night 2026-10-22: REPORT; CHANGELOG corrected — a push of the installer to main publishes nothing (R-110, tag installer-v1.35.0); D4 stays a patch because both halves need a release

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.

.gitea/workflows/gates.yml

@@ -64,6 +64,9 @@ 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)
@@ -108,8 +111,16 @@ jobs:
 
           repo = os.environ.get("GITHUB_REPOSITORY", "?")
           sha  = os.environ.get("GITHUB_SHA", "?")
-          run  = os.environ.get("GITHUB_RUN_NUMBER", "?")
+          # R-1135: the link is for a PERSON, so it is the public address and the run ID.
-          srv  = os.environ.get("GITHUB_SERVER_URL", "https://gitea.dooplex.hu")
+          # 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>",

.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 +1,266 @@
+## Unreleased (2026-10-22 night) — the tunnel's future-stamped handshake is reported (R-1204) — COMMITTED, NOT RELEASED (no agent release tonight); root SSH by key only (R-982, D4) is a READY PATCH, not on main
+
+- **R-982 / D4 — a ready patch, NOT on main.** Built and proven the same night (felhom.eu
+  `audits/night-2026-10-22/P2-D4/`): a bundle drop-in `/etc/ssh/sshd_config.d/10-felhom-root-key-only.conf`
+  (`PermitRootLogin prohibit-password`), a `sshd` content check (exactly `prohibit-password`; `sshd -t -f`), and after
+  an install that wrote it `sshd -t` + `systemctl reload ssh`; tests `RootKeyOnly` (3), red-proved. Live on a fresh
+  nested box from ISO 1.30.0: root password over SSH refused, root key works, the console and the web page still take
+  the password. It was committed (`a0c88e5`) and moved back out (`7651461`): the installer's uninstall must remove every bundle file (felhom.eu `scripts/test_hostinstall.py`), the bundle half reaches a box only with an agent release, and the installer half only with a new `installer-v…` tag (R-110 — a push to `main` alone publishes nothing) — both forbidden tonight. *(Corrected the same night: the first reason written here, „a push of `felhom-host-install.sh` publishes the installer", was wrong — the website serves `/scripts/` from the tag `installer-v1.35.0`, read live in `manifests/webpage.yaml` and the running Deployment.)* Both halves are patches: `felhom-agent-bundle-D4.patch` and `felhom.eu-installer-D4.patch` in that folder; apply them together with the agent release and the next installer tag.
+- **R-1204 — `wireguard.handshake_future_s`.** A box that ran once with its clock ahead keeps a handshake stamped in the
+  future; ep0 refuses every newer one, and `last_handshake_age_s` (clamped at 0) read „just now" for ever. The heartbeat
+  now carries how many seconds the handshake lies ahead (> 60 s only); the age keeps its clamp. The hub alarms on it
+  (felhom.eu hub `wg_tunnel_stale`). Test `TestR1204_AFutureHandshakeIsReported`, red-proved.
+
+## v0.162.0 — every guest may swap as much as its RAM (R-1192, the 8 GB freeze); the host's swap is reported and checked; no false „operator access degraded" on a fresh box (R-1194) (2026-10-22) — RELEASED on the operator's yes; the vouch is the operator's
+
+Released 2026-10-22 07:40 UTC (operator yes, the morning answers) with `scripts/release-agent.sh` from a clean clone at
+`b310773`: binary sha256 `5fd7a23bfe319cb521c4bc399612d5fb3413f1e695f79bc663c386f69c757928` (the same bytes as the
+night's candidate), config bundle sha256 `ac772967fd4669045e837a70bf5acca239c58d55f7f00212e472a8facd60b5a5`, tag
+`v0.162.0`; no `configs/` change since v0.161.0. Delivered by one signed `agent_update` job per box to demo-hp, Tester 1
+and demo-felhom (07:49–07:58 UTC): each `signed op COMPLETED`, `capabilities self-check ok=76 total=76`, `update
+committed`, then `guest-swap: raised` for its 9201 (512 MB → its RAM). Evidence felhom.eu
+`documentation/audits/morning-2026-10-22/release/04-agent-delivery.txt`. **Found at once:** demo-hp (host swap 8191 MB,
+guest swap 25898 MB) and demo-felhom (8191 MB vs 12288 MB) logged the host-swap WARN and the hub mailed
+`host_swap_short` — the check compares the host's swap with the guests' whole swap allowance, which a guest larger
+than the installer's 8 GB host swap always exceeds (felhom.eu register R-1216). Soak (night of 2026-10-21, 8 GB box):
+11 h, no freeze, the host's swap peaked at 2.7 GB used.
+
+
+The freeze of the rehearsal's first afternoon, reproduced and measured on demo-hp (felhom.eu
+`audits/fixes-2026-10-21/P4-freeze/`): Proxmox gives an LXC a 512 MB swap cap when nothing sets `swap`, and the agent
+never set it. On the 8 GB machine (guest 4096 MB) the first uploads made Immich load its analysis models; the guest hit
+its limit with the swap full and could evict only program code — 150 000–600 000 refaults and ~1 GB of disk reads per
+10 s, memory „full" pressure 92–95 %, load 155, the dashboard dead for 6½ minutes, no app killed. Raising the swap cap
+to 4096 MB live ended it within a minute while the uploads ran on.
+
+- `reconcile.buildBringUpConfig`: a new guest gets `swap` = its memory.
+- `POST /guest/memory` (RAM resize): `swap` moves with `memory` (`localapi.GuestSwapMB`).
+- At agent start (one minute in): `localapi.EnsureGuestSwap` raises every LXC in the `felhom` pool whose swap is below
+  its RAM to it — a live cgroup change, no restart; never lowers a higher swap; one guest's failure is logged and the
+  others are still done.
+- Tests `TestR1192_EnsureGuestSwapRaisesEveryLowSwapToItsRAM`, `TestR1192_BringUpSetsSwapToTheRAM`, the resize test's
+  new swap assertion — each red-proved (the write removed → `raised 0, want 2`; `swap=""` on both other paths).
+- Privilege: `swap` is under `VM.Config.Memory`, which the RAM resize already uses; no role, sudoers or bundle change.
+
+Added the night of 2026-10-21 (still unreleased, the same release):
+
+- **The host's swap is reported and checked (R-1192 follow-up).** The guest fix moves a memory burst onto the HOST's
+  swap; a host with none gets nothing from it. The host report carries `host.host_swap_mb` (Proxmox node status — the
+  kernel's SwapTotal, the same number as `/proc/meminfo`) and `host.guest_swap_mb` (the sum of the guests' `swap`; each
+  guest spec also carries `swap_bytes`). When the first is below the second the agent logs one WARN (and one INFO when
+  it is covered again); the hub raises `host_swap_short`. `GET /guest/memory` carries `host_swap_mb` and
+  `guest_swap_mb` for the controller's install page. The agent never creates host swap (a decision — felhom.eu
+  register). Measured the same night: our installer gives an 8 GB / 256 GB machine 7.75 GB of host swap.
+- **R-1194 — no false „operator access degraded" on a fresh box.** The first host report was sent before the
+  felhom-sshd loop's first pass, read a config that was never rendered („config invalid, port 0") and the hub mailed
+  `oob_degraded` — reproduced on a fresh install 2026-10-21 (the report at 20:34:00, the apply 0.4 s later, the daemon
+  healthy; the check runs `sshd -t` through sudo, so the non-root theory was false). The loop now reports no OOB stanza
+  until its first pass has run (a failed pass counts, so a real lasting failure still reaches the hub).
+- Tests `TestR1192_HostReportCarriesSwapAndWarnsOnce`, `TestR1192_GuestMemoryCarriesHostAndGuestSwap`,
+  `TestR1194_NoOOBStanzaBeforeTheFirstPass` — each red-proved.
+
+## v0.161.0 — three fixes (R-921, R-1070, R-1010); two texts made true after 0.160.0 (2026-10-20) — RELEASED on the operator's yes, NOT VOUCHED
+
+Released 2026-10-20 (operator yes, the morning answers) with `scripts/release-agent.sh` from a clean clone at `5152123`:
+binary sha256 `ff4e6dc52dd70234913445ecdeb88d31dea8e394d77c48704d26ff19b999d3b5`, config bundle sha256
+`485f36c957d83f772aa22cac24e5e73f44da5cf8b957b1fd513bec2bc1d12c47`, tag `v0.161.0`. Delivered by one signed
+`agent_update` job per box to demo-hp (06:38 UTC), Tester 1 (06:54) and demo-felhom (06:55): each `signed op COMPLETED`,
+`capabilities self-check ok=76 total=76`, `update committed`. **The vouch is the operator's.** Live on demo-hp: `host_busy`
+answers `{"busy": true, "what": "backup:local"}` during a „Mentés most" whole-guest backup and `{"busy": false}` before
+and after (read from inside 9201 with its own token; felhom.eu `audits/night2-2026-10-20/P1/`). R-1083 (a key for an
+existing guest) is NOT in this release: see the R-1083 row. 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.

