", sanDisk...)
+	f.node("sda3", "sda")
+	for _, d := range []string{"dm-1", "dm-2", "dm-3", "dm-4", "dm-6"} {
+		f.node(d, "")
+	}
+	f.slaves("dm-1", "sda3")
+	f.slaves("dm-2", "sda3")
+	f.slaves("dm-3", "sda3")
+	f.slaves("dm-4", "dm-2", "dm-3")
+	f.slaves("dm-6", "dm-4")
+	f.links["/dev/mapper/pve-root"] = "dm-1"
+	f.links["/dev/mapper/pve-data-tpool"] = "dm-4"
+	f.links["/dev/pve/vm-9201-disk-1"] = "dm-6"
+	f.mount("/dev/mapper/pve-root", "/", "ext4")
+}
+
+func TestIdentity_OneDiskLVM_AllPlacesSameProvenDisk(t *testing.T) {
+	f := newFx(t)
+	oneDiskLVM(f)
+	r := f.resolver()
+	want := []string{"serial:ata:SanDisk_X600:182195804614"}
+	for _, p := range []PlaceIdentity{r.HostPath("/var/lib/vz"), r.Device("/dev/mapper/pve-data-tpool"), r.Device("/dev/pve/vm-9201-disk-1")} {
+		if p.Identity != IdentityProven || !reflect.DeepEqual(p.PhysicalDisks, want) {
+			t.Fatalf("want proven %v, got %+v", want, p)
+		}
+	}
+	diff, proven, why := Relation(r.HostPath("/var/lib/vz"), r.Device("/dev/pve/vm-9201-disk-1"))
+	if diff || !proven || why != "same_disk" {
+		t.Fatalf("root vs guest volume: want same+proven, got diff=%v proven=%v why=%q", diff, proven, why)
+	}
+}
+
+func TestIdentity_TwoDisks_ProvenDifferent(t *testing.T) {
+	f := newFx(t)
+	oneDiskLVM(f)
+	f.node("sdb", "", "ID_BUS=ata", "ID_MODEL=TOSHIBA_MQ04ABF100", "ID_SERIAL_SHORT=65NOP3HDT", "ID_WWN=0x5000039ddb108568", "DEVPATH=/devices/pci0000:00/0000:00:14.0/usb1/1-3/block/sdb")
+	f.mount("/dev/sdb", "/mnt/hdd_1", "ext4")
+	f.mount("/dev/sdb", "/mnt/felhom-drives/hdd_1", "ext4") // the felhom-data bind shows the source device
+	r := f.resolver()
+	drive := r.HostPath("/mnt/felhom-drives/hdd_1/userdata")
+	if drive.Identity != IdentityProven || drive.PhysicalDisks[0] != "serial:ata:TOSHIBA_MQ04ABF100:65NOP3HDT" {
+		t.Fatalf("USB drive passing its identity: want proven, got %+v", drive)
+	}
+	diff, proven, why := Relation(r.Device("/dev/pve/vm-9201-disk-1"), drive)
+	if !diff || !proven || why != "" {
+		t.Fatalf("system vs drive: want different+proven, got %v %v %q", diff, proven, why)
+	}
+}
+
+func TestIdentity_TwoPartitionsOneDisk_Same(t *testing.T) {
+	f := newFx(t)
+	f.node("sdc", "", "ID_BUS=usb", "ID_MODEL=Portable", "ID_SERIAL_SHORT=NAAB7C21")
+	f.node("sdc1", "sdc")
+	f.node("sdc2", "sdc")
+	f.mount("/dev/sdc1", "/mnt/a", "ext4")
+	f.mount("/dev/sdc2", "/mnt/b", "ext4")
+	r := f.resolver()
+	diff, proven, why := Relation(r.HostPath("/mnt/a"), r.HostPath("/mnt/b"))
+	if diff || !proven || why != "same_disk" {
+		t.Fatalf("two partitions of one disk: want same+proven, got %v %v %q", diff, proven, why)
+	}
+}
+
+func TestIdentity_LoopPartitionsOnAFile_ResolveToTheFilesDisk(t *testing.T) {
+	f := newFx(t)
+	f.node("nvme0n1", "", "ID_MODEL=KXG50PNV1T02", "ID_SERIAL_SHORT=58BS11AFT8MQ", "ID_WWN=eui.000000000000001000080d020046c07b")
+	f.mount("/dev/nvme0n1", "/mnt/hdd_1", "ext4")
+	f.node("loop2", "")
+	f.loop("loop2", "/mnt/hdd_1/scratch.img")
+	f.node("loop2p1", "loop2") // the partition step must run BEFORE the loop step (the spike's probe bug)
+	f.mount("/dev/loop2p1", "/tmp/scrA", "ext4")
+	r := f.resolver()
+	p := r.HostPath("/tmp/scrA")
+	if p.Identity != IdentityProven || !reflect.DeepEqual(p.PhysicalDisks, []string{"eui:000000000000001000080d020046c07b"}) {
+		t.Fatalf("loop partition: want the NVMe's eui, got %+v", p)
+	}
+}
+
+func TestIdentity_WWNWithoutVendor(t *testing.T) {
+	f := newFx(t)
+	f.node("sda", "", "ID_BUS=ata", "ID_MODEL=AirDisk_512GB_SSD", "ID_SERIAL_SHORT=QDF922W009654S30EX", "ID_WWN=0x5000000123456f1f")
+	f.node("sdb", "", "ID_BUS=ata", "ID_MODEL=Cheap", "ID_WWN=0x5000000123456f1f")
+	r := f.resolver()
+	if p := r.Device("/dev/sda"); p.Identity != IdentityProven || p.PhysicalDisks[0] != "serial:ata:AirDisk_512GB_SSD:QDF922W009654S30EX" {
+		t.Fatalf("a real serial wins over a vendor-less WWN: got %+v", p)
+	}
+	if p := r.Device("/dev/sdb"); p.Identity != IdentityUnknown || p.Reason != ReasonWeakWWN {
+		t.Fatalf("no serial + vendor-less WWN: want unknown/weak_wwn, got %+v", p)
+	}
+}
+
+func TestIdentity_NoSerial(t *testing.T) {
+	f := newFx(t)
+	f.node("vdb", "", "ID_MODEL=")
+	if p := f.resolver().Device("/dev/vdb"); p.Identity != IdentityUnknown || p.Reason != ReasonNoSerial {
+		t.Fatalf("want unknown/no_serial, got %+v", p)
+	}
+}
+
+func TestIdentity_InsideAVM_EverythingUnknown(t *testing.T) {
+	f := newFx(t)
+	oneDiskLVM(f)
+	f.virt = "kvm"
+	f.node("sdb", "", "ID_BUS=scsi", "ID_MODEL=QEMU_HARDDISK", "ID_SERIAL_SHORT=drive-scsi1")
+	f.mount("/dev/sdb", "/mnt/felhom-drives/adatlemez", "ext4")
+	r := f.resolver()
+	a, b := r.Device("/dev/pve/vm-9201-disk-1"), r.HostPath("/mnt/felhom-drives/adatlemez")
+	for _, p := range []PlaceIdentity{a, b} {
+		if p.Identity != IdentityUnknown || p.Reason != ReasonVM {
+			t.Fatalf("inside a VM: want unknown/vm, got %+v", p)
+		}
+	}
+	diff, proven, why := Relation(a, b)
+	if proven || why != ReasonVM || !diff {
+		t.Fatalf("two virtual disks: want not proven (vm), got %v %v %q", diff, proven, why)
+	}
+}
+
+func TestIdentity_USBBridgeSerial(t *testing.T) {
+	f := newFx(t)
+	f.node("sdd", "", "ID_BUS=usb", "ID_MODEL=External_USB3.0", "ID_SERIAL_SHORT=000000000000")
+	f.node("sde", "", "ID_BUS=usb", "ID_MODEL=External_USB3.0", "ID_SERIAL_SHORT=20231106A1B2")
+	r := f.resolver()
+	if p := r.Device("/dev/sdd"); p.Identity != IdentityUnknown || p.Reason != ReasonUSBBridge {
+		t.Fatalf("a bridge placeholder serial: want unknown/usb_bridge, got %+v", p)
+	}
+	if p := r.Device("/dev/sde"); p.Identity != IdentityProven {
+		t.Fatalf("a USB disk with a plausible serial: want proven, got %+v", p)
+	}
+}
+
+func TestIdentity_NetworkShareAndUnmapped(t *testing.T) {
+	f := newFx(t)
+	oneDiskLVM(f)
+	f.mount("//192.168.0.123/TesztFilmek", "/mnt/felhom-drives/omv-filmek", "cifs")
+	r := f.resolver()
+	n := r.HostPath("/mnt/felhom-drives/omv-filmek/x")
+	if n.Identity != IdentityUnknown || n.Reason != ReasonNetwork || len(n.PhysicalDisks) != 0 {
+		t.Fatalf("network share: want unknown/network with no disks, got %+v", n)
+	}
+	_, proven, why := Relation(r.HostPath("/"), n)
+	if proven || why != ReasonNetwork {
+		t.Fatalf("system vs network: want not proven (network), got %v %q", proven, why)
+	}
+	if _, proven, why := Relation(PlaceIdentity{}, n); proven || why != "unmapped" {
+		t.Fatalf("an uncomputed side: want unmapped, got %v %q", proven, why)
+	}
+}

