hub v0.66.0 + ISO v1.20.0: customer self-bind (R-27 slice 1)

Let a customer bind their own freshly-installed appliance without the
operator: operator "Send self-bind link" mints a 7-day tokenized
capability link, emailed (Hungarian, sibling sender) to the customer, who
opens a public /bind/<token> page and proves two factors — the console
pairing code shown on the box screen + their retrieval passphrase — and
the hub stages the bind via the same BindAppliance (provenance
customer_selfbind). The box's ~30s appliance poll delivers.

Viktor's three rulings verbatim: console pairing code (no appliance list
ever rendered), operator-sent tokenized link, 5-attempt lockout ->
"call support". Wrong code == wrong passphrase (one generic failure, no
oracle, both factors compared unconditionally); expiry falls back to
operator-bind unchanged.

THE TRAP: one public prefix /bind/, exempt from auth+CSRF at both /login
gate sites via a single isPublicBindPath predicate (tight trailing-slash
match; ServeMux ..-cleans; handler rejects '/' in token). 9 tests
(Scenarios A-F + F1/F2); 4 red-proofs verified red-then-green (lockout,
oracle, widened-prefix, single-active). GC verdict: no appliance GC ->
the 7-day TTL stands alone. Controller/agent untouched; R-27b deferred.

Green: full hub build/vet/test (17 ok) + bash -n + hub confirm gate.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017qDiBqKKQ5vPB5fXBqu7Kp
This commit is contained in:
2026-07-17 23:56:53 +02:00
parent c6d7a69e6d
commit 592818492c
24 changed files with 1304 additions and 138 deletions
+21
View File
@@ -3,6 +3,27 @@
> Created with the REUSE.md rollout (2026-07-03). Authoritative history: `hub/CHANGELOG.md` (hub), > Created with the REUSE.md rollout (2026-07-03). Authoritative history: `hub/CHANGELOG.md` (hub),
> `website/CHANGELOG.md`, `scripts/CHANGELOG.md`; end-of-task detail in `REPORT.md`. > `website/CHANGELOG.md`, `scripts/CHANGELOG.md`; end-of-task detail in `REPORT.md`.
- **2026-07-17 — CUSTOMER SELF-BIND shipped (hub v0.66.0 + ISO scripts v1.20.0, R-27 slice 1).** A
customer binds their OWN freshly-installed appliance without the operator. Operator clicks **"Send
self-bind link"** on the customer Setup tab → hub mints a **7-day tokenized capability link**
emails it (Hungarian, sibling sender, NOT via the claim engine) → customer opens the **public,
Hungarian `/bind/<token>`** page (no login — the URL token IS the auth) → enters the **console
pairing code** (shown on the box screen) + their **retrieval passphrase** → hub stages the bind via
the same `BindAppliance` (provenance `customer_selfbind`) → the box's ~30 s appliance poll delivers.
**Viktor's three rulings, verbatim:** (a) *"only their own visible"* → console pairing code, **no
appliance list ever rendered**; (b) *first-box entry* → operator-sent tokenized capability link over
Hungarian email; (c) *lockout after 5 failed attempts* → token locks, "call support". Wrong code and
wrong passphrase = **one identical generic failure** (no oracle; both factors compared
unconditionally); expiry falls back to operator-bind unchanged. **THE TRAP (§9.2):** one public
prefix `/bind/`, exempted from auth+CSRF at both `/login` gate sites via a single `isPublicBindPath`
predicate (tight trailing-slash match; ServeMux `..`-cleans; handler rejects `/` in the token) — 4
red-proofs verified red-then-green (lockout→C1, oracle→B, widened-prefix→E, single-active→C4).
**GC verdict:** no appliance-staleness GC exists (`applianceStaleAfter` is a display badge only) → the
7-day token TTL stands alone, single-active-per-customer, no reaper needed. **Controller/agent
untouched; R-27b (controller second-box dismissable prompt) DEFERRED, mechanism sketched.** Green:
full hub `build/vet/test` (17 ok, +9 self-bind tests) + `bash -n`. **LIVE validation pending deploy
(this session).** See REPORT.md.
- **2026-07-17 — PBS DR STORAGE VISIBILITY + Offsite tab split + dual gauges shipped (hub v0.65.0 + - **2026-07-17 — PBS DR STORAGE VISIBILITY + Offsite tab split + dual gauges shipped (hub v0.65.0 +
tenantsync v1.2.0, R-5), LIVE.** Scoping correction (Viktor): "restic box" and "PBS box" are NOT two tenantsync v1.2.0, R-5), LIVE.** Scoping correction (Viktor): "restic box" and "PBS box" are NOT two
Hetzner Storage Boxes — restic = subaccounts on the shared Hetzner box (Hetzner API, v0.64.0); **PBS DR = Hetzner Storage Boxes — restic = subaccounts on the shared Hetzner box (Hetzner API, v0.64.0); **PBS DR =
+62 -102
View File
@@ -2,119 +2,79 @@
> **Overwrite** this file with a summary of the most recent task only (uniform with the other repos; not cumulative). The cumulative hub history lives in [hub/CHANGELOG.md](hub/CHANGELOG.md); the scripts history lives in [scripts/CHANGELOG.md](scripts/CHANGELOG.md). > **Overwrite** this file with a summary of the most recent task only (uniform with the other repos; not cumulative). The cumulative hub history lives in [hub/CHANGELOG.md](hub/CHANGELOG.md); the scripts history lives in [scripts/CHANGELOG.md](scripts/CHANGELOG.md).