CONTEXT.md

@@ -1,5 +1,33 @@
 # CONTEXT — felhom-agent working state
 
+> **2026-10-20 — v0.161.0 RELEASED (operator yes), NOT VOUCHED; on demo-hp, Tester 1, demo-felhom** (signed jobs, all
+> committed, 76/76). R-921 `host_busy` (proven live on demo-hp during a press), R-1070 the tunnel `absent` state (9202's
+> real Docker 29 answer is the fixture row), R-1010 the janitor's one `force=1`. Vouch: version 0.161.0, sha256
+> `ff4e6dc5…99d3b5`, bundle `485f36c9…12c47`. R-1083 left out on purpose: an agent key for 9202 would make its
+> controller book whole-guest backups through the agent, one tier writing to ep0 — the R-1083 row says what is needed.
+> **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),

REPORT.md

@@ -1,12 +1,19 @@
-# REPORT — the agent on Go 1.27.2 (R-972, 2026-10-16) — NOT released
+# REPORT — night 2026-10-22: one change on `main`, one ready patch, nothing released
 
-Written in ASD-STE100 Simplified Technical English. Part of burndown 2 (felhom.eu `REPORT-burndown2-2026-10-16.md`).
+**What:** the night brief's agent parts. No agent release, no vouch (both forbidden tonight). Design read:
+`03-host-agent.md` (the WireGuard tunnel, the config bundle), `06-offsite-connectivity.md`.
 
-- **What changed:** `go.mod` `go 1.27.2` and golang.org/x/crypto v0.58.0. No code change.
+- **R-1204, `a0c88e5` (on `main`, unreleased).** The heartbeat's `wireguard` block gains `handshake_future_s` when the
-- **Proof:** `go build ./...`, `go vet ./...`, `go test ./...` rc 0 on Go 1.27.2; `agent_gates.py --fast` all OK.
+  last handshake lies more than 60 s in the future (`wgtunnel.handshakeFuture`) — the clock-ahead lock-out that the
-- **Not done, on purpose:** no release and no deploy — agent releases wait for the operator's vouch. R-972 stays
+  clamped `last_handshake_age_s` hid. Red-proved (`TestR1204_AFutureHandshakeIsReported`). The hub's alarm on it is
-  NARROWED for the agent: the next vouched release carries the new Go.
+  hub 0.166.0 (felhom.eu, built, not deployed).
-
+- **R-982 / D4 (operator 15:47: root SSH from the LAN by key only) — a ready patch, NOT on main (`7651461` moved it
-## Observations
+  out).** A bundle drop-in `/etc/ssh/sshd_config.d/10-felhom-root-key-only.conf` (`PermitRootLogin prohibit-password`),
-
+  checked and reloaded by `felhom-os-apply`; proven on a fresh install from ISO 1.30.0 (nested VM 9343, destroyed):
-1. The release script needs no change: it builds with the toolchain `go.mod` names. NOT-A-FINDING: GOTOOLCHAIN=auto on DooPlex fetched go1.27.2 for this build and will for a release build.
+  password SSH refused, key works, the console and the PVE web login keep the password. Patches:
+  felhom.eu `audits/night-2026-10-22/P2-D4/felhom-agent-bundle-D4.patch` + `felhom.eu-installer-D4.patch`. They ship
+  with the next agent release (the operator's vouch) and the next installer tag.
+- **Correction:** the CHANGELOG first said a push of the installer would publish it. Wrong: `/scripts/` is served from
+  the tag `installer-v1.35.0` (R-110). The decision to keep D4 as a patch stands on the other two reasons.
+- **Gates:** `agent_gates.py --fast` green before each push. CI by commit: felhom.eu `REPORT-night-2026-10-22.md`.
+- **Teardown:** VM 9343 destroyed on demo-hp (own id); its hub appliance (id 49) waits for the operator's Discard.

REUSE.md

@@ -114,8 +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) |
+| `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); subprocess stdout is SECRET-BEARING → parsed then zeroed, never logged |
+| 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 |
@@ -174,6 +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

@@ -194,7 +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)")
@@ -404,6 +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",
@@ -830,8 +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
@@ -906,6 +936,22 @@ func runDaemon(cfg config.Config, logger *slog.Logger, logRing *applog.Ring) int
 	// network and stays under the hub's 45-minute host_stale (its
 	// alerting.stale_threshold; host_down at 90). On a box without the kernel lane (an older wrapper, a BYO host) the check is refused and logged.
 	go osLeg.KernelAfterBoot(ctx, 0, osupdate.KernelJudge{Wait: osupdate.DefaultKernelJudgeWait})
+	// R-1192: every guest may swap as much as its RAM — raised once at start (a live change, no restart). A minute
+	// after start so the guest has settled; a failure is logged and the next start retries.
+	go func() {
+		select {
+		case <-ctx.Done():
+			return
+		case <-time.After(time.Minute):
+		}
+		sctx, cancel := context.WithTimeout(ctx, 2*time.Minute)
+		defer cancel()
+		if n, err := localapi.EnsureGuestSwap(sctx, px, reconcile.DefaultPool, logger); err != nil {
+			logger.Warn("guest-swap: check skipped", "err", err)
+		} else {
+			logger.Info("guest-swap: checked", "raised", n)
+		}
+	}()
 
 	// Reconcile (slice 4) runs alongside the hub loop, sharing the per-guest queue
 	// (doc 03 §10). At slice 4 the desired-state provider is empty (no hub serving
@@ -1197,6 +1243,9 @@ 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()
@@ -1206,6 +1255,7 @@ 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)
@@ -1913,10 +1963,7 @@ func buildLocalAPIServer(cfg config.Config, px *proxmox.Client, store *backup.St
 	var tierRunners []*backup.BackupRunner
 	var tierTargets []string
 	for _, t := range backupTiers {
-		prune := ""
+		prune := t.PruneSpec() // primary: keep-daily=N (R-1149); an extra tier: keep-last=N; 0 → no prune
-		if t.KeepLast > 0 {
-			prune = fmt.Sprintf("keep-last=%d", t.KeepLast)
-		}
 		// 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
@@ -1934,7 +1981,7 @@ 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
@@ -2826,6 +2873,20 @@ 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
@@ -2976,16 +3037,36 @@ 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).
@@ -3049,10 +3130,7 @@ func runSelftestEscrowCreate(ctx context.Context, cfg config.Config, logger *slo
 		return 2
 	}
 
-	out, cerr := escrowCeremony(ctx, cfg, logger, escrowCeremonyOpts{
+	out, cerr := escrowCeremony(ctx, cfg, logger, escrowOptsFromCLI(storage, paperkey, offline, upload, identityBundlePath, jsonMode))
-		storage: storage, paperkey: paperkey, offline: offline, upload: upload,
-		identityBundlePath: identityBundlePath,
-	})
 	if cerr != nil {
 		switch cerr.kind {
 		case "usage":
@@ -3090,6 +3168,9 @@ 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 {
@@ -3243,13 +3324,10 @@ type escrowUploadRequest struct {
 	ResticPwSHA256 string `json:"restic_pw_sha256,omitempty"`
 }
 
-// uploadEscrowBlob PUTs the opaque blob (and, for 10D, the identity blob) to
+// escrowUploadBody is the hub upload body for one ceremony's result: the opaque blob (and, for 10D, the
-// the hub, authed with the per-host key. The hub stores ciphertext + non-secret fields; no usable
+// identity blob) plus non-secret fields. Ciphertext only — useless without R; no usable secret leaves
-// secret leaves the agent.
+// the agent. putEscrowBody sends it, authed with the per-host key.
-func uploadEscrowBlob(ctx context.Context, cfg config.Config, res escrow.CreateResult, resticPwSHA256 string) error {
+func escrowUploadBody(res escrow.CreateResult, resticPwSHA256 string) []byte {
-	if cfg.Hub.URL == "" || cfg.Hub.HostID == "" || cfg.Hub.APIKey == "" {
-		return fmt.Errorf("hub not configured (url/host_id/api_key)")
-	}
 	upReq := escrowUploadRequest{
 		BlobB64:        base64.StdEncoding.EncodeToString(res.Blob),
 		KeyFingerprint: res.KeyFingerprint,
@@ -3261,6 +3339,15 @@ 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 {