internal/storage/intent.go

@@ -17,12 +17,12 @@ import (
 // Four states (3A):
 //   - new            : durable-id NOT in the store. Never auto-mounted (the user must enroll it).
 //   - enrolled       : desired = mounted + bound into the guest. Drift (present && unmounted) with
-//                      this intent → reconcile.
+//     this intent → reconcile.
 //   - ejected        : an intentional, temporary unmount, set ONLY via the official eject endpoint.
-//                      Present && unmounted → left alone. CLEARED to `enrolled` when the drive goes
-//                      physically ABSENT, so a replug auto-mounts (the replug rule, for free).
+//     Present && unmounted → left alone. CLEARED to `enrolled` when the drive goes
+//     physically ABSENT, so a replug auto-mounts (the replug rule, for free).
 //   - decommissioned : permanent. Never auto-mounted; SURVIVES absent/present; cleared only by an
-//                      explicit re-commission.
+//     explicit re-commission.
 //
 // The elegant invariant: intent is recorded ONLY through the official enroll/eject/decommission
 // paths. "Is it unmounted" ≠ "did someone officially eject it."

internal/storage/mkfs_guarded_suite_test.go

@@ -0,0 +1,24 @@
+package storage
+
+import (
+	"os/exec"
+	"path/filepath"
+	"runtime"
+	"testing"
+)
+
+// The root mkfs wrapper's own tests (configs/test_felhom_mkfs_guarded.py, agent v0.158.0 — a sibling partition that
+// is Felhom's own drive no longer blocks formatting the part next to it; every other mount still refuses) run with
+// the Go suite.
+func TestMkfsGuardedPythonSuite(t *testing.T) {
+	py, err := exec.LookPath("python3")
+	if err != nil {
+		t.Skip("python3 not found")
+	}
+	_, file, _, _ := runtime.Caller(0)
+	script := filepath.Join(filepath.Dir(file), "..", "..", "configs", "test_felhom_mkfs_guarded.py")
+	out, err := exec.Command(py, script).CombinedOutput()
+	if err != nil {
+		t.Fatalf("wrapper suite failed: %v\n%s", err, out)
+	}
+}

internal/storage/netverify_test.go

@@ -93,11 +93,11 @@ systemd-1 /mnt/felhom-drives/live autofs rw,relatime,fd=86 0 0
 		where string
 		want  bool
 	}{
-		{"/mnt/felhom-drives/live", true},   // real nfs4 (the autofs line for the same path must not confuse it)
-		{"/mnt/felhom-drives/smb", true},    // cifs
-		{"/mnt/felhom-drives/idle", false},  // autofs trigger ONLY — idle automount is NOT mounted
-		{"/mnt/felhom-drives/disk", false},  // a local fs at the path is not a network mount
-		{"/mnt/felhom-drives/nope", false},  // absent
+		{"/mnt/felhom-drives/live", true},  // real nfs4 (the autofs line for the same path must not confuse it)
+		{"/mnt/felhom-drives/smb", true},   // cifs
+		{"/mnt/felhom-drives/idle", false}, // autofs trigger ONLY — idle automount is NOT mounted
+		{"/mnt/felhom-drives/disk", false}, // a local fs at the path is not a network mount
+		{"/mnt/felhom-drives/nope", false}, // absent
 	}
 	for _, tc := range cases {
 		if got := networkMountedIn(procMounts, tc.where); got != tc.want {

internal/storage/observe.go

@@ -40,6 +40,10 @@ type Observer struct {
 	// R-672 (v0.133.0): a thin pool crossing thinPoolAlarmFraction (data OR metadata) requests an
 	// out-of-band host report at once, so the hub's storage-fill alarm sees it in seconds instead of at
 	// the next 15-minute report. Rising edge per pool; re-armed below thinPoolRearmFraction.
+	// ident (agent v0.157.0, `16` §3) computes each target's physical-disk identity. nil → the fields stay
+	// empty (the pre-0.157 shape), which every consumer reads as "not computed".
+	ident *IdentityResolver
+
 	highMu     sync.Mutex
 	high       map[string]bool
 	onThinHigh func()
@@ -50,6 +54,51 @@ const (
 	thinPoolRearmFraction = 0.85
 )
 
