BIGNIGHT: R-513 filed (P1 security: FileBrowser admin/admin on every box; demo-hp login page public)

documentation/backlog/OPEN-ITEMS.md

@@ -716,6 +716,7 @@ class (an image `VOLUME` at an unmounted path) is still live — `immich-server`
 | **R-510** | **[P1-HIGH] `tester-1`'s tunnel now has its route, and still gives a fresh box 502: the route sends traffic to `https://traefik` WITH certificate checking, and traefik answers the name `traefik` with its default certificate.** MEASURED 2026-09-14 (BIGNIGHT, VM 333, ISO 1.27.1, controller 0.242.0), after the operator's R-505 fix: from DooPlex `https://felhom.enkicsifelhom.hu` → **502 ×3** (18:07:48Z, `server: cloudflare`). The box's `cloudflared` logs `Request failed … tls: failed to verify certificate: x509: certificate is valid for 544346c4….traefik.default, not traefik … ingressRule=0 originService=https://traefik`. Traefik itself holds a valid Let's Encrypt `CN=*.enkicsifelhom.hu` (openssl on 127.0.0.1:443 with SNI). **Control:** demo-hp's working tunnel config reads `{"hostname":"*.enkisfelhom.hu", "originRequest":{"noTLSVerify":true}, "service":"https://traefik"}` — the same route WITH `noTLSVerify`. So the Cloudflare-side public hostname for `*.enkicsifelhom.hu` lacks „No TLS Verify" (or an origin server name). The Cloudflare side is not visible to the session; the inference rests on the log line and the control. Intervention **I2** of the big night: the claim and every dashboard request go to the guest's LAN address with the name forced. **Fix:** operator ticks „No TLS Verify" on that public hostname, then day-0 A.1 names the setting beside the route. | **WAITING-ON-OPERATOR — rank P1-HIGH; owner: operator (Cloudflare route), CC (day-0 A.1 wording after)** |
 | **R-511** | **[P2-MEDIUM] A customer whose box is rebuilt keeps its ep0 PBS token, and then the DR tier can be neither provisioned nor re-issued: the hub's error advises the one action that refuses.** MEASURED 2026-09-14 (BIGNIGHT, VM 333, `tester-1`, DR tier ticked): on the new box's WireGuard registration the hub logged `[ERROR] pbsdr auto-provision for tester-1 (WG-registration hook): the endpoint already holds a PBS token for tester-1 but the hub has no descriptor — use the explicit "Re-issue PBS credentials" action — save the customer config to retry`. The operator's `POST /configs/tester-1/pbsdr-reissue` → **400 `No provisioned PBS DR tier for this customer`** (`hub/internal/web/pbsdr.go` ~411). The token was left by the doorstep walk's host delete (a host delete does not deprovision tenancy; only RESET does, which also removes the tunnel). So a box rebuilt for an existing customer — the reinstall journey — has no whole-guest off-site tier and no button that restores it. **Fix shape:** let re-issue adopt an existing endpoint token when the descriptor is absent (the message already assumes it does), or have host delete offer to drop the PBS token. | **READY — rank P2-MEDIUM; owner: CC (hub)** |
 | **R-512** | **[P2-MEDIUM] Vaultwarden is installed with open registration, and the one control the page tells the customer to use to close it is read-only.** MEASURED 2026-09-14 (BIGNIGHT, VM 333, controller 0.242.0, catalog vaultwarden 1.36.0-alpine): the deploy form sends `SIGNUPS_ALLOWED=true` (catalog default); after install, „Vaultwarden — Beállítások" says „Ez az alkalmazás már telepítve van. **Az alábbi beállítások csak olvashatók.**" and, below it, „Regisztráció engedélyezése — Igen / Nem — Új fiókok regisztrálásának engedélyezése. **Az első fiók létrehozása után állítsd 'Nem'-re.**" At 18:27:20Z a stranger with no invite and no login registered `idegen.probe@example.com` → **200** (`phase3/vaultwarden-stranger-signup.txt`). Once the dashboard is reachable through the tunnel, anyone who guesses `vault.<domain>` can open an account on the household's password server. A route does exist (the Vaultwarden admin panel's own setting, token under „Megjelenítés"), and no screen names it. **Fix shape:** default `SIGNUPS_ALLOWED=false` with an invite-first first step, or make that one field editable after install; the page must not instruct an act it forbids. | **READY — rank P2-MEDIUM; owner: CC (catalog + controller)** |
+| **R-513** | **[P1-HIGH — SECURITY] Every box's file manager (FileBrowser, `files.<domain>`, a launcher tile) accepts the login `admin` / `admin`, and on demo-hp that login page is on the public internet.** MEASURED 2026-09-14 (BIGNIGHT): `POST /api/auth/login?username=admin` with `X-Password: admin` → **200 + a session token** on VM 333 (fresh ISO 1.27.1 install, controller 0.242.0) and on demo-hp guests **9201 and 9202** (loopback, `Host: files.enkisfelhom.hu`); negative control `admin` / wrong → **401** on all three. On VM 333 that token lists both sources — „Adatlemez" (the data drive's `userdata`: documents, media, photos…) and „Beolvasás" (with `paperless`) — `GET /api/users?id=self` 200. **Public exposure, measured by GET only:** `https://files.enkisfelhom.hu/` from DooPlex through Cloudflare → 200, FileBrowser Quantum, `passwordAvailable:true, noAuth:false` (`phase3/filebrowser-public-reachability-demo-hp.txt`); no login was attempted over the internet. The generated `config.yaml` sets no admin credential (FileBrowser's own default applies); no screen shows the customer any FileBrowser login. Geo-restriction narrows who can reach it, it does not authenticate. **Not changed tonight** (9201's standing state is fenced; no product code). **Fix shape:** the controller sets a generated admin password (or proxy auth behind the dashboard session) at stack creation and on every existing box, and shows it where the customer finds app credentials. | **READY — rank P1-HIGH; owner: CC (controller) · operator (rotate on live boxes first)** |
 
 <!-- DUE-CHECKS-BEGIN — machine-readable. Parsed by scripts/due_checks_gate.py.
      One row per dated check. The R-number must have a row above. Dates are UTC.