## Hub v0.65.0 — PBS DR storage visibility (ep0 `usage` op) + Offsite tab split (Restic / PBS DR) + dual dashboard gauges (R-5) — 2026-07-17 ## Hub v0.66.0 — Customer self-bind (R-27 slice 1) — 2026-07-17
### 1. Baselines used A customer can now bind their **own** freshly-installed appliance without the operator. The operator
sends a **7-day tokenized capability link** by email; the customer opens a **public, Hungarian**
`/bind/<token>` page (no login — the URL token IS the capability), enters the **console pairing code**
shown on the box screen plus their **retrieval passphrase**, and the hub stages the bind through the
same `BindAppliance` the operator uses. The box's ~30 s appliance poll then delivers credentials and
day-0 install proceeds. The controller and agent are **untouched**.
- **felhom.eu** @ `3588a31` (v0.64.0 REPORT), clean on `main`, local == origin re-confirmed. Viktor's three rulings, honoured verbatim:
- **Hub** v0.64.0 → **v0.65.0**. Manifest `felhom-hub:0.64.0``:0.65.0`. 1. **"Only their own visible"** → possession is proven by the console pairing code; **no appliance
- **`scripts/felhom-tenantsync.sh`** v1.1.0 → **v1.2.0** (adds the read-only `usage` op). list is ever rendered** on any public surface.
- READ-ONLY against ep0 and Hetzner throughout (no mutation, no admin token; the usage op is a pure `df`). 2. **First-box entry** → an operator-sent tokenized capability link over Hungarian email (claim-engine
delivery pattern, a **sibling** sender in `web/`+`notify/` — deliberately NOT routed through the
claim engine).
3. **Lockout after 5 failed attempts** → the token locks; the page says *call support*.
### 2. Phase-0 probe (the gate — PASSED) ### Components
On ep0 (`felhom-hetzner`, root SSH, PBS **4.2.3**): `proxmox-backup-manager datastore usage` does not exist | Part | Area | What |
in PBS 4.2, so `df` on the datastore path is the authoritative read-only source. `datastore list |---|---|---|
--output-format json` → `felhom-offsite` path `/srv/pbs-felhom`; `df -B1 --output=size,used,avail /srv/pbs-felhom`: | 1 | store + api + bootstrap | `pairing_code` on `appliance_registrations`; additive `pairing_code` in the register response; stable across idempotent re-register; console banner (`felhom-bootstrap.sh`, ISO **v1.20.0**) |
| 2 | store + notify + web | `selfbind_tokens` table (sha256-at-rest, single-active/customer, 5-attempt lock, one-shot); operator **"Send self-bind link"** button; Hungarian email (sibling sender); F1/F2 honesty |
| 3 | web | public `/bind/<token>` page; **THE TRAP** auth+CSRF exemption via one `isPublicBindPath` predicate; two-factor unconditional compare; provenance event `customer_selfbind`; own rate limiter |
| 4 | tests | Scenarios AF + F1/F2 (9 tests) + **4 red-proofs** |
| 5 | docs + deploy | this REPORT, CHANGELOGs, CONTEXT, capability-map, ROADMAP; build 0.66.0, manifest bump, sync; live validation |
``` ### Security properties (spec §9)
1B-blocks Used Avail
39990112256 7628091392 30686175232
```
All **bytes** (total 37.2 GiB / used 7.1 GiB → ~19%), read-only, in the existing sudo context, no admin - **THE TRAP (§9.2):** exactly one new public prefix `/bind/`, exempted from operator auth and CSRF at
token. Gate PASSED → the op was built (Option A). The v1.2.0 op emits exactly these as the two gate sites `/login` occupies, through a **single** `isPublicBindPath` definition. Matched
`{"status":"ok","total","used","avail"}` — verified against the real ep0 before writing the harness. tightly — trailing slash (no `/bindsecret` sibling), the `http.ServeMux` `..`-cleans before the
handler sees the path, and the handler rejects any token containing `/`. Scenario E asserts a public
link renders logged-out while `/`, `/hosts`, `/customers/…`, `/configuration`, `/offsite` still
redirect to `/login`; Scenario D is its companion.
- **No oracle:** an unknown token folds into "expired" (no *was-this-real* signal); wrong code and
wrong passphrase yield one byte-identical generic failure (both factors compared unconditionally
before deciding); the page renders no appliance data in any state.
- **Custody:** `sha256(token)` at rest, prefix-only in logs, raw token never in logs/events; the
passphrase is never logged/echoed/persisted; only attempt COUNTS are logged (`attempt N/5 … <8hex>…`).
- **No customer-login system** was built; the capability link is the whole auth model. A cross-site
POST without both secrets only burns attempts (accepted + documented).
### 3. Files created / modified ### GC verdict (spec §3)
**Code + script + runbook (commit `7f11cfb`):** There is **no appliance-staleness garbage collection** in the hub — `applianceStaleAfter` (7 d) is a
- `scripts/felhom-tenantsync.sh` → v1.2.0 (read-only `usage` op) `+ scripts/tenantsync-usage-harness.sh` (new Group-A test). DISPLAY badge only; `pruneAll` and `PurgeExpiredLogBundles` touch host reports and log bundles, not
- `hub/internal/tenantsync/client.go``Usage()` + `BoxUsage` + `ErrUsageUnsupported` + response fields `+ client_test.go`. appliances or self-bind tokens. The **7-day token TTL stands alone** and needs no reaper: single-active
- `hub/internal/monitor/pbsdr_box.go` (new) `+ _test.go``PBSDRBoxChecker` (usageReader seam, throttle, 3 states, bands). per customer means at most one row per customer, a re-mint deletes the prior row, and an expired row
- `hub/internal/notify/dispatcher_test.go` — the pbsdr-box operator-only test. simply reads as expired (no security or storage pressure).