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"`
+	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" {
+	if spec := (config.BackupConfig{}).PruneBackupsSpec(); spec != "keep-daily=3" {
-		t.Fatalf("default PruneBackupsSpec = %q, want keep-last=3", spec)
+		t.Fatalf("default PruneBackupsSpec = %q, want keep-daily=3 (R-1149)", spec)
 	}
 }

internal/backup/r1012_thincheck_test.go

@@ -10,8 +10,9 @@ import (
 )
 
 // 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 before the run and once, FRESH, after the failure.
+// 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 (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{
@@ -29,15 +30,17 @@ func TestBackup_SnapshotFailure_NamesTheOrphan(t *testing.T) {
 			if !strings.Contains(rec.Error, why) {
 				t.Fatalf("the failure does not name the orphan: %q", rec.Error)
 			}
-			if calls != 2 {
+			if calls != 1 {
-				t.Fatalf("thin check ran %d times, want 2 (before the run and after the failure)", calls)
+				t.Fatalf("thin check ran %d times, want 1 (after the failure only)", calls)
 			}
 		})
 	}
 }
 
-// A successful backup keeps a clean record (the pre-check only logs); a stop-mode backup takes no snapshot, so the
+// A successful backup never runs the check (it takes a metadata snapshot of the live pool, and nobody needs the answer)
-// check never runs; an unwired runner behaves exactly as before.
+// 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}}}
@@ -45,8 +48,8 @@ func TestBackup_ThinCheck_SilentWhenNotApplicable(t *testing.T) {
 	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 != 1 {
+	if err != nil || !rec.Success || rec.Error != "" || calls != 0 {
-		t.Fatalf("success: err=%v rec=%+v calls=%d (want one pre-check, clean record)", err, rec, calls)
+		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())

internal/backup/runner.go

@@ -62,8 +62,9 @@ type BackupRunner struct {
 	// 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 before a snapshot-mode backup (a WARN in the log) and again when the vzdump fails, so the failure the
+	// runs ONLY when a snapshot-mode vzdump has failed (start or task), so the failure the controller mails to the
-	// controller mails to the operator says why. It never removes anything. nil = not wired.
+	// 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
@@ -117,8 +118,8 @@ func (r *BackupRunner) SetBeforeBackup(fn func(ctx context.Context, storage stri
 // 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. A FRESH check (not the one from
+// explainFailure appends the orphan finding to a failed snapshot-mode backup's error — the only place the check runs, so
-// before the run): the failure itself may be what made the orphan. Bounded, so a stuck helper cannot hold the job.
+// 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
@@ -221,11 +222,8 @@ func (r *BackupRunner) backup(ctx context.Context, vmid int, onSnapshot func())
 		return rec, fmt.Errorf("backup: %s", rec.Error)
 	}
 
-	if r.thinCheck != nil && r.mode == proxmox.ModeSnapshot {
+	// R-1012: no orphan check here. It takes a metadata snapshot of the live pool, and its answer is only needed when the
-		if why := r.thinCheck(ctx, vmid); why != "" {
+	// snapshot step has failed — explainFailure runs it then (pinned by TestBackup_ThinCheck_SilentWhenNotApplicable).
-			r.logger.Warn("backup: an orphan thin device may make this snapshot backup fail (R-1012)", "vmid", vmid, "finding", why)
-		}
-	}
 	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)

internal/capability/manifest.go

@@ -91,7 +91,7 @@ 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 around a whole-guest backup (read-only, R-1012)", "/usr/local/sbin/felhom-thin-check", []string{"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, ""},

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.
+// CeremonyOutputVersion is the current CeremonyOutput.Version value. 2 (R-978): json mode no longer
-const CeremonyOutputVersion = 1
+// 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/felhomsshd/loop.go

@@ -22,6 +22,11 @@ type Loop struct {
 	mu      sync.Mutex
 	fetched bool // a desired-state document has been delivered (false = never a teardown signal)
 	block   *hub.WireWireguard
+	// passed (R-1194) is set once the first reconcile pass has run. Until then the instance has never been rendered,
+	// so its health says nothing — and the first host report, sent before that pass, used to read „config invalid,
+	// port 0" and raise a false oob_degraded on EVERY fresh box (rehearsal 2026-10-20; reproduced 2026-10-21: the
+	// report at 20:34:00 read an absent config, the apply landed 0.4 s later).
+	passed bool
 
 	nudge chan struct{}
 }
@@ -85,6 +90,7 @@ func (l *Loop) reconcile(ctx context.Context) {
 	if !fetched {
 		applyBlock = nil
 	}
+	defer l.markPassed()
 	port, err := l.mgr.Apply(ctx, applyBlock)
 	if err != nil {
 		return // Apply logged; a claim/exhaustion or install error — retry next tick
@@ -106,8 +112,23 @@ func oobPeerIP(block *hub.WireWireguard) string {
 	return block.OOBPeerIP
 }
 
-// OOBStatus implements the hub collector's reporter seam (Part 4).
+// markPassed records that a reconcile pass has run (R-1194) — after a failed Apply too, so a real, lasting failure
+// is still reported from the next report on.
+func (l *Loop) markPassed() {
+	l.mu.Lock()
+	l.passed = true
+	l.mu.Unlock()
+}
+
+// OOBStatus implements the hub collector's reporter seam (Part 4). nil — no stanza, which the hub does not evaluate —
+// until the first reconcile pass has run (R-1194).
 func (l *Loop) OOBStatus(ctx context.Context) *hub.OOBStatus {
+	l.mu.Lock()
+	passed := l.passed
+	l.mu.Unlock()
+	if !passed {
+		return nil
+	}
 	_, block := l.snapshot()
 	return l.mgr.Status(ctx, block)
 }

internal/felhomsshd/r1194_first_report_test.go

@@ -0,0 +1,29 @@
+package felhomsshd
+
+import (
+	"context"
+	"testing"
+	"time"
+)
+
+// R-1194: on every fresh box the first host report — sent before the OOB loop's first pass — read „config invalid,
+// port 0" (the config was not rendered yet) and the hub mailed oob_degraded. Reproduced 2026-10-21 on a fresh box
+// (felhom.eu audits/night-2026-10-21/P5-checks/): the report at 20:34:00, the apply 0.4 s later, the daemon healthy.
+// The loop now reports NO stanza (the hub does not evaluate a missing one) until a pass has run — a failed pass
+// counts, so a real lasting failure is still reported from the next report on.
+// RED-PROOF: with OOBStatus not gated on `passed`, the first assertion fails (a stanza with config_invalid).
+func TestR1194_NoOOBStanzaBeforeTheFirstPass(t *testing.T) {
+	r := newScriptRunner()
+	m := NewManager(r, t.TempDir(), nil)
+	m.isActive = func(context.Context) bool { return false }
+	m.isFailed = func(context.Context) bool { return false }
+	m.now = func() time.Time { return time.Unix(1783270000, 0) }
+	l := NewLoop(m, nil, time.Minute, nil)
+	if st := l.OOBStatus(context.Background()); st != nil {
+		t.Fatalf("a stanza before the first pass reads a config that was never rendered: %+v", st)
+	}
+	l.reconcile(context.Background()) // fails or succeeds — either way a pass has run
+	if st := l.OOBStatus(context.Background()); st == nil {
+		t.Fatal("after the first pass the health must be reported (a real failure must reach the hub)")
+	}
+}

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") {
+		if e := strings.ToLower(stderr); strings.Contains(e, "no such object") || strings.Contains(e, "no such container") {
-			return TunnelNotRunning, "no cloudflared container in the guest"
+			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/collect.go

@@ -11,6 +11,7 @@ import (
 	"os"
 	"strings"
 	"sync"
+	"sync/atomic"
 	"time"
 