+// SetIdentity wires the physical-disk identity resolver (main: NewIdentityResolver()).
+func (o *Observer) SetIdentity(r *IdentityResolver) { o.ident = r }
+
+// placeIdentity fills a target's identity fields (`16` §3). Network and remote stores are fixed
+// "unknown/network"; a dir storage is judged only on its OWN mount (the same rule as backing_device —
+// never the containing filesystem, so an absent drive is "not computed", never the root disk); the
+// builtin local and an lvmthin pool are the system disk(s) they sit on.
+func (o *Observer) placeIdentity(tgt *hub.StorageTarget, s proxmox.Storage) {
+	if o.ident == nil {
+		return
+	}
+	var p PlaceIdentity
+	switch tgt.Type {
+	case hub.StorageTypeNFS, hub.StorageTypeCIFS, hub.StorageTypePBS:
+		tgt.Kind = PlaceKindNetwork
+		p = NetworkPlace()
+	case hub.StorageTypeLocal:
+		tgt.Kind = PlaceKindSystem
+		if s.Path != "" {
+			p = o.ident.HostPath(s.Path)
+		}
+	case hub.StorageTypeLVMThin:
+		tgt.Kind = PlaceKindSystem
+		if s.VGName != "" && s.ThinPool != "" {
+			p = o.ident.Device("/dev/mapper/" + dmEscape(s.VGName) + "-" + dmEscape(s.ThinPool) + "-tpool")
+		}
+	default:
+		if tgt.BackingDevice == "" {
+			return // not its own mount (absent, or a dir on root): not computed
+		}
+		tgt.Kind = PlaceKindDisk
+		if _, known := wholeDiskOf(tgt.BackingDevice); known && !reWholeDisk.MatchString(tgt.BackingDevice) {
+			tgt.Kind = PlaceKindPartition
+		}
+		p = o.ident.Device(tgt.BackingDevice)
+	}
+	if !p.Computed() {
+		return
+	}
+	tgt.PhysicalDisks, tgt.Identity, tgt.IdentityReason = p.PhysicalDisks, p.Identity, p.Reason
+}
+
+// dmEscape doubles dashes the way device-mapper names an LV (pve + vm-1-disk-0 → pve-vm--1--disk--0).
+func dmEscape(s string) string { return strings.ReplaceAll(s, "-", "--") }
+
 // SetThinHighTrigger wires the out-of-band report request (main: the storage trigger channel).
 func (o *Observer) SetThinHighTrigger(f func()) { o.onThinHigh = f }
 
@@ -303,6 +352,9 @@ func (o *Observer) build(s proxmox.Storage, mounts []Mount) observed {
 		o.noteThinFill(s.Storage+"/data", frac)
 	}
 
+	// Physical-disk identity (`16` §3) — after typ/backing are settled; never from a caller's claim.
+	o.placeIdentity(&tgt, s)
+
 	// SMART-only device hint (v0.95.0): a dir-storage that lives INSIDE a shared filesystem (the
 	// builtin `local` on the LVM root) has empty backing by design, yet we can still read the PHYSICAL
 	// disk's SMART by resolving its containing filesystem. Gated to catDir + no own backing + reachable,

internal/storage/observe_identity_test.go

@@ -0,0 +1,73 @@
+package storage
+
+import (
+	"context"
+	"reflect"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// The report rows carry the `16` §3 identity per storage type: local + lvmthin → the system disk; a dir
+// on its own partition → that disk, kind "partition"; nfs/pbs → network; an UNMOUNTED dir → not computed
+// (never the root disk — the same rule as backing_device).
+func TestObserve_FillsPhysicalDiskIdentity(t *testing.T) {
+	f := newFx(t)
+	oneDiskLVM(f)
+	f.node("sdb", "", "ID_BUS=ata", "ID_MODEL=TOSHIBA_MQ04ABF100", "ID_SERIAL_SHORT=65NOP3HDT")
+	f.node("sdb1", "sdb")
+	f.mount("/dev/sdb1", "/mnt/usb-backup", "ext4")
+	res := f.resolver()
+
+	api := &fakeStorageAPI{
+		node: "demo",
+		cluster: []proxmox.Storage{
+			{Storage: "local", Type: "dir", Path: "/var/lib/vz"},
+			{Storage: "local-lvm", Type: "lvmthin", VGName: "pve", ThinPool: "data"},
+			{Storage: "usb-backup", Type: "dir", Path: "/mnt/usb-backup"},
+			{Storage: "gone", Type: "dir", Path: "/mnt/gone"},
+			{Storage: "nfs-arch", Type: "nfs", Server: "10.0.0.5", Export: "/x"},
+		},
+		nodeSt: []proxmox.Storage{
+			{Storage: "local", Type: "dir", Path: "/var/lib/vz", Active: 1},
+			{Storage: "local-lvm", Type: "lvmthin", Active: 1},
+			{Storage: "usb-backup", Type: "dir", Path: "/mnt/usb-backup", Active: 1},
+			{Storage: "gone", Type: "dir", Path: "/mnt/gone", Active: 1},
+			{Storage: "nfs-arch", Type: "nfs", Active: 1},
+		},
+	}
+	host := &fakeHostReader{
+		mounts: []Mount{
+			{Device: "/dev/mapper/pve-root", MountPoint: "/", FSType: "ext4"},
+			{Device: "/dev/sdb1", MountPoint: "/mnt/usb-backup", FSType: "ext4"},
+		},
+		exists: map[string]bool{"/dev/sdb1": true, "/dev/mapper/pve-root": true},
+	}
+	ob := NewObserver(api, host, nil, quietLogger())
+	ob.SetIdentity(res)
+	got, err := ob.Observe(context.Background())
+	if err != nil {
+		t.Fatal(err)
+	}
+	m := byName(got)
+	ssd := []string{"serial:ata:SanDisk_X600:182195804614"}
+	for _, n := range []string{"local", "local-lvm"} {
+		if tg := m[n]; tg.Kind != "system" || tg.Identity != "proven" || !reflect.DeepEqual(tg.PhysicalDisks, ssd) {
+			t.Errorf("%s: want system/proven on the SSD, got kind=%q id=%q disks=%v", n, tg.Kind, tg.Identity, tg.PhysicalDisks)
+		}
+	}
+	if tg := m["usb-backup"]; tg.Kind != "partition" || tg.Identity != "proven" || tg.PhysicalDisks[0] != "serial:ata:TOSHIBA_MQ04ABF100:65NOP3HDT" {
+		t.Errorf("usb-backup: got %+v", tg)
+	}
+	if tg := m["nfs-arch"]; tg.Kind != "network" || tg.IdentityReason != "network" {
+		t.Errorf("nfs: got kind=%q reason=%q", tg.Kind, tg.IdentityReason)
+	}
+	if tg := m["gone"]; tg.Identity != "" || len(tg.PhysicalDisks) != 0 {
+		t.Errorf("an unmounted dir must not be computed (never root's disk), got %+v", tg)
+	}
+	// Without a resolver the fields stay empty (the pre-0.157 shape).
+	plain, _ := NewObserver(api, host, nil, quietLogger()).Observe(context.Background())
+	if tg := byName(plain)["local"]; tg.Identity != "" || tg.Kind != "" {
+		t.Errorf("no resolver: want empty identity, got %+v", tg)
+	}
+}