- `hub/cmd/hub/main.go``Alerting.PBSDRBoxFill*` config; checker construction with the tenantsync client; 60 s sweep; `SetPBSDRBox`.
- `hub/internal/web/{server.go,offsite.go,pbsdr_box.go(new),pbsdr_box_render_test.go(new),offsite_box_render_test.go,render_test.go}` — PBS view/tile, tab wiring, two-gauge payload, render tests.
- `hub/internal/web/templates/{offsite.html,dashboard.html,style.css}` — Restic/PBS-DR tabs, two gauges, gauge container CSS.
- `documentation/runbooks/offsite-endpoint.md` §10 — v1.2.0 install steps (no sudoers/authorized_keys change).
- `hub/CHANGELOG.md`.
**Docs (commit `a9cd308`):** REPORT (this), CONTEXT, capability-map, ROADMAP. ### Verification
**Manifest (commit `ab91e49`):** `manifests/hub.yaml``:0.65.0`.
### 4. Test results + red-proof outcomes - **Green gate:** `go build ./... && go vet ./... && go test ./...` — all hub packages pass (17 ok),
including 9 new self-bind tests. `bash -n` clean on both ISO scripts. Hub confirm gate OK.
- **Red-proofs (all confirmed red-then-green):**
- lockout removed (`locked = false`) → **C1** red.
- oracle introduced (distinct state on wrong code) → **B** red.
- `/bind/` prefix widened (drop the slash) → **E** red (`/bindsecret` leaks past auth).
- single-active DELETE dropped → **C4** red (the prior link still resolves).
- **D2 regression evidence:** with an operator password configured, `/`, `/hosts`, `/customers/acme`,
`/configuration`, `/offsite` all `302 → /login` through `RequireAuth`, while `/bind/<token>` renders
`200` — proving the exemption is tight (`TestSelfBind_D_AuthGateIntact`, `TestSelfBind_E_TheTrap`).
10 Go tests + a bash harness, all green: ### Live validation
| Test | Group | Asserts | Filled in after deploy — see the CONTEXT.md live-state note.
|------|-------|---------|
| `TestUsage_Op` | B | usage op parses total/used/avail (bytes) |
| `TestUsage_UnknownOpTypedUnsupported` | B | ep0 ≤ v1.1.0 `unknown op` → typed `ErrUsageUnsupported` |
| `TestPBSDRBox_Throttle` | C1 | ≈4 usage calls over an hour of 60 s sweeps (not ≈60) |
| `TestPBSDRBox_FillBands` | C2/C3 | 75→no; 82→warn; 85 in-band→no re-emit; 92→crit; 70→re-arm; re-breach→emit; scope "pbsdr-box" |
| `TestPBSDRBox_Unavailable` | C4 | ErrUsageUnsupported → "unavailable" state, NO alert, no band |
| `TestPBSDRBox_DegradedKeepsLast` | C5 | exec error → degraded, keeps last values, no band transition, no emit |
| `TestProcessEvent_PBSDRBoxScopeOperatorOnly` | C3 | a pbsdr-box event → operator channel ONLY |
| `TestPBSDRPanel_OK` / `_Unavailable` / `_NotConfigured` | D | panel renders ok / the pending-update message / not-configured |
| `tenantsync-usage-harness.sh` | A | usage JSON + exit 0 + **zero mutation** (stub-logged) + provision regression |
**Red-proofs (run-fail-restore), all confirmed red then restored:** ### Not in this slice (deferred)
| # | Removed / broken | Test | Result | - **R-27b** — the controller's second-box flow (a dismissable "bind another box" prompt + bind-later
|---|------------------|------|--------| under settings). Mechanism sketched; not built. The controller/agent were not touched.
| A | the usage op emits a mutation (`acl update`) | harness | FAILED ("usage ran mutations: … acl update …") → restored | - **Multi-box per link** — one link binds one box (single-active, one-shot). Binding several boxes for
| C-esc | the escalation-only guard | `PBSDRBox_FillBands` | FAILED ("same-band must NOT re-emit, got 2") → restored | one customer = repeated operator sends. Noted on the ROADMAP.
| C-unavail | unavailable drives a fill band | `PBSDRBox_Unavailable` | FAILED ("must yield 'unavailable', got State:ok … critical") → restored | - Attaching the appliance's stored SSH host key to the host record on bind (no clean hand-off surface
today — same open item as slice C).
Full suite: `go build ./... && go vet ./... && go test ./...` (17 packages) + `bash -n` + hub confirm gate —
all green. Test count: +10 Go (+2 tenantsync, +4 monitor, +1 notify, +3 web) + 1 bash harness.
### 5. Deploy verification
Built `felhom-hub:0.65.0` on 180 (source `7f11cfb`). **ep0 updated to tenantsync v1.2.0 THIS session** (root
SSH, `.bak-1.1.0` kept; `tr -d '\r'``bash -n` OK → `install -m0755`; NO sudoers/authorized_keys change).
On-box verify: `echo '{"op":"usage"}' | /usr/local/bin/felhom-tenantsync`
`{"status":"ok","total":39990112256,"used":7628349440,"avail":30685917184}`; `fingerprint` regression OK.