 	"gitea.dooplex.hu/admin/felhom-agent/internal/capability"
@@ -136,6 +137,8 @@ type Collector struct {
 	selfSHA          func() string // R-349: sha256 of the running binary; default runningBinarySHA256
 	logger           *slog.Logger
 	now              func() time.Time
+	// swapShortWarned (R-1192): the host-swap-short WARN is logged once per state change, not every report.
+	swapShortWarned atomic.Bool
 }
 
 // NewCollector builds a collector. hostID echoes config.Hub.HostID; agentVersion is
@@ -375,6 +378,8 @@ func (c *Collector) Collect(ctx context.Context) (*HostReport, error) {
 	}
 
 	host := hostMetrics(c.px.Node(), ns)
+	guests := c.collectGuests(ctx)
+	c.applySwap(&host, ns, guests)          // R-1192: host swap vs the guests' swap allowance
 	host.CPUTempC = c.cpuTempC(ctx)         // slice 9: operator freebie — temp now rides the hub report too
 	host.WrapperSHA256 = pbsWrapperSHA256() // R-50b(a): make privileged-artifact drift answerable
 	report := &HostReport{
@@ -383,7 +388,7 @@ func (c *Collector) Collect(ctx context.Context) (*HostReport, error) {
 		AgentVersion: c.agentVersion,
 		AgentSHA256:  c.agentSHA256(),
 		Host:         host,
-		Guests:       c.collectGuests(ctx),
+		Guests:       guests,
 		// storage_targets populated this slice (slice 5) via the observer; the rest stay
 		// defined-but-empty (slice 6). Non-nil so they marshal as [].
 		StorageTargets: c.collectStorage(ctx),
@@ -567,6 +572,7 @@ func (c *Collector) collectGuests(ctx context.Context) []Guest {
 				Cores:       cfg.Cores,
 				MemoryBytes: g.MaxMem,
 				DiskBytes:   g.MaxDisk,
+				SwapBytes:   cfg.Swap * (1 << 20),
 			}
 		}
 		guests = append(guests, entry)

internal/hub/hostswap.go

@@ -0,0 +1,39 @@
+package hub
+
+import (
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1192 follow-up (night 2026-10-21): the guest swap fix (every guest may swap as much as its RAM,
+// localapi/guestswap.go) moves a memory burst onto the HOST's swap. Measured on demo-hp the same day: our installer
+// gives an 8 GB / 256 GB machine 7.75 GB of host swap, so a 4096 MB guest's 4096 MB swap fits. A host with no swap
+// (a BYO install, or swap turned off) gets nothing from the fix. So the report carries both numbers, the agent says it
+// once in its log, and the hub raises host_swap_short. The agent never creates host swap itself: changing a host its
+// owner set up is a decision (felhom.eu register).
+
+// HostSwapShort reports whether the host's swap is below the guests' swap allowance. A host whose guests have no
+// swap allowance (0) is never short.
+func HostSwapShort(hostMB, guestMB int64) bool { return guestMB > 0 && hostMB < guestMB }
+
+// applySwap fills the host block's swap numbers from the node status and the guests' specs, and logs the short
+// state once per change.
+func (c *Collector) applySwap(h *HostMetrics, ns proxmox.NodeStatus, guests []Guest) {
+	hostMB := ns.Swap.Total / (1 << 20)
+	var guestMB int64
+	for _, g := range guests {
+		if g.Spec != nil {
+			guestMB += g.Spec.SwapBytes / (1 << 20)
+		}
+	}
+	h.HostSwapMB, h.GuestSwapMB = &hostMB, &guestMB
+	if HostSwapShort(hostMB, guestMB) {
+		if c.swapShortWarned.CompareAndSwap(false, true) {
+			c.logger.Warn("host-swap: the host's swap is below its guests' swap allowance — a memory burst can freeze the box (R-1192); the hub is told (host_swap_short)",
+				"host_swap_mb", hostMB, "guest_swap_mb", guestMB)
+		}
+		return
+	}
+	if c.swapShortWarned.CompareAndSwap(true, false) {
+		c.logger.Info("host-swap: the host's swap covers its guests' swap allowance again", "host_swap_mb", hostMB, "guest_swap_mb", guestMB)
+	}
+}

internal/hub/r1192_hostswap_test.go

@@ -0,0 +1,42 @@
+package hub
+
+import (
+	"bytes"
+	"log/slog"
+	"strings"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1192 follow-up (night 2026-10-21): the host report carries the host's swap and the guests' swap allowance, and the
+// agent says ONCE in its log when the host's swap is below it (and once when it is covered again).
+// RED-PROOF: with applySwap not called the numbers are nil; with the CompareAndSwap removed the WARN repeats.
+func TestR1192_HostReportCarriesSwapAndWarnsOnce(t *testing.T) {
+	var buf bytes.Buffer
+	c := &Collector{logger: slog.New(slog.NewTextHandler(&buf, nil))}
+	guests := []Guest{{VMID: 9201, Spec: &GuestSpec{SwapBytes: 4096 << 20}}, {VMID: 9202, Spec: nil}}
+	var ns proxmox.NodeStatus // a host with no swap
+	for i := 0; i < 3; i++ {
+		h := HostMetrics{}
+		c.applySwap(&h, ns, guests)
+		if h.HostSwapMB == nil || h.GuestSwapMB == nil || *h.HostSwapMB != 0 || *h.GuestSwapMB != 4096 {
+			t.Fatalf("report numbers: host %v guest %v", h.HostSwapMB, h.GuestSwapMB)
+		}
+	}
+	if n := strings.Count(buf.String(), "host-swap: the host's swap is below"); n != 1 {
+		t.Fatalf("the short WARN must be logged once, got %d:\n%s", n, buf.String())
+	}
+	ns.Swap.Total = 7935 << 20 // the installer's 8 GB host
+	h := HostMetrics{}
+	c.applySwap(&h, ns, guests)
+	if *h.HostSwapMB != 7935 || HostSwapShort(*h.HostSwapMB, *h.GuestSwapMB) {
+		t.Fatalf("a covered host must not be short: %v / %v", *h.HostSwapMB, *h.GuestSwapMB)
+	}
+	if !strings.Contains(buf.String(), "covers its guests' swap allowance again") {
+		t.Fatalf("the recovery must be said once:\n%s", buf.String())
+	}
+	if HostSwapShort(0, 0) {
+		t.Fatal("a host whose guests have no swap allowance is never short")
+	}
+}

internal/hub/report.go

@@ -292,6 +292,10 @@ type WireguardStatus struct {
 	Registered        bool   `json:"registered"` // the registration marker exists
 	Active            bool   `json:"active"`     // wg-quick@wg-felhom is-active
 	LastHandshakeAgeS *int64 `json:"last_handshake_age_s,omitempty"`
+	// HandshakeFutureS (R-1204): seconds the latest handshake lies in the FUTURE by this box's clock — set (> 0) only
+	// then. A box that ran once with its clock ahead keeps a handshake stamped in the future; WireGuard on ep0 then
+	// refuses every newer handshake, and LastHandshakeAgeS (clamped at 0) reads „just now" for ever. The hub alarms on it.
+	HandshakeFutureS *int64 `json:"handshake_future_s,omitempty"`
 	AssignedIP       string `json:"assigned_ip,omitempty"` // from the marker, e.g. "10.77.0.2/32"
 }
 
@@ -338,6 +342,11 @@ type HostMetrics struct {
 	//
 	// Empty = unreadable/absent, which the hub treats as UNKNOWN, never as drift.
 	WrapperSHA256 string `json:"wrapper_sha256,omitempty"`
+	// HostSwapMB / GuestSwapMB (R-1192 follow-up, night 2026-10-21): the host's swap (Proxmox node status — the
+	// kernel's SwapTotal) and the sum of its guests' swap allowance. The guest swap fix (swap = RAM) needs swap ON THE
+	// HOST behind it; the hub raises host_swap_short when the first is below the second. nil = not read (older agent).
+	HostSwapMB  *int64 `json:"host_swap_mb,omitempty"`
+	GuestSwapMB *int64 `json:"guest_swap_mb,omitempty"`
 }
 
 // Guest is one LXC. The agent reports vmid; the hub derives the guest PK
@@ -355,11 +364,12 @@ type GuestSpec struct {
 	Cores       int   `json:"cores"`
 	MemoryBytes int64 `json:"memory_bytes"`
 	DiskBytes   int64 `json:"disk_bytes"`
+	SwapBytes   int64 `json:"swap_bytes,omitempty"` // R-1192: the guest's swap allowance (PVE config `swap`, MB → bytes)
 }
 