internal/storage/partition_durable_test.go

@@ -0,0 +1,112 @@
+package storage
+
+import (
+	"os"
+	"path/filepath"
+	"runtime"
+	"strings"
+	"testing"
+)
+
+// The data-bearing partition path binds the disk id + the part number + the fs-UUID (felhom.eu architecture/16 §6,
+// agent v0.158.0). The wipe re-resolve (localapi antiRetargetResolveExpect) formats only when resolve(id) succeeds AND
+// DeviceDurableID(that device) == id — the round trip these tests pin, in the three ways a target can move.
+
+type partFixture struct{ base, root string }
+
+func newPartFixture(t *testing.T) partFixture {
+	t.Helper()
+	if runtime.GOOS == "windows" {
+		t.Skip("symlink fixtures run on Linux")
+	}
+	f := partFixture{base: t.TempDir()}
+	f.root = filepath.Join(f.base, "disk")
+	for _, d := range []string{"sdb", "sdb1", "sdb2", "sdc2"} {
+		if err := os.WriteFile(filepath.Join(f.base, d), []byte("x"), 0o600); err != nil {
+			t.Fatal(err)
+		}
+	}
+	os.MkdirAll(filepath.Join(f.root, "by-id"), 0o755)
+	os.MkdirAll(filepath.Join(f.root, "by-uuid"), 0o755)
+	withDevDiskRoot(t, f.root)
+	return f
+}
+
+func (f partFixture) link(t *testing.T, sub, name, dev string) {
+	t.Helper()
+	p := filepath.Join(f.root, sub, name)
+	os.Remove(p)
+	if err := os.Symlink(filepath.Join(f.base, dev), p); err != nil {
+		t.Fatal(err)
+	}
+}
+
+func (f partFixture) dev(name string) string { return filepath.Join(f.base, name) }
+
+func (f partFixture) standard(t *testing.T) {
+	f.link(t, "by-id", "wwn-0x5000aaaa", "sdb")
+	f.link(t, "by-id", "ata-TOSHIBA_MQ04_65NOP-part1", "sdb1")
+	f.link(t, "by-id", "wwn-0x5000aaaa-part1", "sdb1")
+	f.link(t, "by-id", "wwn-0x5000aaaa-part2", "sdb2")
+	f.link(t, "by-uuid", "1111-films", "sdb2")
+}
+
+func roundTrip(id string) (string, bool) {
+	dev, err := ResolveDurableDevice(id)
+	if err != nil {
+		return "", false
+	}
+	again, err := DeviceDurableID(dev)
+	return dev, err == nil && again == id
+}
+
+func TestPartitionWipeID_BindsDiskPartAndFilesystem(t *testing.T) {
+	f := newPartFixture(t)
+	f.standard(t)
+	id, err := DeviceDurableID(f.dev("sdb2"))
+	if err != nil || id != "byid:wwn-0x5000aaaa-part2+fsuuid:1111-films" {
+		t.Fatalf("a part with a filesystem: want disk id + part + fs-UUID, got %q %v", id, err)
+	}
+	if dev, ok := roundTrip(id); !ok || dev != f.dev("sdb2") {
+		t.Fatalf("the unchanged part must round-trip, got %q %v", dev, ok)
+	}
+	// A blank part (no filesystem) and a whole disk keep the plain by-id binding.
+	if id1, _ := DeviceDurableID(f.dev("sdb1")); id1 != "byid:wwn-0x5000aaaa-part1" {
+		t.Fatalf("a blank part: want the plain part link, got %q", id1)
+	}
+	if idDisk, _ := DeviceDurableID(f.dev("sdb")); idDisk != "byid:wwn-0x5000aaaa" {
+		t.Fatalf("a whole disk keeps today's id, got %q", idDisk)
+	}
+}
+
+func TestPartitionWipeID_ReformattedPartRefuses(t *testing.T) {
+	f := newPartFixture(t)
+	f.standard(t)
+	id, _ := DeviceDurableID(f.dev("sdb2"))
+	os.Remove(filepath.Join(f.root, "by-uuid", "1111-films"))
+	f.link(t, "by-uuid", "2222-other", "sdb2") // someone reformatted the part since the household confirmed
+	if _, ok := roundTrip(id); ok {
+		t.Fatalf("a reformatted part must not round-trip (the wipe would hit a different filesystem)")
+	}
+}
+
+func TestPartitionWipeID_LinkMovedToAnotherDiskRefuses(t *testing.T) {
+	f := newPartFixture(t)
+	f.standard(t)
+	id, _ := DeviceDurableID(f.dev("sdb2"))
+	f.link(t, "by-id", "wwn-0x5000aaaa-part2", "sdc2") // a different node took the part link; the fs is still sdb2
+	if _, err := ResolveDurableDevice(id); err == nil || !strings.Contains(err.Error(), "different devices") {
+		t.Fatalf("the part link and its filesystem on different devices must refuse, got %v", err)
+	}
+}
+
+func TestPartitionWipeID_RefusesAForgedCompound(t *testing.T) {
+	f := newPartFixture(t)
+	f.standard(t)
+	f.link(t, "by-uuid", "3333-whole", "sdb") // a filesystem straight on the whole disk: both halves name sdb
+	for _, id := range []string{"byid:wwn-0x5000aaaa+fsuuid:3333-whole", "byid:wwn-0x5000aaaa-part2+fsuuid:../x"} {
+		if _, err := ResolveDurableDevice(id); err == nil {
+			t.Fatalf("%q must be refused", id)
+		}
+	}
+}