Manifest bump (`ab91e49`) → ArgoCD hard-refresh → sync → rollout succeeded; image `:0.65.0`. Startup log:
```
[INFO] felhom-hub 0.65.0 starting
[INFO] PBS DR tenantsync enabled (endpoint 167.233.158.164:22, user felhom-peersync; …)
[INFO] PBS-DR box checker initialized: fill warn=80% crit=90%, refresh 15m0s
```
### 6. Live-rendered values (the PROVEN-LIVE evidence)
Because ep0 was updated to v1.2.0, the PBS gauge shows **real** numbers. The hub UI is operator-password-gated
(CC can't screenshot), so the in-cluster checker's first live poll is the proof — the exact numbers the PBS
panel + gauge render:
```
[INFO] PBS-DR box refreshed: 19.1% full (7.1 GB of 37.2 GB)
```
Live-confirmed end-to-end (hub checker → tenantsync `usage` op → ep0 `df` → snapshot): **7.1 GB of 37.2 GB,
19.1%** — matching the Phase-0 probe. The restic gauge (v0.64.0) continues to render its own live numbers.
Both tabs (Restic / PBS DR) and both dashboard gauges are live; the panel-render is UNIT-verified
(`TestPBSDRPanel_*`), the data proven live.
### 7. NOT yet live-validated
- **The UI render** (tabs + gauges) could not be screenshotted (hub UI password-gated). It is unit-verified
(render tests) and the data behind both gauges is proven live (§6).
- **The PBS fill alert leg** (`pbsdr_box_fill` operator email) is NOT fired live: the datastore is nominal
(19.1%, far from 80/90%). Unit + red-proof verified; a live-fire would email Viktor (offered on demand).
- **The graceful "unavailable" state** was exercised in unit tests, not live — ep0 was updated straight to
v1.2.0, so the hub never saw the old-script path in production (the hub-independence property is verified
by construction + `TestUsage_UnknownOpTypedUnsupported` + `TestPBSDRBox_Unavailable`).
### 8. Observations
- **Separate PBS threshold pair:** shipped as `Alerting.PBSDRBoxFill{Warn,Crit}Percent` defaulting to the
same 80/90 as restic — independently tunable later without touching the restic policy. Whether Viktor
wants a different PBS pair is his call (one-liner). No oversubscription concept for PBS (namespaces, not
quotas) — fill only, as ruled.
- **Dashboard decimal:** the gauges use `%.0f` (e.g. "19%"), so a near-empty box shows "0%" not "0.2%" —
matches the restic tile's existing rounding. The panels show the finer detail.
- The `usage` op is `df`-based because PBS 4.2 has no native `datastore usage` command; if a future PBS adds
one with dedup-aware logical sizing, the op can switch source with no hub/client change (same JSON shape).
@@ -31,8 +31,9 @@
|---|---|---|---|---| |---|---|---|---|---|
| Appliance day-0 install: golden image → first boot → auto-confirm (zero clicks) → claimable box | installer, agent, hub, golden | **PROVEN-LIVE** (nested VM) | `DRILL-day0-vm-2026-07-12`, `DRILL-day0-take2-2026-07-12` | First firing on real customer hardware pending → R-1 | | Appliance day-0 install: golden image → first boot → auto-confirm (zero clicks) → claimable box | installer, agent, hub, golden | **PROVEN-LIVE** (nested VM) | `DRILL-day0-vm-2026-07-12`, `DRILL-day0-take2-2026-07-12` | First firing on real customer hardware pending → R-1 |
| BYO install: `--mode byo`, mandatory caps, host-mutation disclosure, coexistence guards | installer v1.15+, agent | **PARTIAL** | `DRILL-GL6-2026-07-08` (demo box); GL-8 coexistence fixes | Peti clean-slate reinstall on proxmox2 is the first real BYO run of the current path → R-1 | | BYO install: `--mode byo`, mandatory caps, host-mutation disclosure, coexistence guards | installer v1.15+, agent | **PARTIAL** | `DRILL-GL6-2026-07-08` (demo box); GL-8 coexistence fixes | Peti clean-slate reinstall on proxmox2 is the first real BYO run of the current path → R-1 |
| Bare-metal Felhom ISO (blank hardware → zero-touch auto-install → first-boot `host-install`); selectable UEFI loader; **universal secret-free / operator-bind** mode | scripts v1.19.0 (`scripts/iso/`) + hub v0.62.0 + assistant container | **PARTIAL** (nested VM 310/311; live endpoints) | slice A `SPIKE-baremetal-iso-2026-07-16` (build gate, disk-filter fail-safe, stub→host-install fetch); slice B RUNBOOK-B (shim boots+installs OVMF SB-enforcing + SeaBIOS; `--loader mkimage` boots+installs SB-off; mkimage SB-enforcing **FAILS** `Access Denied`; surgery byte-identical); **slice C (2026-07-17): the GENERIC secret-free ISO** — box self-registers as an unclaimed appliance (`POST /api/v1/appliance/register`, one-shot poll delivery, 404-no-oracle — all live-verified through the public ingress), operator binds on the Hosts page, hub delivers credentials once; bootstrap harness proves direct(zero-appliance-calls)/pairing/delivery; artifact proven secret-free (baked env = hub URL only) | **F1 loader caveat:** `--loader mkimage` fixes cheap AMI firmware that can't USB-boot the stock GRUB — UNSIGNED → **Secure Boot must be OFF**; default `shim` keeps SB. **Slice C bind is operator-password-gated** (CC stages, Viktor binds) → the live boot→register→bind→day-0 composition + physical N100 boot fold into the supervised rehearsal (R-1). Customer-facing **self-bind page** = future item (registered) | | Bare-metal Felhom ISO (blank hardware → zero-touch auto-install → first-boot `host-install`); selectable UEFI loader; **universal secret-free / operator-bind** mode | scripts v1.19.0 (`scripts/iso/`) + hub v0.62.0 + assistant container | **PARTIAL** (nested VM 310/311; live endpoints) | slice A `SPIKE-baremetal-iso-2026-07-16` (build gate, disk-filter fail-safe, stub→host-install fetch); slice B RUNBOOK-B (shim