 // 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"
 }
 

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/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
+//   - Crash-safety is IN-MEMORY BY DESIGN: an agent restart loses R AND the package it opens,
-//     on the hub; a re-run supersedes it). Status reporting "none" after a restart is the honest
+//     which is SAFE (R-978: the package was never uploaded, so the hub still holds the previous one
-//     answer. No journal, deliberately — persistence is the one property R must never have.
+//     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
+		job.Uploaded = false // R-978: uploaded by the claim
-		s.finishEscrowCeremony(job, out.RecoveryCode)
+		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`) {
+	if !strings.Contains(body, `"restic_pw_sealed":true`) || !strings.Contains(body, `"uploaded":false`) {
-		t.Fatalf("summary fields missing: %s", body)
+		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/guestmemory.go

@@ -57,6 +57,10 @@ type MemoryInfo struct {
 	MaxMB       int64 `json:"max_mb"`
 	FloorMB     int64 `json:"floor_mb"`
 	Running     bool  `json:"running"`
+	// HostSwapMB / GuestSwapMB (R-1192 follow-up): the host's swap and this guest's swap allowance, so the controller
+	// can tell the household when the host has too little swap behind the guest (its install page). nil = not read.
+	HostSwapMB  *int64 `json:"host_swap_mb,omitempty"`
+	GuestSwapMB *int64 `json:"guest_swap_mb,omitempty"`
 }
 
 // memoryResizeRequest is the POST /guest/memory body.
@@ -94,6 +98,8 @@ type memoryBounds struct {
 	maxMB       int64
 	floorMB     int64
 	running     bool
+	hostSwapMB  int64
+	guestSwapMB int64
 }
 
 func (b memoryBounds) refusal(code string) memoryRefusal {
@@ -104,7 +110,9 @@ func (b memoryBounds) refusal(code string) memoryRefusal {
 }
 
 func (b memoryBounds) info(vmid int) MemoryInfo {
+	hostSwap, guestSwap := b.hostSwapMB, b.guestSwapMB
 	return MemoryInfo{
+		HostSwapMB: &hostSwap, GuestSwapMB: &guestSwap,
 		VMID: vmid, AllocatedMB: b.allocatedMB, UsageMB: b.usageMB, HostTotalMB: b.hostTotalMB,
 		MinMB: b.minMB, MaxMB: b.maxMB, FloorMB: b.floorMB, Running: b.running,
 	}
@@ -139,6 +147,8 @@ func (s *Server) readMemoryBounds(ctx context.Context, vmid int) (memoryBounds,
 		hostTotalMB: node.Memory.Total / mib, // bytes → MB, floor (conservative for max)
 		minMB:       minGuestMemoryMB,
 		running:     st.Status == "running",
+		hostSwapMB:  node.Swap.Total / mib, // R-1192 follow-up: the kernel's SwapTotal, as Proxmox reports it
+		guestSwapMB: cfg.Swap,              // PVE config swap is MB
 	}
 	b.maxMB = b.hostTotalMB - hostReserveMB
 	b.floorMB = b.usageMB + shrinkUsageMarginMB
@@ -225,7 +235,8 @@ func (s *Server) handleGuestMemoryResize(w http.ResponseWriter, r *http.Request,
 	}
 
 	// Apply. SetConfig may be synchronous (empty UPID) or return a task to wait on.
-	upid, err := s.mem.SetConfig(r.Context(), vmid, map[string]string{"memory": fmt.Sprintf("%d", target)})
+	// R-1192: the swap moves with the memory (GuestSwapMB) — a grown guest with the old swap cap thrashes the same way.
+	upid, err := s.mem.SetConfig(r.Context(), vmid, map[string]string{"memory": fmt.Sprintf("%d", target), "swap": fmt.Sprintf("%d", GuestSwapMB(target))})
 	if err != nil {
 		s.logger.Error("local-api: guest-memory SetConfig", "vmid", vmid, "target_mb", target, "err", err)
 		writeErr(w, http.StatusBadGateway, "memory resize failed: "+err.Error())

internal/localapi/guestmemory_test.go

@@ -25,12 +25,14 @@ type fakeMemory struct {
 	setUPID      string
 	setErr       error
 	applyReflect bool
+	hostSwapB    int64 // R-1192 follow-up: node swap total
+	guestSwapMB  int64
 }
 
 func (f *fakeMemory) GuestConfig(_ context.Context, _ int) (proxmox.GuestConfig, error) {
 	f.mu.Lock()
 	defer f.mu.Unlock()
-	return proxmox.GuestConfig{Memory: f.allocMB}, nil
+	return proxmox.GuestConfig{Memory: f.allocMB, Swap: f.guestSwapMB}, nil
 }
 func (f *fakeMemory) GuestStatus(_ context.Context, vmid int) (proxmox.Guest, error) {
 	f.mu.Lock()
@@ -42,6 +44,7 @@ func (f *fakeMemory) NodeStatus(_ context.Context) (proxmox.NodeStatus, error) {
 	defer f.mu.Unlock()
 	var ns proxmox.NodeStatus
 	ns.Memory.Total = f.hostTotalB
+	ns.Swap.Total = f.hostSwapB
 	return ns, nil
 }
 func (f *fakeMemory) SetConfig(_ context.Context, _ int, params map[string]string) (string, error) {
@@ -151,6 +154,10 @@ func TestGuestMemory_GrowAndShrink(t *testing.T) {
 		if f.lastParams["memory"] != "12288" {
 			t.Errorf("SetConfig param = %q, want 12288", f.lastParams["memory"])
 		}
+		// R-1192: the swap moves with the memory.
+		if f.lastParams["swap"] != "12288" {
+			t.Errorf("SetConfig swap = %q, want 12288 (swap = RAM)", f.lastParams["swap"])
+		}
 	})
 	t.Run("shrink above floor", func(t *testing.T) {
 		f := stdFixture() // floor = max(2048, 3000+512) = 3512

internal/localapi/guestswap.go

@@ -0,0 +1,73 @@
+package localapi
+
+import (
+	"context"
+	"fmt"
+	"log/slog"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// ── R-1192 (2026-10-21): the guest may swap as much as its RAM ────────────────────────────────────────────────────
+//
+// MEASURED (felhom.eu audits/fixes-2026-10-21/P4-freeze/): on the minimum machine (8 GB host, guest 4096 MB) the
+// household's first uploads made Immich load its analysis models (+1.15 GB) on top of 8 apps' 1.75 GB; the guest hit
+// its 4096 MB limit with its swap capped at Proxmox's LXC default of 512 MB — and full. The kernel could then evict
+// only file pages: ~360 MB was left for the program code of ~25 containers, re-read from disk without end (150 000–
+// 600 000 refaults and ~1 GB of reads per 10 s; memory „full" pressure 92–95 %, load 155, the dashboard dead, 0 OOM
+// kills). Raising the cap to 4096 MB live ended it in under a minute (memory „full" 96 % → 1.7 %, dashboard back
+// 42 s later) while the uploads ran on. The host had 7.4 GB of swap free the whole time.
+//
+// The agent never set `swap` at all, so every guest had the 512 MB default. Now:
+//   - a new guest is created with swap = its memory (reconcile.bringup);
+//   - a RAM resize sets swap with the memory (handleGuestMemoryResize);
+//   - at agent start, every LXC in the felhom pool whose swap is below its memory is raised to it (this file).
+// Raising swap is a live cgroup change (memory.swap.max), no restart. It never lowers a swap someone set higher.
+
+// GuestSwapMB is the swap a guest of memoryMB gets: the same as its RAM.
+func GuestSwapMB(memoryMB int64) int64 { return memoryMB }
+
+// SwapOps is the Proxmox surface EnsureGuestSwap needs (satisfied by *proxmox.Client).
+type SwapOps interface {
+	Pool(ctx context.Context, name string) (proxmox.PoolInfo, error)
+	GuestConfig(ctx context.Context, vmid int) (proxmox.GuestConfig, error)
+	SetConfig(ctx context.Context, vmid int, params map[string]string) (string, error)
+	WaitTask(ctx context.Context, upid string, opts proxmox.WaitOptions) (proxmox.TaskStatus, error)
+}
+
+// EnsureGuestSwap raises the swap of every LXC guest in pool to GuestSwapMB(memory) when it is below. Returns how
+// many guests were raised. One guest's failure is logged and the others are still handled.
+func EnsureGuestSwap(ctx context.Context, ops SwapOps, pool string, logger *slog.Logger) (int, error) {
+	p, err := ops.Pool(ctx, pool)
+	if err != nil {
+		return 0, fmt.Errorf("guest swap: read pool %s: %w", pool, err)
+	}
+	raised := 0
+	for _, m := range p.Members {
+		if m.Type != "lxc" || m.VMID <= 0 {
+			continue
+		}
+		cfg, err := ops.GuestConfig(ctx, m.VMID)
+		if err != nil {
+			logger.Warn("guest-swap: config read failed (left as it is)", "vmid", m.VMID, "err", err)
+			continue
+		}
+		want := GuestSwapMB(cfg.Memory)
+		if cfg.Memory <= 0 || cfg.Swap >= want {
+			logger.Debug("guest-swap: nothing to do", "vmid", m.VMID, "memory_mb", cfg.Memory, "swap_mb", cfg.Swap)
+			continue
+		}
+		upid, err := ops.SetConfig(ctx, m.VMID, map[string]string{"swap": fmt.Sprintf("%d", want)})
+		if err == nil && upid != "" {
+			_, err = ops.WaitTask(ctx, upid, proxmox.WaitOptions{})
+		}
+		if err != nil {
+			logger.Warn("guest-swap: raise failed (left as it is)", "vmid", m.VMID, "from_mb", cfg.Swap, "to_mb", want, "err", err)
+			continue
+		}
+		raised++
+		logger.Info("guest-swap: raised (R-1192 — a guest swap cap below its RAM turned a memory burst into a freeze)",
+			"vmid", m.VMID, "from_mb", cfg.Swap, "to_mb", want)
+	}
+	return raised, nil
+}