internal/storage/partitions.go

@@ -0,0 +1,123 @@
+package storage
+
+import (
+	"encoding/json"
+	"os"
+	"path/filepath"
+	"strconv"
+	"strings"
+)
+
+// Partitions as places (agent v0.158.0) — felhom.eu architecture/16-storage-physical-identity.md §6.
+//
+// A disk that passed the WHOLE-disk unclaimed filter (classifyClaim: not the system disk, no LVM/ZFS/RAID/LUKS/swap
+// member on any node, no foreign mount on any node) also lists its partitions, each with its own state:
+//
+//	empty       no filesystem, no children           → „format this part" (self-serve: the agent's probe says blank) —
+//	                                                   only while NO part of the disk is mounted (Formattable)
+//	filesystem  ext4/xfs (mountable) or another fs   → „use this part" when mountable; „format this part" only through
+//	                                                   the customer-confirmed data-bearing path (the agent's own probe)
+//	in_use      already Felhom's own drive           → shown, never offered (it is a registered drive already)
+//
+// A partition with children (LVM, a mapper) is not listed. The design's separate „a filesystem with no files" state is
+// folded into `filesystem` on purpose: telling the two apart needs a root read-only mount, and the data-bearing path's
+// confirmation is the safer default for both (`16` §13). Creating, deleting or resizing a partition: never here (D5).
+// The format binding is the partition's own by-id link (…-partN: the disk id + the partition number) via
+// DeviceDurableID, re-resolved before mkfs (wipe_reresolve.go); the attach key is its fs-UUID.
+type CandidatePartition struct {
+	Device    string `json:"device"` // e.g. /dev/sdb2
+	Number    int    `json:"number,omitempty"`
+	SizeBytes int64  `json:"size_bytes"`
+	FSType    string `json:"fstype,omitempty"`
+	State     string `json:"state"` // empty | filesystem | in_use
+	Mountable bool   `json:"mountable"`
+	DurableID string `json:"durable_id,omitempty"` // uuid:<fs-uuid> — the attach key ("" when empty)
+	MountedAt string `json:"mounted_at,omitempty"` // in_use: Felhom's own mount
+	// Formattable is false when ANY node of the disk is mounted (another part in use): the root mkfs wrapper refuses
+	// every mount outside /mnt/felhom-drives on the disk and deliberately does not trust a sibling that claims to be
+	// Felhom's own (a compromised agent could fake that). Such a part can still be USED (attach erases nothing).
+	Formattable bool `json:"formattable"`
+}
+
+const (
+	PartEmpty      = "empty"
+	PartFilesystem = "filesystem"
+	PartInUse      = "in_use"
+)
+
+// candidatePartitions is the pure lister: the whole disk's lsblk tree → its partitions. sizeOf/numberOf read /sys,
+// uuidOf the by-uuid links, owned is the set of mountpoints corroborated as Felhom's own (felhomOwnedMounts).
+func candidatePartitions(lsblkJSON []byte, wholeName string, sizeOf func(string) int64, numberOf func(string) int,
+	uuidOf func(string) (string, bool), owned map[string]bool) []CandidatePartition {
+	var doc struct {
+		BlockDevices []lsblkDev `json:"blockdevices"`
+	}
+	if json.Unmarshal(lsblkJSON, &doc) != nil {
+		return nil
+	}
+	var out []CandidatePartition
+	for _, d := range doc.BlockDevices {
+		if d.Name != wholeName {
+			continue
+		}
+		diskMounted := d.MountPoint != ""
+		var walk func(n lsblkDev)
+		walk = func(n lsblkDev) {
+			if n.MountPoint != "" {
+				diskMounted = true
+			}
+			for _, c := range n.Children {
+				walk(c)
+			}
+		}
+		for _, c := range d.Children {
+			walk(c)
+		}
+		for _, p := range d.Children {
+			if len(p.Children) > 0 || memberFSTypes[p.FSType] {
+				continue // LVM/mapper on it, or a member: not a place (the disk filter refuses these anyway)
+			}
+			c := CandidatePartition{Device: "/dev/" + p.Name, Number: numberOf(p.Name), SizeBytes: sizeOf(p.Name), FSType: p.FSType}
+			switch {
+			case p.MountPoint != "" && (underFelhomDrives(p.MountPoint) || owned[p.MountPoint]):
+				c.State, c.MountedAt = PartInUse, p.MountPoint
+			case p.MountPoint != "":
+				continue // mounted by something else: never offered (the disk filter refuses the disk anyway)
+			case p.FSType == "":
+				c.State = PartEmpty
+			default:
+				c.State = PartFilesystem
+				c.Mountable = mountableFSTypes[p.FSType]
+				if u, ok := uuidOf(c.Device); ok && u != "" {
+					c.DurableID = "uuid:" + u
+				}
+			}
+			c.Formattable = c.State != PartInUse && !diskMounted
+			out = append(out, c)
+		}
+	}
+	return out
+}
+
+// readSysPartNumber reads /sys/class/block/<part>/partition (0 when unreadable).
+func readSysPartNumber(name string) int {
+	b, err := os.ReadFile(filepath.Join("/sys/class/block", name, "partition"))