boots+installs OVMF SB-enforcing + SeaBIOS; `--loader mkimage` boots+installs SB-off; mkimage SB-enforcing **FAILS** `Access Denied`; surgery byte-identical); **slice C (2026-07-17): the GENERIC secret-free ISO** — box self-registers as an unclaimed appliance (`POST /api/v1/appliance/register`, one-shot poll delivery, 404-no-oracle — all live-verified through the public ingress), operator binds on the Hosts page, hub delivers credentials once; bootstrap harness proves direct(zero-appliance-calls)/pairing/delivery; artifact proven secret-free (baked env = hub URL only) | **F1 loader caveat:** `--loader mkimage` fixes cheap AMI firmware that can't USB-boot the stock GRUB — UNSIGNED → **Secure Boot must be OFF**; default `shim` keeps SB. **Slice C bind is operator-password-gated** (CC stages, Viktor binds) → the live boot→register→bind→day-0 composition + physical N100 boot fold into the supervised rehearsal (R-1). Customer-facing **self-bind page = R-27 slice 1 SHIPPED (hub v0.66.0, 2026-07-17)** — see the dedicated self-bind row |
| Customer claim: one-time emailed code → customer sets own password (bcrypt, operator never sees it) | controller v0.122, hub v0.50 | **PROVEN-LIVE** (drill VM) | `DRILL-day0-vm-2026-07-12` §10/F-4 (gate ON via real edge; claimed, code consumed) | Never executed by a non-Viktor human → R-3. (Dropped mis-cited `CAMPAIGN-4` F-C — that is the escrow-claim 502, not password claim) | | Customer claim: one-time emailed code → customer sets own password (bcrypt, operator never sees it) | controller v0.122, hub v0.50 | **PROVEN-LIVE** (drill VM) | `DRILL-day0-vm-2026-07-12` §10/F-4 (gate ON via real edge; claimed, code consumed) | Never executed by a non-Viktor human → R-3. (Dropped mis-cited `CAMPAIGN-4` F-C — that is the escrow-claim 502, not password claim) |
| Customer binds their own appliance (self-service): operator-sent 7-day tokenized capability link → public two-factor `/bind/<token>` (console pairing code + retrieval passphrase) → hub stages the bind, no operator | hub v0.66.0 + ISO scripts v1.20.0 | **IMPLEMENTED** | hub v0.66.0 (`web/selfbind.go`, `store/selfbind.go`; Scenarios AF + F1/F2; 4 red-proofs verified red — THE TRAP `/bind/` exemption, no-oracle, lockout, single-active); GC verdict §3 (no appliance GC → TTL stands alone) | R-27 **slice 1**. No appliance list ever rendered; wrong code == wrong passphrase (one generic failure); 5-attempt lockout → call support; expiry falls back to operator-bind. **Live first-run + new-ISO console banner pending** (folds into the supervised rehearsal R-1). **R-27b** (controller second-box dismissable prompt) deferred; **multi-box-per-link** = repeated operator sends |
| Escrow ceremony: customer-facing wizard, one-shot R claim, operator zero-knowledge | controller v0.127, agent v0.88/0.89 | **PROVEN-LIVE** (drill VM, endpoint-exact) | agent v0.88.0 REPORT (ceremony ~4s, one-shot claim 200→410, R absent from every payload); `SPIKE-controller-escrow-2026-07-13` | Endpoints driven on the drill VM; customer-facing **browser wizard** leg not yet live-validated. First supervised ceremony with a real customer pending → R-1. **agent v0.89.0:** `/escrow/preflight` `pbs_storage_id` row now live-reloads (reads current agent.json) — a pbsdr convergence that seeds the id flips it green with NO service restart. **hub v0.60.0 (data-first retention):** a re-escrow with a DIFFERENT sealed passphrase no longer destroys the old blob — the hub RETAINS it (`host_escrow_superseded`), so a previous passphrase stays recoverable with its recovery code (turns the reinstall-orphan incident from "history destroyed" into "history recoverable"). Guided-recovery flow = R-26. Red-proof `TestSaveHostEscrow_RetainsSuperseded`. **hub v0.60.1 — custody survives the host lifecycle:** host deletion (with the escrow ack) DEMOTES the current blob to retained custody (moved into `host_escrow_superseded`, never destroyed; existing superseded rows spared); the customer Danger-zone Delete is the one true purge point (cascades both escrow tables incl. already-deleted hosts). No operator path through host lifecycle can lose a blob. Red-proofs `TestDeleteHost_DemotesEscrowNeverDestroys` + `TestDeleteCustomer_PurgesEscrowCustody` | | Escrow ceremony: customer-facing wizard, one-shot R claim, operator zero-knowledge | controller v0.127, agent v0.88/0.89 | **PROVEN-LIVE** (drill VM, endpoint-exact) | agent v0.88.0 REPORT (ceremony ~4s, one-shot claim 200→410, R absent from every payload); `SPIKE-controller-escrow-2026-07-13` | Endpoints driven on the drill VM; customer-facing **browser wizard** leg not yet live-validated. First supervised ceremony with a real customer pending → R-1. **agent v0.89.0:** `/escrow/preflight` `pbs_storage_id` row now live-reloads (reads current agent.json) — a pbsdr convergence that seeds the id flips it green with NO service restart. **hub v0.60.0 (data-first retention):** a re-escrow with a DIFFERENT sealed passphrase no longer destroys the old blob — the hub RETAINS it (`host_escrow_superseded`), so a previous passphrase stays recoverable with its recovery code (turns the reinstall-orphan incident from "history destroyed" into "history recoverable"). Guided-recovery flow = R-26. Red-proof `TestSaveHostEscrow_RetainsSuperseded`. **hub v0.60.1 — custody survives the host lifecycle:** host deletion (with the escrow ack) DEMOTES the current blob to retained custody (moved into `host_escrow_superseded`, never destroyed; existing superseded rows spared); the customer Danger-zone Delete is the one true purge point (cascades both escrow tables incl. already-deleted hosts). No operator path through host lifecycle can lose a blob. Red-proofs `TestDeleteHost_DemotesEscrowNeverDestroys` + `TestDeleteCustomer_PurgesEscrowCustody` |