internal/localapi/guestswap_test.go

@@ -0,0 +1,71 @@
+package localapi
+
+import (
+	"context"
+	"errors"
+	"io"
+	"log/slog"
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1192: the freeze measured on the minimum machine (felhom.eu audits/fixes-2026-10-21/P4-freeze/) — a guest whose
+// swap is capped below its RAM evicts program code instead of idle memory. EnsureGuestSwap raises every pool guest's
+// swap to its RAM, never lowers one, skips non-LXC members, and survives one guest's failure.
+// COMPANION RED-PROOF: return 0 before the SetConfig call → „raised 0, want 2".
+type fakeSwapOps struct {
+	cfg     map[int]proxmox.GuestConfig
+	failSet map[int]bool
+	set     map[int]map[string]string
+}
+
+func (f *fakeSwapOps) Pool(context.Context, string) (proxmox.PoolInfo, error) {
+	return proxmox.PoolInfo{Members: []proxmox.PoolMember{{VMID: 9201, Type: "lxc"}, {VMID: 9202, Type: "lxc"}, {VMID: 300, Type: "qemu"},
+		{Type: "storage"}, {VMID: 9203, Type: "lxc"}, {VMID: 9204, Type: "lxc"}}}, nil
+}
+func (f *fakeSwapOps) GuestConfig(_ context.Context, vmid int) (proxmox.GuestConfig, error) {
+	c, ok := f.cfg[vmid]
+	if !ok {
+		return c, errors.New("no such guest")
+	}
+	return c, nil
+}
+func (f *fakeSwapOps) SetConfig(_ context.Context, vmid int, p map[string]string) (string, error) {
+	if f.failSet[vmid] {
+		return "", errors.New("pve said no")
+	}
+	if f.set == nil {
+		f.set = map[int]map[string]string{}
+	}
+	f.set[vmid] = p
+	return "", nil
+}
+func (f *fakeSwapOps) WaitTask(context.Context, string, proxmox.WaitOptions) (proxmox.TaskStatus, error) {
+	return proxmox.TaskStatus{}, nil
+}
+
+func TestR1192_EnsureGuestSwapRaisesEveryLowSwapToItsRAM(t *testing.T) {
+	f := &fakeSwapOps{cfg: map[int]proxmox.GuestConfig{
+		9201: {Memory: 4096, Swap: 512},  // the measured box: raised
+		9202: {Memory: 2048, Swap: 4096}, // already higher: never lowered
+		9203: {Memory: 6144, Swap: 0},    // none: raised
+		9204: {Memory: 4096, Swap: 512},  // the write fails: logged, the others still done
+	}, failSet: map[int]bool{9204: true}}
+	n, err := EnsureGuestSwap(context.Background(), f, "felhom", slog.New(slog.NewTextHandler(io.Discard, nil)))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if n != 2 {
+		t.Fatalf("raised %d, want 2 (9201, 9203)", n)
+	}
+	if f.set[9201]["swap"] != "4096" || f.set[9203]["swap"] != "6144" {
+		t.Fatalf("writes = %v", f.set)
+	}
+	if _, touched := f.set[9202]; touched {
+		t.Fatal("a swap above the RAM was lowered")
+	}
+	if len(f.set[9201]) != 1 {
+		t.Fatalf("the swap raise wrote more than the swap: %v", f.set[9201])
+	}
+}

internal/localapi/r1192_hostswap_test.go

@@ -0,0 +1,39 @@
+package localapi
+
+import (
+	"encoding/json"
+	"testing"
+)
+
+// R-1192 follow-up (night 2026-10-21): GET /guest/memory carries the host's swap and this guest's swap allowance, so
+// the controller's install page can say when the host has no swap behind the guest. 0 is a real answer (no swap);
+// absent means an older agent.
+// RED-PROOF: without the two MemoryInfo fields the keys are missing → fails on „host_swap_mb absent".
+func TestR1192_GuestMemoryCarriesHostAndGuestSwap(t *testing.T) {
+	for _, c := range []struct {
+		name            string
+		hostB, guestMB  int64
+		wantHost, wantG int64
+	}{
+		{"installer host (7.75 GB swap), guest swap = RAM", 7935 * mib, 8192, 7935, 8192},
+		{"a host with NO swap", 0, 8192, 0, 8192},
+	} {
+		f := stdFixture()
+		f.hostSwapB, f.guestSwapMB = c.hostB, c.guestMB
+		w := do(t, memServer(t, f).Handler(), "GET", "/guest/memory", "A", "")
+		var m struct {
+			Data map[string]any `json:"data"`
+		}
+		if err := json.Unmarshal(w.Body.Bytes(), &m); err != nil {
+			t.Fatal(err)
+		}
+		hs, ok := m.Data["host_swap_mb"].(float64)
+		if !ok {
+			t.Fatalf("%s: host_swap_mb absent: %s", c.name, w.Body.String())
+		}
+		gs, _ := m.Data["guest_swap_mb"].(float64)
+		if int64(hs) != c.wantHost || int64(gs) != c.wantG {
+			t.Fatalf("%s: host %v guest %v, want %d / %d", c.name, hs, gs, c.wantHost, c.wantG)
+		}
+	}
+}

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 →
@@ -302,6 +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)
@@ -426,8 +435,17 @@ 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
+	// 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
@@ -603,6 +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).
@@ -878,6 +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
@@ -941,6 +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.
@@ -1314,6 +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) {
@@ -1333,6 +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

@@ -2,6 +2,7 @@ package localapi
 
 import (
 	"context"
+	"errors"
 	"fmt"
 	"log/slog"
 	"os"
@@ -160,6 +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)
 }
@@ -334,8 +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
 	}
@@ -346,6 +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_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/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