+	if err != nil {
+		return 0
+	}
+	n, _ := strconv.Atoi(strings.TrimSpace(string(b)))
+	return n
+}
+
+// readSysPartSize reads a partition's size in bytes from /sys/class/block/<part>/size (512-byte sectors).
+func readSysPartSize(name string) int64 {
+	b, err := os.ReadFile(filepath.Join("/sys/class/block", name, "size"))
+	if err != nil {
+		return 0
+	}
+	s, err := strconv.ParseInt(strings.TrimSpace(string(b)), 10, 64)
+	if err != nil {
+		return 0
+	}
+	return s * 512
+}

internal/storage/partitions_test.go

@@ -0,0 +1,91 @@
+package storage
+
+import (
+	"strings"
+	"testing"
+)
+
+// Partitions as places (felhom.eu architecture/16 §6, agent v0.158.0): every partition of a free disk with its own
+// state; Felhom's own drive shown but never offered; anything foreign, layered or a member never listed.
+
+const partTree = `{"blockdevices":[
+ {"name":"sdb","fstype":null,"mountpoint":null,"children":[
+   {"name":"sdb1","fstype":"ext4","mountpoint":"/mnt/felhom-drives/hdd_1"},
+   {"name":"sdb2","fstype":null,"mountpoint":null},
+   {"name":"sdb3","fstype":"ntfs","mountpoint":null},
+   {"name":"sdb4","fstype":"xfs","mountpoint":null},
+   {"name":"sdb5","fstype":"ext4","mountpoint":"/mnt/films"},
+   {"name":"sdb6","fstype":"ext4","mountpoint":"/srv/data"},
+   {"name":"sdb7","fstype":"LVM2_member","mountpoint":null,"children":[{"name":"vg-lv","fstype":"ext4","mountpoint":null}]},
+   {"name":"sdb8","fstype":"crypto_LUKS","mountpoint":null}
+ ]},
+ {"name":"sdc","fstype":null,"mountpoint":null,"children":[{"name":"sdc1","fstype":null,"mountpoint":null}]}
+]}`
+
+func listParts(owned map[string]bool) map[string]CandidatePartition {
+	got := candidatePartitions([]byte(partTree), "sdb",
+		func(n string) int64 { return int64(len(n)) << 30 },
+		func(n string) int { return int(n[len(n)-1] - '0') },
+		func(dev string) (string, bool) { return "uuid-of-" + strings.TrimPrefix(dev, "/dev/"), true },
+		owned)
+	m := map[string]CandidatePartition{}
+	for _, p := range got {
+		m[p.Device] = p
+	}
+	return m
+}
+
+func TestPartitions_StatesPerPartition(t *testing.T) {
+	m := listParts(map[string]bool{"/mnt/films": true})
+	want := map[string]string{"/dev/sdb1": PartInUse, "/dev/sdb2": PartEmpty, "/dev/sdb3": PartFilesystem,
+		"/dev/sdb4": PartFilesystem, "/dev/sdb5": PartInUse}
+	if len(m) != len(want) {
+		t.Fatalf("want exactly %d listed partitions, got %+v", len(want), m)
+	}
+	for dev, st := range want {
+		if m[dev].State != st {
+			t.Fatalf("%s: want %s, got %+v", dev, st, m[dev])
+		}
+	}
+	if m["/dev/sdb2"].DurableID != "" || m["/dev/sdb2"].Mountable {
+		t.Fatalf("an empty part has no attach key: %+v", m["/dev/sdb2"])
+	}
+	if !m["/dev/sdb4"].Mountable || m["/dev/sdb4"].DurableID != "uuid:uuid-of-sdb4" || m["/dev/sdb4"].Number != 4 {
+		t.Fatalf("an xfs part is usable by its fs-UUID: %+v", m["/dev/sdb4"])
+	}
+	if m["/dev/sdb3"].Mountable {
+		t.Fatalf("ntfs is a filesystem the agent does not mount as-is: %+v", m["/dev/sdb3"])
+	}
+}
+
+// A raw /mnt/<name> mount counts as Felhom's own ONLY when corroborated (felhomOwnedMounts); else it is foreign and
+// the part is not listed. Positive control: TestPartitions_StatesPerPartition (the same mount, corroborated).
+func TestPartitions_UncorroboratedMountIsNotOffered(t *testing.T) {
+	m := listParts(nil)
+	if _, ok := m["/dev/sdb5"]; ok {
+		t.Fatalf("an uncorroborated /mnt/films mount must not be listed, got %+v", m["/dev/sdb5"])
+	}
+	for _, dev := range []string{"/dev/sdb6", "/dev/sdb7", "/dev/sdb8", "/dev/sdc1"} {
+		if _, ok := m[dev]; ok {
+			t.Fatalf("%s must never be listed (foreign mount, LVM, LUKS, another disk)", dev)
+		}
+	}
+}
+
+// The root wrapper refuses any format on a disk with a mounted node (and trusts no „it is Felhom's own" claim), so a
+// part next to a mounted one is never offered for formatting — only for use. Positive control: an unmounted disk.
+func TestPartitions_FormattableOnlyWhenNothingOnTheDiskIsMounted(t *testing.T) {
+	m := listParts(map[string]bool{"/mnt/films": true})
+	for dev, p := range m {
+		if p.Formattable {
+			t.Fatalf("%s: a part of a disk with a mounted part must not be formattable: %+v", dev, p)
+		}
+	}
+	free := `{"blockdevices":[{"name":"sdc","fstype":null,"mountpoint":null,"children":[
+	  {"name":"sdc1","fstype":null,"mountpoint":null},{"name":"sdc2","fstype":"ext4","mountpoint":null}]}]}`
+	got := candidatePartitions([]byte(free), "sdc", func(string) int64 { return 1 << 30 }, func(string) int { return 1 },
+		func(string) (string, bool) { return "u", true }, nil)
+	if len(got) != 2 || !got[0].Formattable || !got[1].Formattable {
+		t.Fatalf("parts of an unmounted disk are formattable, got %+v", got)
+	}
+}