| DR tier by default: PBS + WireGuard base infra on every install, hub-controlled activation | installer v1.15, agent v0.86, hub v0.51 | **IMPLEMENTED** | `DRILL-day0-take2-2026-07-12` §2 (WG enabled both modes, PBS-DR descriptor auto-provisioned ~1s after WG registration, zero operator steps); ships installer v1.15/agent v0.86/hub v0.51 | Live only on demo/drill fleet. (Cited spike was slice-0 mechanics — shipped nothing; corrected. Candidate upgrade to PROVEN-LIVE — see REPORT.) **agent v0.89.0 closes the F4 non-default-storage-id gap (R-22) — PROVEN-LIVE 2026-07-17:** the reconcile self-grants the ACL through the root wrapper on a pre-check 403 instead of dead-locking. Reproduced F4 on the demo (marker moved aside = reinstall fresh-state + felhom-offsite ACLs revoked) → next reconcile tick `pbsdr: pre-check 403 … self-granting … (R-22)``converged state=adopted` in ~3 s, ACLs self-restored, `pvesm status felhom-offsite`=active, zero operator action. No more one-shot `pveum` grant | | DR tier by default: PBS + WireGuard base infra on every install, hub-controlled activation | installer v1.15, agent v0.86, hub v0.51 | **IMPLEMENTED** | `DRILL-day0-take2-2026-07-12` §2 (WG enabled both modes, PBS-DR descriptor auto-provisioned ~1s after WG registration, zero operator steps); ships installer v1.15/agent v0.86/hub v0.51 | Live only on demo/drill fleet. (Cited spike was slice-0 mechanics — shipped nothing; corrected. Candidate upgrade to PROVEN-LIVE — see REPORT.) **agent v0.89.0 closes the F4 non-default-storage-id gap (R-22) — PROVEN-LIVE 2026-07-17:** the reconcile self-grants the ACL through the root wrapper on a pre-check 403 instead of dead-locking. Reproduced F4 on the demo (marker moved aside = reinstall fresh-state + felhom-offsite ACLs revoked) → next reconcile tick `pbsdr: pre-check 403 … self-granting … (R-22)``converged state=adopted` in ~3 s, ACLs self-restored, `pvesm status felhom-offsite`=active, zero operator action. No more one-shot `pveum` grant |
| Customer RESET (middle lifecycle tier: host delete < RESET < customer Delete): one operator action → pre-first-install; all operational state destroyed, identity + basic config survive | hub v0.61.0, felhom-tenantsync v1.1.0 | **PROVEN-LIVE (external teardown)** | hub v0.61.0 REPORT; **ep0 live drill 2026-07-17** (throwaway `drill-reset-01` with a real backup: deprovision `deleted:true` destroyed the namespace + backup group + token, idempotent re-run `deleted:false`, all 3 real tenants + shared user survived); red-proofs (ack-gate, partial-failure resumability) + orchestration/store/offsite/render tests | External teardown FIRST, DB purge LAST, every leg idempotent; refuses while any host row exists; separate escrow-custody ack; clears claim (fresh code next onboarding); keeps the offsite tier CHOICE, drops provisioned fields. **Not live-clicked:** the web POST is password-gated (CC verifies via render + httptest orchestration tests); the Hetzner sub-account delete is unit-tested + a faithful mirror of the live-proven `ReissueCredentials`. **Consistency gap → R-25b:** the Danger-zone DELETE leaves host rows and doesn't run this teardown | | Customer RESET (middle lifecycle tier: host delete < RESET < customer Delete): one operator action → pre-first-install; all operational state destroyed, identity + basic config survive | hub v0.61.0, felhom-tenantsync v1.1.0 | **PROVEN-LIVE (external teardown)** | hub v0.61.0 REPORT; **ep0 live drill 2026-07-17** (throwaway `drill-reset-01` with a real backup: deprovision `deleted:true` destroyed the namespace + backup group + token, idempotent re-run `deleted:false`, all 3 real tenants + shared user survived); red-proofs (ack-gate, partial-failure resumability) + orchestration/store/offsite/render tests | External teardown FIRST, DB purge LAST, every leg idempotent; refuses while any host row exists; separate escrow-custody ack; clears claim (fresh code next onboarding); keeps the offsite tier CHOICE, drops provisioned fields. **Not live-clicked:** the web POST is password-gated (CC verifies via render + httptest orchestration tests); the Hetzner sub-account delete is unit-tested + a faithful mirror of the live-proven `ReissueCredentials`. **Consistency gap → R-25b:** the Danger-zone DELETE leaves host rows and doesn't run this teardown |
+2 -1
View File
@@ -48,7 +48,8 @@
| ID | Item | Size | Status | Notes | | ID | Item | Size | Status | Notes |
|----|------|------|--------|-------| |----|------|------|--------|-------|
| R-26 | **Guided old-history recovery via a retained superseded escrow + the recovery code.** Enabled by hub v0.60.0 (Part B) which now RETAINS superseded escrow blobs (`host_escrow_superseded`, `ListSupersededEscrow`). Build the flow that, given the customer's recovery code, unwraps a retained old blob → recovers the old repo passphrase → mounts/reads the moved-aside `.orphaned-<date>` repo for restore. | M | idea (enabled by v0.60.0) | Turns "history recoverable in principle" into a real customer-drivable path; pairs with the controller v0.142.0 orphaned-repo move-aside. Origin `DIAGNOSE-offbox-repo-orphaned-2026-07-17` | | R-26 | **Guided old-history recovery via a retained superseded escrow + the recovery code.** Enabled by hub v0.60.0 (Part B) which now RETAINS superseded escrow blobs (`host_escrow_superseded`, `ListSupersededEscrow`). Build the flow that, given the customer's recovery code, unwraps a retained old blob → recovers the old repo passphrase → mounts/reads the moved-aside `.orphaned-<date>` repo for restore. | M | idea (enabled by v0.60.0) | Turns "history recoverable in principle" into a real customer-drivable path; pairs with the controller v0.142.0 orphaned-repo move-aside. Origin `DIAGNOSE-offbox-repo-orphaned-2026-07-17` |