@@ -645,6 +645,9 @@ func buildBringUpConfig(spec BringUpSpec, cfg proxmox.GuestConfig) map[string]st
 	}
 	if spec.MemoryMB > 0 {
 		params["memory"] = strconv.Itoa(spec.MemoryMB)
+		// R-1192: swap = RAM. Left unset, Proxmox gives an LXC 512 MB, and a guest that fills its RAM can then evict
+		// only program code — the measured freeze (localapi.GuestSwapMB carries the reasoning and the numbers).
+		params["swap"] = strconv.Itoa(spec.MemoryMB)
 	}
 	for i, m := range spec.Mounts {
 		// backup=1 for data-bearing mounts: extra LXC mountpoints default to backup=0 = EXCLUDED

internal/reconcile/bringup_swap_test.go

@@ -0,0 +1,19 @@
+package reconcile
+
+import (
+	"testing"
+
+	"gitea.dooplex.hu/admin/felhom-agent/internal/proxmox"
+)
+
+// R-1192: a new guest is created with swap = its RAM (left unset, Proxmox gives an LXC 512 MB — the measured freeze,
+// felhom.eu audits/fixes-2026-10-21/P4-freeze/). COMPANION RED-PROOF: drop params["swap"] in buildBringUpConfig.
+func TestR1192_BringUpSetsSwapToTheRAM(t *testing.T) {
+	p := buildBringUpConfig(BringUpSpec{MemoryMB: 4096}, proxmox.GuestConfig{})
+	if p["memory"] != "4096" || p["swap"] != "4096" {
+		t.Fatalf("bring-up params memory=%q swap=%q, want 4096/4096", p["memory"], p["swap"])
+	}
+	if q := buildBringUpConfig(BringUpSpec{}, proxmox.GuestConfig{}); q["swap"] != "" {
+		t.Fatalf("no memory in the spec → no swap written either, got %q", q["swap"])
+	}
+}

internal/storage/hostops.go

@@ -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)
+	bout, _, berr := h.runner.Run(ctx, h.bins.Blkid, "-p", "-o", "export", device)
-	for k, v := range parseBlkidExport(bout) {
+	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.
+// whether it has child partitions, is mounted (itself or any child), and its fstype/pttype. parsed=false
-func parseLsblkDevice(out []byte) (hasChildren, mounted bool, fstype, pttype string) {
+// 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/wgtunnel/manager.go

@@ -661,6 +661,9 @@ func (m *Manager) Status(ctx context.Context) *hub.WireguardStatus {
 			if age, ok := parseHandshakeAge(string(out), m.now()); ok {
 				st.LastHandshakeAgeS = &age
 			}
+			if fut, ok := handshakeFuture(string(out), m.now()); ok {
+				st.HandshakeFutureS = &fut
+			}
 		}
 	}
 	if st.Pubkey == "" && !st.Registered && !st.Active {
@@ -669,6 +672,28 @@ func (m *Manager) Status(ctx context.Context) *hub.WireguardStatus {
 	return st
 }
 
+// handshakeFuture (R-1204) reports how many seconds the latest handshake lies in the future, ok only when it does
+// (by more than a minute — clock jitter is not the case). parseHandshakeAge clamps that case to 0 („just now"), which
+// is what made a box whose clock had run ahead look healthy while ep0 refused its tunnel. Pinned by
+// TestR1204_AFutureHandshakeIsReported.
+func handshakeFuture(out string, now time.Time) (int64, bool) {
+	for _, line := range strings.Split(strings.TrimSpace(out), "\n") {
+		f := strings.Fields(line)
+		if len(f) != 2 {
+			continue
+		}
+		epoch, err := strconv.ParseInt(f[1], 10, 64)
+		if err != nil || epoch <= 0 {
+			continue
+		}
+		if fut := epoch - now.Unix(); fut > 60 {
+			return fut, true
+		}
+		return 0, false
+	}
+	return 0, false
+}
+
 // parseHandshakeAge parses `wg show <if> latest-handshakes` output ("<peer-pub>\t<epoch>").
 // epoch 0 = no handshake yet → not ok (nil in the report; nil ≠ 0).
 func parseHandshakeAge(out string, now time.Time) (int64, bool) {

internal/wgtunnel/manager_test.go

@@ -696,3 +696,31 @@ func TestInitialResolveFailureNoTeardown(t *testing.T) {
 		t.Errorf("did not recover after DNS returned: %v", rr.calls)
 	}
 }
+
+// R-1204: a box that ran once with its clock ahead keeps a handshake stamped in the FUTURE — ep0 then refuses every
+// newer one. The age stays clamped at 0 (unchanged contract), and the new field says how far ahead it lies; a
+// handshake in the past (or within a minute of now) never sets it.
+// RED-PROOF: with Status not setting HandshakeFutureS the first assertion fails — seen failing 2026-10-22.
+func TestR1204_AFutureHandshakeIsReported(t *testing.T) {
+	fh := &fakeHub{}
+	m, _, active, _ := testManager(t, fh)
+	ctx := context.Background()
+	m.Apply(ctx, false, nil)
+	*active = true
+	m.now = func() time.Time { return time.Unix(1783107115-2*24*3600, 0) } // the clock came back two days earlier
+	st := m.Status(ctx)
+	if st.HandshakeFutureS == nil || *st.HandshakeFutureS != 2*24*3600 {
+		t.Fatalf("handshake_future_s = %v, want %d", st.HandshakeFutureS, 2*24*3600)
+	}
+	if st.LastHandshakeAgeS == nil || *st.LastHandshakeAgeS != 0 {
+		t.Fatalf("the age stays clamped at 0: %v", st.LastHandshakeAgeS)
+	}
+	m.now = func() time.Time { return time.Unix(1783107115+100, 0) }
+	if st = m.Status(ctx); st.HandshakeFutureS != nil {
+		t.Fatalf("a past handshake must not set handshake_future_s: %v", *st.HandshakeFutureS)
+	}
+	m.now = func() time.Time { return time.Unix(1783107115-30, 0) } // 30 s of jitter is not the case
+	if st = m.Status(ctx); st.HandshakeFutureS != nil {
+		t.Fatalf("clock jitter must not set handshake_future_s")
+	}
+}

scripts/agent_gates.py

@@ -15,6 +15,9 @@ Gates (all must pass; **non-zero exit on any failure**):
                      or because someone is offline on a train. CI runs the FULL set for exactly this
                      reason: it is the machine that can afford a network check, and it is the half
                      that emails when something is wrong.
+  …  security-scan   govulncheck (called vulnerabilities) + gitleaks on the pushed commits, WARN-FIRST: it lists,
+                     it never refuses (R-986). Shared: felhom.eu/scripts/security_scan_gate.py. In the fast set
+                     although it reads vuln.go.dev — offline it says NOT CHECKED, so it cannot stop a push.
   …  gate-decoys     the decoy suite (test_gate_decoys.py): every gate here still convicts its decoys
                      (R-1116). ~20 s in the workspace, so the full run only: not the hook, but CI
                      runs the full set on every push. Scratch clones and a local fake Gitea only.
@@ -49,6 +52,9 @@ SHARED_INSTRUCTIONS = os.path.join(
 # R-389 — shared, like the two above: it lives in felhom.eu/scripts/ and is never copied.
 SHARED_OBSERVATIONS = os.path.join(
     os.path.dirname(ROOT), "felhom.eu", "scripts", "observations_gate.py")
+# R-986 — shared too: the dependency and secret scanner, warn-first.
+SHARED_SECURITY = os.path.join(
+    os.path.dirname(ROOT), "felhom.eu", "scripts", "security_scan_gate.py")
 
 # (label, absolute script path, args, fast)
 GATES = [
@@ -67,6 +73,13 @@ GATES = [
     # runs the FULL set, so every push is still judged by it. It plants only in scratch clones and asks a fake
     # Gitea on 127.0.0.1 (never the network, never the real tree); it does not call this runner (no recursion).
     ("gate-decoys", os.path.join(ROOT, "scripts", "test_gate_decoys.py"), [], False),
+    # R-1160 (2026-10-19) — every Go file is gofmt-clean; the controller's gofmt_gate.py shape. Fast: one gofmt -l.
+    # Where the runner has no Go it says NOT CHECKED out loud and passes; the pre-push hook (with Go) enforces it.
+    ("gofmt", os.path.join(ROOT, "scripts", "gofmt_gate.py"), [], True),
+    # R-986 (2026-10-19) — govulncheck + gitleaks, WARN-FIRST (exit 0 whatever it finds). Fast although it reads
+    # vuln.go.dev: it cannot refuse, so a network blink is a NOT CHECKED line. ~4 s here. CI: gitleaks fetched pinned
+    # (FELHOM_SCAN_FETCH=1), no Go, so govulncheck says NOT CHECKED there.
+    ("security-scan", SHARED_SECURITY, [ROOT], True),
 ]
 