internal/storage/watchdog.go

@@ -96,11 +96,11 @@ type Watchdog struct {
 	now       func() time.Time
 	spawn     func(func()) // spawn a background task (overridable in tests; default `go f()`)
 
-	intent   IntentReader      // P3: gate re-mount by drive intent (nil = ungated legacy)
+	intent   IntentReader           // P3: gate re-mount by drive intent (nil = ungated legacy)
 	onAbsent func(durableID string) // P3: called when a known target goes ABSENT (clears `ejected`)
 
 	mu             sync.Mutex
-	last           map[string]bool      // name -> last observed present (only for seen targets)
+	last           map[string]bool // name -> last observed present (only for seen targets)
 	lastFire       time.Time
 	fired          bool                 // lastFire is valid
 	pending        bool                 // a transition is awaiting the debounce window

internal/thincheck/thincheck.go

@@ -0,0 +1,243 @@
+// Package thincheck finds an ORPHAN thin device: one the LVM thin pool holds that LVM itself no longer knows (R-1012).
+//
+// A full host root while vzdump takes its storage snapshot can leave such a device behind (demo-hp 2026-10-14: the
+// pool held devices 1, 2 and 4; LVM knew 1 and 2). From then on every snapshot backup of that guest fails at the
+// snapshot step („Creation of new snapshot 4 of device 2 failed") and nothing in the product said why. This package
+// NAMES the orphan, in plain words, in the backup's failure — it never removes one: the repair
+// (`dmsetup message <pool>-tpool 0 "delete <id>"`) stays a human step, decided by the operator (`03` whole-guest
+// backup; runbook `whole-guest-backup-orphan-thin.md`).
+//
+// The device list comes from the root helper `felhom-thin-check <vg> <pool>` (config bundle), which reads the pool
+// through a metadata snapshot and LVM's list before and after. An id is an orphan only when the pool holds it and
+// NEITHER LVM list names it, on two reads a few seconds apart — so a volume being created at that moment is never
+// reported.
+package thincheck
+
+import (
+	"context"
+	"fmt"
+	"log/slog"
+	"sort"
+	"strconv"
+	"strings"
+	"time"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// HelperPath is where the config bundle installs the root helper.
+const HelperPath = "/usr/local/sbin/felhom-thin-check"
+
+// Pool names one LVM thin pool.
+type Pool struct{ VG, Pool string }
+
+func (p Pool) String() string { return p.VG + "/" + p.Pool }
+
+// Reading is one helper run, parsed.
+type Reading struct {
+	PoolIDs   []int
+	LVMBefore map[int]string
+	LVMAfter  map[int]string
+}
+
+// Parse reads the helper's output. ok=false unless the closing "ok" line is there (a cut-off run is no reading).
+func Parse(out string) (Reading, bool) {
+	r := Reading{LVMBefore: map[int]string{}, LVMAfter: map[int]string{}}
+	done := false
+	for _, line := range strings.Split(out, "\n") {
+		f := strings.Fields(line)
+		switch {
+		case len(f) == 1 && f[0] == "ok":
+			done = true
+		case len(f) == 2 && f[0] == "pool":
+			if id, err := strconv.Atoi(f[1]); err == nil {
+				r.PoolIDs = append(r.PoolIDs, id)
+			}
+		case len(f) == 3 && (f[0] == "lvm-before" || f[0] == "lvm-after"):
+			id, err := strconv.Atoi(f[1])
+			if err != nil {
+				continue
+			}
+			if f[0] == "lvm-before" {
+				r.LVMBefore[id] = f[2]
+			} else {
+				r.LVMAfter[id] = f[2]
+			}
+		}
+	}
+	return r, done
+}
+
+// Orphans returns the ids the pool holds that neither LVM list names, sorted.
+func (r Reading) Orphans() []int {
+	var out []int
+	for _, id := range r.PoolIDs {
+		_, b := r.LVMBefore[id]
+		_, a := r.LVMAfter[id]
+		if !b && !a {
+			out = append(out, id)
+		}
+	}
+	sort.Ints(out)
+	return out
+}
+
+// Runner is the privileged exec surface (proxmox.ExecRunner in sudo mode).
+type Runner interface {
+	Run(ctx context.Context, name string, args ...string) ([]byte, []byte, error)
+}
+
+// StorageLister is the read the checker needs to map a guest's volumes to their thin pools (*proxmox.Client).
+type StorageLister interface {
+	GuestConfig(ctx context.Context, vmid int) (proxmox.GuestConfig, error)
+	ListStorage(ctx context.Context) ([]proxmox.Storage, error)
+}
+
+// Checker runs the check for one guest's pools.
+type Checker struct {
+	API    StorageLister
+	Runner Runner
+	Helper string        // "" → HelperPath
+	Pause  time.Duration // between the two reads when the first finds an orphan; 0 → 3 s
+	Logger *slog.Logger
+}
+
+// Finding is one pool with orphans.
+type Finding struct {
+	Pool    Pool
+	Orphans []int
+}
+
+// PoolsOf maps the guest's volume-backed disks (rootfs, mpN) to the lvmthin pools they live on.
+func (c *Checker) PoolsOf(ctx context.Context, vmid int) ([]Pool, error) {
+	cfg, err := c.API.GuestConfig(ctx, vmid)
+	if err != nil {
+		return nil, err
+	}
+	stores, err := c.API.ListStorage(ctx)
+	if err != nil {
+		return nil, err
+	}
+	thin := map[string]Pool{}
+	for _, s := range stores {
+		if s.Type == "lvmthin" && s.VGName != "" && s.ThinPool != "" {
+			thin[s.Storage] = Pool{VG: s.VGName, Pool: s.ThinPool}
+		}
+	}
+	vols := []string{cfg.RootFS}
+	mps := cfg.MountPoints()
+	keys := make([]string, 0, len(mps))
+	for k := range mps {
+		keys = append(keys, k)
+	}
+	sort.Strings(keys)
+	for _, k := range keys {
+		vols = append(vols, mps[k])
+	}
+	seen := map[Pool]bool{}
+	var out []Pool
+	for _, v := range vols {
+		store, _, ok := strings.Cut(strings.Split(v, ",")[0], ":")
+		if !ok || strings.HasPrefix(store, "/") {
+			continue // a bind mount (a path) is no volume
+		}
+		if p, isThin := thin[store]; isThin && !seen[p] {
+			seen[p] = true
+			out = append(out, p)
+		}
+	}
+	return out, nil
+}
+
+// read runs the helper once.
+func (c *Checker) read(ctx context.Context, p Pool) (Reading, error) {
+	helper := c.Helper
+	if helper == "" {
+		helper = HelperPath
+	}
+	out, stderr, err := c.Runner.Run(ctx, helper, p.VG, p.Pool)
+	if err != nil {
+		return Reading{}, fmt.Errorf("felhom-thin-check %s: %w: %s", p, err, strings.TrimSpace(string(stderr)))
+	}
+	r, ok := Parse(string(out))
+	if !ok {
+		return Reading{}, fmt.Errorf("felhom-thin-check %s: the output has no closing ok line", p)
+	}
+	return r, nil
+}
+
+// Check returns the pools of vmid that hold an orphan (two reads, intersected). A pool that cannot be read is logged
+// and left out: the check FAILS OPEN — it explains a failure, it never stops a backup.
+func (c *Checker) Check(ctx context.Context, vmid int) []Finding {
+	logger := c.Logger
+	if logger == nil {
+		logger = slog.Default()
+	}
+	pools, err := c.PoolsOf(ctx, vmid)
+	if err != nil {
+		logger.Warn("thincheck: cannot map the guest's disks to thin pools — not checked", "vmid", vmid, "err", err)
+		return nil
+	}
+	pause := c.Pause
+	if pause <= 0 {
+		pause = 3 * time.Second
+	}
+	var found []Finding
+	for _, p := range pools {
+		first, err := c.read(ctx, p)
+		if err != nil {
+			logger.Warn("thincheck: thin pool not read — not checked", "vmid", vmid, "pool", p.String(), "err", err)
+			continue
+		}
+		o1 := first.Orphans()
+		if len(o1) == 0 {
+			logger.Debug("thincheck: thin pool and LVM agree", "vmid", vmid, "pool", p.String(), "devices", len(first.PoolIDs))
+			continue
+		}
+		select {
+		case <-ctx.Done():
+			return found
+		case <-time.After(pause):
+		}
+		second, err := c.read(ctx, p)
+		if err != nil {
+			logger.Warn("thincheck: second read failed — not reported", "vmid", vmid, "pool", p.String(), "err", err)
+			continue
+		}
+		in2 := map[int]bool{}
+		for _, id := range second.Orphans() {
+			in2[id] = true
+		}
+		var both []int
+		for _, id := range o1 {
+			if in2[id] {
+				both = append(both, id)
+			}
+		}
+		if len(both) > 0 {
+			found = append(found, Finding{Pool: p, Orphans: both})
+			logger.Warn("thincheck: ORPHAN thin device(s) in the pool — LVM does not know them; NOT removed (a human step)",
+				"vmid", vmid, "pool", p.String(), "orphans", both)
+		}
+	}
+	return found
+}
+
+// Describe is the operator sentence for a backup failure. Plain words; names the pool and the ids; says nothing was
+// removed and where the repair is written down.
+func Describe(f []Finding) string {
+	if len(f) == 0 {
+		return ""
+	}
+	var parts []string
+	for _, x := range f {
+		ids := make([]string, len(x.Orphans))
+		for i, id := range x.Orphans {
+			ids[i] = strconv.Itoa(id)
+		}
+		parts = append(parts, fmt.Sprintf("thin pool %s holds device %s that LVM does not know", x.Pool, strings.Join(ids, ", ")))
+	}
+	return "probable cause: " + strings.Join(parts, "; ") +
+		" (an orphan snapshot, usually left when the host disk filled during a backup). Snapshot backups on this pool " +
+		"fail until an operator removes it; Felhom did not remove anything. Repair: runbook whole-guest-backup-orphan-thin.md"
+}