| R-27 | **Customer-facing self-bind page (R-21 slice C follow-on).** Today an unclaimed appliance is bound by the OPERATOR on the Hosts page (hub v0.62.0). Build the customer-facing flow so a customer can claim/bind their own freshly-installed box (e.g. enter a claim code / the appliance's displayed pairing id → the hub binds it to their account → delivery proceeds). Turns "operator binds every box" into true self-service onboarding. | M | idea | Origin: hub v0.62.0 slice C (operator-bind ruling; self-bind deliberately deferred). Reuses the appliance_registrations + one-shot delivery machinery; adds a customer-auth surface + a pairing-id/claim-code channel. Pairs with the claim engine | | R-27 | **Customer-facing self-bind page (R-21 slice C follow-on).** Today an unclaimed appliance is bound by the OPERATOR on the Hosts page (hub v0.62.0). Build the customer-facing flow so a customer can claim/bind their own freshly-installed box. | M | **SHIPPED (slice 1, hub v0.66.0, 2026-07-17)** | **Slice 1 = the FIRST-box flow, DONE:** operator "Send self-bind link" → 7-day tokenized capability link over Hungarian email → public two-factor `/bind/<token>` (console pairing code shown on the box screen + retrieval passphrase) → hub stages the bind via the same `BindAppliance`, no operator. Viktor's 3 rulings verbatim (console code not a list; operator-sent link; 5-attempt lockout→call support). No oracle; expiry falls back to operator-bind; THE TRAP `/bind/` exemption tight (single predicate, 4 red-proofs). GC verdict §3 (no appliance GC → the 7-day TTL stands alone). **Live first-run + new-ISO console banner fold into the supervised rehearsal (R-1).** Reused the appliance_registrations + one-shot delivery machinery; the capability link is the whole customer-auth surface (no login system built). Controller/agent untouched. See hub v0.66.0 REPORT. **Multi-box per link is out of scope by design** (single-active, one-shot → binding several boxes = repeated operator sends) |
| R-27b | **Customer self-bind, second-box flow (controller side).** For a customer who ALREADY has a bound box and installs another, the controller shows a dismissable "bind another box" prompt (and a bind-later entry under settings) that walks to the hub `/bind/` page — so a returning customer isn't emailed a fresh operator-sent link for every box. Mechanism sketched in the hub v0.66.0 REPORT; NOT built (R-27 slice 1 deliberately did not touch the controller). | M | idea (minted by hub v0.66.0) | Origin: hub v0.66.0 slice-1 ship (first-box only). Reuses the same `/bind/` public page + tokenized-link machinery; adds a controller-side entry point + the operator "mint a link for an existing customer" affordance |
| R-25b | **Customer DELETE ↔ RESET consistency.** The middle-tier Customer RESET (hub v0.61.0) runs the full external teardown (Hetzner sub-account/box + PBS namespace/groups/token) and refuses while any host row exists. The Danger-zone DELETE still (a) leaves host rows and (b) does NOT run that teardown — it purges escrow custody + drops the config only. Decide the model: DELETE requires a prior RESET, or DELETE subsumes RESET's teardown, or they stay orthogonal (RESET = recycle-in-place, DELETE = escrow-purge). | S | idea | Origin: hub v0.61.0 RESET ship. Flips a future "customer fully offboarded (external resources released)" map row. Cheap once the model is chosen | | R-25b | **Customer DELETE ↔ RESET consistency.** The middle-tier Customer RESET (hub v0.61.0) runs the full external teardown (Hetzner sub-account/box + PBS namespace/groups/token) and refuses while any host row exists. The Danger-zone DELETE still (a) leaves host rows and (b) does NOT run that teardown — it purges escrow custody + drops the config only. Decide the model: DELETE requires a prior RESET, or DELETE subsumes RESET's teardown, or they stay orthogonal (RESET = recycle-in-place, DELETE = escrow-purge). | S | idea | Origin: hub v0.61.0 RESET ship. Flips a future "customer fully offboarded (external resources released)" map row. Cheap once the model is chosen |
| R-25 | **Device-node TOCTOU hardening (drive init).** Graduate the controller v0.141.0 Observation: the `format → resolveEnrollUUID(path) → AssignDisk(uuid)` sequence has a narrow /dev-re-enumeration window (agent-guarded on the destructive format via anti-retarget durable-id; benign fs-UUID mount). Bind resolve+assign to the format's durable-id so the mount can't target a moved node. | S | idea | From the v0.141.0 F6 commit's security-review finding (`felhom-controller` REPORT). Low real risk (single-operator, agent-guarded), but cheap to close | | R-25 | **Device-node TOCTOU hardening (drive init).** Graduate the controller v0.141.0 Observation: the `format → resolveEnrollUUID(path) → AssignDisk(uuid)` sequence has a narrow /dev-re-enumeration window (agent-guarded on the destructive format via anti-retarget durable-id; benign fs-UUID mount). Bind resolve+assign to the format's durable-id so the mount can't target a moved node. | S | idea | From the v0.141.0 F6 commit's security-review finding (`felhom-controller` REPORT). Low real risk (single-operator, agent-guarded), but cheap to close |