 VERDICT = {0: "OK", 1: "FAILED", 2: "INCONCLUSIVE"}

scripts/gofmt_gate.py

@@ -0,0 +1,79 @@
+#!/usr/bin/env python3
+"""gofmt_gate.py — every Go file in this module is gofmt-clean (R-1160; the controller's R-454 gate, same shape).
+
+MEASURED 2026-10-19: 14 agent files were not gofmt-clean and nothing noticed (fixed by one formatting-only commit,
+1edfd8e). The controller's history below is why a gate and not a one-off clean-up.
+
+MEASURED 2026-09-02: five internal/web test files had been gofmt-unclean for an unknown length of
+time and nothing noticed — `go vet` does not check formatting and no gate did. By 2026-10-05 the five
+were clean and TWELVE others were not: the count only grows while nothing looks, and every `gofmt -l`
+someone runs by hand is then noise that hides a real one.
+
+Run from the repo root:  python3 scripts/gofmt_gate.py
+Exit 0 clean · 1 convicted (lists the files) · 2 inconclusive (gofmt not found — never a pass).
+
+gofmt is the Go toolchain's own formatter (not a shell utility), found on PATH or under
+`go env GOROOT`/bin. Decoy: scripts/test_gate_decoys.py ("gofmt"), in a scratch copy of this script.
+"""
+import os
+import shutil
+import subprocess
+import sys
+
+ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+ROOTS = ["cmd", "internal"]
+
+
+def find_gofmt():
+    p = shutil.which("gofmt")
+    if p:
+        return p
+    go = shutil.which("go")
+    if go:
+        try:
+            root = subprocess.check_output([go, "env", "GOROOT"], text=True).strip()
+            cand = os.path.join(root, "bin", "gofmt")
+            if os.path.isfile(cand):
+                return cand
+        except Exception:
+            pass
+    return None
+
+
+def in_ci():
+    """True on the Gitea Actions runner (act sets GITHUB_ACTIONS, Gitea sets GITEA_ACTIONS)."""
+    return os.environ.get("GITEA_ACTIONS") == "true" or os.environ.get("GITHUB_ACTIONS") == "true"
+
+
+def main():
+    gofmt = find_gofmt()
+    if not gofmt:
+        # The CI runner may have no Go toolchain (felhom.eu CLAUDE.md, R-887 note). There the gate
+        # says so OUT LOUD and passes: the pre-push hook on the developer's machine, where Go always exists, is where
+        # formatting is enforced. Anywhere else a missing gofmt stays INCONCLUSIVE — never a silent pass.
+        # Pinned by the "gofmt: CI without Go" and "gofmt: dev without Go" decoys in test_gate_decoys.py.
+        if in_ci():
+            print("gofmt gate NOT CHECKED in CI: no Go toolchain on the runner — the pre-push hook (with Go) enforces it")
+            return 0
+        print("gofmt gate INCONCLUSIVE: gofmt not found on PATH or under `go env GOROOT`/bin")
+        return 2
+    roots = [r for r in ROOTS if os.path.isdir(os.path.join(ROOT, r))]
+    if not roots:
+        print("gofmt gate INCONCLUSIVE: none of %s exists under %s" % (ROOTS, ROOT))
+        return 2
+    p = subprocess.run([gofmt, "-l"] + roots, cwd=ROOT, capture_output=True, text=True)
+    if p.returncode != 0:
+        print("gofmt gate INCONCLUSIVE: gofmt exited %d: %s" % (p.returncode, p.stderr.strip()[:400]))
+        return 2
+    bad = [l for l in p.stdout.splitlines() if l.strip()]
+    if bad:
+        for f in bad:
+            print("  not gofmt-clean: %s" % f)
+        print("GOFMT GATE FAILED: %d file(s) — run `gofmt -w <file>` (formatting only, no behaviour change)" % len(bad))
+        return 1
+    print("gofmt gate OK — every Go file under %s is gofmt-clean" % ", ".join(roots))
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/test_gate_decoys.py

@@ -66,6 +66,14 @@ COVERS = {
                     "inside an HTML comment (R-426)",
     "observations": "R-419 in THIS repo's REPORT.md: an Observations note SAYING it carries no marker - vs the "
                     "two genuine markers (R-426)",
+    "gofmt": "R-1160: an unformatted .go file planted under internal/ of a SCRATCH copy is convicted by name - vs the "
+             "same tree formatted (passes); with no gofmt reachable, CI passes OUT LOUD and a dev machine is "
+             "INCONCLUSIVE, never a silent pass",
+    "security-scan": "R-986: the shared gate's own decoys (felhom.eu scripts/test_security_scan_gate.py, 8 cases in "
+                     "SCRATCH git repos: a planted token listed by rule and file, its value never printed; a second "
+                     "secret on another line of an accepted file still listed; a called govulncheck finding listed, "
+                     "an imported-only one not; no tool is NOT CHECKED out loud), run from here; plus the gate on THIS "
+                     "repo's tree: exit 0 and its summary line (warn-first - the listing is the fact)",
     "gate-decoys": "this suite, run as a gate of agent_gates.py's full run, which CI runs on every push (R-1116): "
                    "every case above is a decoy of its own gate, and ONE case that stops convicting turns the run "
                    "red - red-proved 2026-10-17 by flipping a reuse-refs FACT case to expect rc 0",
@@ -291,6 +299,32 @@ def release_cases(base, ws):
 
 
 # ── the shared felhom.eu scripts, against THIS repo's inputs ─────────────────────────────────────
+def gofmt_cases(ws):
+    """R-1160: the gofmt gate in a scratch root (its ROOT is its own grandparent directory), never the real tree."""
+    root = os.path.join(ws, "gofmt-root")
+    os.makedirs(os.path.join(root, "scripts"))
+    os.makedirs(os.path.join(root, "internal", "x"))
+    shutil.copy(os.path.join(ROOT, "scripts", "gofmt_gate.py"), os.path.join(root, "scripts", "gofmt_gate.py"))
+    gate = os.path.join(root, "scripts", "gofmt_gate.py")
+    planted = os.path.join(root, "internal", "x", "zz_gofmt_decoy.go")
+    io.open(planted, "w", encoding="utf-8").write(u"package x\n\nfunc  F( )  int {return 1}\n")
+    if shutil.which("gofmt") or shutil.which("go"):
+        rc, out = run([sys.executable, gate], root)
+        report("gofmt: FACT: an unformatted file under internal/", rc, out, 1, ("zz_gofmt_decoy.go", "GOFMT GATE FAILED"))
+        io.open(planted, "w", encoding="utf-8").write(u"package x\n\nfunc F() int { return 1 }\n")
+        rc, out = run([sys.executable, gate], root)
+        report("gofmt: GENUINE: the same file formatted", rc, out, 0, ("gofmt gate OK",))
+    else:
+        print("  --  gofmt: not runnable here (no Go toolchain) - the two cases below pin that mode")
+    for ci, want_rc, want, name in ((True, 0, "NOT CHECKED in CI", "gofmt: CI without Go"),
+                                    (False, 2, "INCONCLUSIVE", "gofmt: dev without Go")):
+        env = {"PATH": os.path.join(ws, "empty-path"), "HOME": ws}
+        if ci:
+            env["GITEA_ACTIONS"] = "true"
+        rc, out = run([sys.executable, gate], root, env=env)
+        report(name, rc, out, want_rc, (want,))
+
+
 def shared_cases(ws):
     """A scratch WORKSPACE: a clone of this repo beside symlinks to the siblings, because the shared
     scripts reach across (REUSE.md cites hub paths; instructions_gate reads the workspace CLAUDE.md)."""
@@ -343,6 +377,20 @@ def shared_cases(ws):
          u"the same minute.**\n", 0)
 
 
+def security_cases():
+    """R-986: the shared scanner's decoys (they build their own scratch repos), then the gate on this repo."""
+    suite = os.path.join(SHARED, "test_security_scan_gate.py")
+    gate = os.path.join(SHARED, "security_scan_gate.py")
+    for f in (suite, gate):
+        if not os.path.isfile(f):
+            fails.append("shared %s is MISSING beside this clone - a failure, never a skip" % f)
+            return
+    rc, out = run([sys.executable, suite], ROOT)
+    report("security-scan: the shared gate's 8 decoys", rc, out, 0, ("all 8 cases passed",))
+    rc, out = run([sys.executable, gate, ROOT], ROOT)
+    report("security-scan: GENUINE: this repo, warn-first", rc, out, 0, ("security-scan:", "govulncheck"))
+
+
 def main():
     srv = Server(("127.0.0.1", 0), Handler)
     threading.Thread(target=srv.serve_forever, daemon=True).start()
@@ -353,6 +401,8 @@ def main():
         published_cases(base)
         release_cases(base, ws)
         shared_cases(ws)
+        gofmt_cases(ws)
+        security_cases()
     finally:
         srv.shutdown()
         srv.server_close()