internal/thincheck/thincheck_test.go

@@ -0,0 +1,184 @@
+package thincheck
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"io"
+	"log/slog"
+	"os/exec"
+	"path/filepath"
+	"reflect"
+	"runtime"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// The helper's output on demo-hp 2026-10-14, before the repair: the pool held 1, 2 and 4; LVM knew 1 and 2
+// (audits/power-cut-fixes-2026-10-14/E/thin-ls.txt).
+const orphanOut = `lvm-before 1 vm-9201-disk-0
+lvm-before 2 vm-9201-disk-1
+pool 1
+pool 2
+pool 4
+lvm-after 1 vm-9201-disk-0
+lvm-after 2 vm-9201-disk-1
+ok
+`
+
+const healthyOut = `lvm-before 1 vm-9201-disk-0
+lvm-before 2 vm-9201-disk-1
+pool 1
+pool 2
+lvm-after 1 vm-9201-disk-0
+lvm-after 2 vm-9201-disk-1
+ok
+`
+
+func quiet() *slog.Logger { return slog.New(slog.NewTextHandler(io.Discard, nil)) }
+
+func TestParse_TheOrphanOf20261014(t *testing.T) {
+	r, ok := Parse(orphanOut)
+	if !ok {
+		t.Fatal("a complete reading was not accepted")
+	}
+	if got := r.Orphans(); !reflect.DeepEqual(got, []int{4}) {
+		t.Fatalf("orphans = %v, want [4]", got)
+	}
+	h, _ := Parse(healthyOut)
+	if got := h.Orphans(); len(got) != 0 {
+		t.Fatalf("a healthy pool reported orphans %v", got)
+	}
+}
+
+// A volume LVM created WHILE the pool was read appears only in the after-list: never an orphan.
+func TestOrphans_AVolumeCreatedDuringTheReadIsNotAnOrphan(t *testing.T) {
+	r, _ := Parse(strings.Replace(orphanOut, "lvm-after 2 vm-9201-disk-1\n", "lvm-after 2 vm-9201-disk-1\nlvm-after 4 snap_vm-9201-disk-1_vzdump\n", 1))
+	if got := r.Orphans(); len(got) != 0 {
+		t.Fatalf("orphans = %v, want none", got)
+	}
+}
+
+// A run cut off before its closing "ok" (the helper died, the output is partial) is no reading at all.
+func TestParse_NoClosingLineIsNoReading(t *testing.T) {
+	if _, ok := Parse(strings.TrimSuffix(orphanOut, "ok\n")); ok {
+		t.Fatal("a cut-off run was accepted as a reading")
+	}
+}
+
+type fakeAPI struct {
+	cfg    proxmox.GuestConfig
+	stores []proxmox.Storage
+}
+
+func (f *fakeAPI) GuestConfig(context.Context, int) (proxmox.GuestConfig, error) { return f.cfg, nil }
+func (f *fakeAPI) ListStorage(context.Context) ([]proxmox.Storage, error)        { return f.stores, nil }
+
+func guest9201() *fakeAPI {
+	extra := map[string]json.RawMessage{
+		"mp0": json.RawMessage(`"local-lvm:vm-9201-disk-1,mp=/var/lib/felhom,backup=1,size=20G"`),
+		"mp8": json.RawMessage(`"/mnt/felhom-drives,mp=/mnt/felhom-drives"`),
+		"mp9": json.RawMessage(`"local:9201/vm-9201-disk-2.raw,mp=/x"`),
+	}
+	return &fakeAPI{
+		cfg: proxmox.GuestConfig{RootFS: "local-lvm:vm-9201-disk-0,size=8G", Extra: extra},
+		stores: []proxmox.Storage{
+			{Storage: "local", Type: "dir", Path: "/var/lib/vz"},
+			{Storage: "local-lvm", Type: "lvmthin", VGName: "pve", ThinPool: "data"},
+			{Storage: "felhom-pbs", Type: "pbs"},
+		},
+	}
+}
+
+type seqRunner struct {
+	outs  []string
+	errs  []error
+	calls [][]string
+}
+
+func (r *seqRunner) Run(_ context.Context, name string, args ...string) ([]byte, []byte, error) {
+	r.calls = append(r.calls, append([]string{name}, args...))
+	i := len(r.calls) - 1
+	var err error
+	if i < len(r.errs) {
+		err = r.errs[i]
+	}
+	if i < len(r.outs) {
+		return []byte(r.outs[i]), nil, err
+	}
+	return nil, nil, err
+}
+
+func TestPoolsOf_RootfsAndVolumeMountpointsOnly(t *testing.T) {
+	c := &Checker{API: guest9201(), Runner: &seqRunner{}, Logger: quiet()}
+	pools, err := c.PoolsOf(context.Background(), 9201)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if !reflect.DeepEqual(pools, []Pool{{VG: "pve", Pool: "data"}}) {
+		t.Fatalf("pools = %v, want [pve/data] once (bind mounts and dir volumes are not thin)", pools)
+	}
+}
+
+// The consequence: an orphan seen on BOTH reads becomes a finding whose sentence names the pool and the id, and says
+// nothing was removed. The only command ever run is the read-only helper with <vg> <pool>.
+// RED-PROOF (P1/redproof-thincheck.txt): with Orphans() returning nil this test fails.
+func TestCheck_OrphanOnBothReads_IsNamed(t *testing.T) {
+	rr := &seqRunner{outs: []string{orphanOut, orphanOut}}
+	c := &Checker{API: guest9201(), Runner: rr, Pause: 1, Logger: quiet()}
+	f := c.Check(context.Background(), 9201)
+	if len(f) != 1 || f[0].Pool.String() != "pve/data" || !reflect.DeepEqual(f[0].Orphans, []int{4}) {
+		t.Fatalf("findings = %+v", f)
+	}
+	want := [][]string{{HelperPath, "pve", "data"}, {HelperPath, "pve", "data"}}
+	if !reflect.DeepEqual(rr.calls, want) {
+		t.Fatalf("calls = %v, want only the read-only helper twice", rr.calls)
+	}
+	d := Describe(f)
+	for _, s := range []string{"thin pool pve/data holds device 4 that LVM does not know", "Felhom did not remove anything",
+		"whole-guest-backup-orphan-thin.md"} {
+		if !strings.Contains(d, s) {
+			t.Fatalf("Describe = %q, missing %q", d, s)
+		}
+	}
+}
+
+// An id that is gone on the second read (a volume mid-creation) is not reported; a healthy pool is read ONCE.
+func TestCheck_TransientOrHealthy_NothingReported(t *testing.T) {
+	rr := &seqRunner{outs: []string{orphanOut, healthyOut}}
+	if f := (&Checker{API: guest9201(), Runner: rr, Pause: 1, Logger: quiet()}).Check(context.Background(), 9201); len(f) != 0 {
+		t.Fatalf("a transient id was reported: %+v", f)
+	}
+	rr = &seqRunner{outs: []string{healthyOut}}
+	if f := (&Checker{API: guest9201(), Runner: rr, Logger: quiet()}).Check(context.Background(), 9201); len(f) != 0 || len(rr.calls) != 1 {
+		t.Fatalf("healthy: findings %+v, calls %v", f, rr.calls)
+	}
+	if Describe(nil) != "" {
+		t.Fatal("no finding must describe as empty")
+	}
+}
+
+// The helper refusing (a metadata snapshot held by someone else, exit 3) or failing: no finding, no panic — the check
+// fails open; it explains a failure, it never stops a backup.
+func TestCheck_HelperRefuses_FailsOpen(t *testing.T) {
+	rr := &seqRunner{errs: []error{errors.New("exit status 3")}}
+	if f := (&Checker{API: guest9201(), Runner: rr, Logger: quiet()}).Check(context.Background(), 9201); len(f) != 0 {
+		t.Fatalf("findings = %+v", f)
+	}
+}
+
+// The root helper's own tests (configs/test_felhom_thin_check.py) run with the Go suite.
+func TestRootHelperPythonSuite(t *testing.T) {
+	py, err := exec.LookPath("python3")
+	if err != nil {
+		t.Skip("python3 not found")
+	}
+	_, file, _, _ := runtime.Caller(0)
+	script := filepath.Join(filepath.Dir(file), "..", "..", "configs", "test_felhom_thin_check.py")
+	out, err := exec.Command(py, script).CombinedOutput()
+	if err != nil {
+		t.Fatalf("helper suite failed: %v\n%s", err, out)
+	}
+}