| R-24 | **Guest RAM resize (live) — SHIPPED (agent v0.90.0 + controller v0.143.0, 2026-07-17).** The customer right-sizes the guest's memory from the controller's Rendszer page; the agent enforces every bound fresh + applies via PVE `SetConfig` (live cgroup, no reboot, Phase-0 proven). **Framing note:** the original hub-desired-state framing is SUPERSEDED by Viktor's controller-direct ruling (2026-07-17) — the resize is controller→agent local-API, never through the hub. Memory only (cores stay observation). Deployed + live-validated on the demo (refusals render Hungarian end-to-end); a successful grow on a normal-sized box folds into the rehearsal. **Cores/live-resize as hub desired-state is NOT built** (deferred, low demand). | M | **SHIPPED** | See felhom-agent + felhom-controller REPORTs; capability-map row "customer right-sizes guest RAM". | | R-24 | **Guest RAM resize (live) — SHIPPED (agent v0.90.0 + controller v0.143.0, 2026-07-17).** The customer right-sizes the guest's memory from the controller's Rendszer page; the agent enforces every bound fresh + applies via PVE `SetConfig` (live cgroup, no reboot, Phase-0 proven). **Framing note:** the original hub-desired-state framing is SUPERSEDED by Viktor's controller-direct ruling (2026-07-17) — the resize is controller→agent local-API, never through the hub. Memory only (cores stay observation). Deployed + live-validated on the demo (refusals render Hungarian end-to-end); a successful grow on a normal-sized box folds into the rehearsal. **Cores/live-resize as hub desired-state is NOT built** (deferred, low demand). | M | **SHIPPED** | See felhom-agent + felhom-controller REPORTs; capability-map row "customer right-sizes guest RAM". |
+53
View File
@@ -1,5 +1,58 @@
# Felhom Hub — Changelog # Felhom Hub — Changelog
## v0.66.0 — Customer self-bind (R-27 slice 1): tokenized capability link + public two-factor `/bind/` page (2026-07-17)
Lets a customer bind their OWN freshly-installed appliance without the operator. Until now every
box booted from the universal secret-free ISO (R-21 slice C) had to be bound by the operator on the
Hosts page; this adds the self-service path. **Viktor's three rulings, each honoured verbatim:**
(a) *"only their own visible"* → the customer proves possession with the **console pairing code**
shown on the box screen — **no appliance list is ever rendered** on any public surface; (b) *first-box
entry* → an **operator-sent 7-day tokenized capability link** over Hungarian email (the claim-engine
delivery pattern, a sibling sender — NOT routed through the claim engine); (c) **lockout after 5
failed attempts** → the token locks and the page says *"call support"*. Wrong code and wrong
passphrase produce **one identical generic failure** (no oracle); an expired link falls back to
operator-bind, unchanged. Does **not** touch the controller or agent. Green: `go build/vet/test`
(full hub suite + 9 new self-bind tests), all 4 red-proofs verified red-then-green, hub confirm gate.
- **Pairing code (Part 1).** `POST /api/v1/appliance/register` now returns an additive `pairing_code`
(6 chars, ambiguity-free alphabet, `ABC-234` display) minted once at first registration and stable
across the idempotent re-register/upsert (backfilled if a pre-existing row had none). Persisted on
`appliance_registrations.pairing_code`; shown in the operator Hosts "Unclaimed appliances" table.
The bootstrap (`felhom-bootstrap.sh`, ISO v1.20.0) parses it and prints a Hungarian **console
banner** to `/dev/console` so the customer can read it off the screen. Old ISOs ignore the field;
an old hub omits it and the banner prints nothing — additive both ways.
- **Capability token (Part 2).** New `selfbind_tokens` table: `sha256(token)` at rest (never the
token), **single-active per customer** (a re-mint deletes the prior row in one tx — an old link dies
the instant a new one is sent), `attempts` locking at 5, one-shot `consumed_at`, `emailed_at`
honesty. Operator **"Send self-bind link"** button on the customer Setup tab
(`POST /customers/{id}/selfbind-link`) mints a 256-bit token and emails
`https://hub.felhom.eu/bind/<token>` (Hungarian, adult tone, no emoji, names both factors + the
7-day + 5-attempt limits). **F1** (no registered email → nothing minted, LOUD flash) and **F2**
(email send fails → the just-minted token is deleted, not left silently live) are both honest.
- **Public bind page (Part 3) — THE TRAP (§9.2).** One new public prefix `/bind/`, exempted from
operator auth AND CSRF at the two gate sites the `/login` exemption occupies, via a **single**
predicate `isPublicBindPath` (matched tightly: trailing slash → no sibling like `/bindsecret`; the
ServeMux `..`-cleans before we see the path → no traversal reach; the handler also rejects a token
containing `/`). `GET` renders form/consumed/locked/expired; `POST` normalizes both inputs, compares
**both factors unconditionally** (constant-time passphrase vs the customer's retrieval passphrase;
the ONE bindable appliance carrying the console code), then decides — identical generic failure
either way. On success: the **same `BindAppliance`** the operator uses, a provenance event with
source `customer_selfbind`, one-shot consume, and *"A doboz kb. egy percen belül folytatja a
telepítést."* The box's ~30 s appliance poll picks up the delivery. Own per-IP rate limiter; the
page is self-contained (it cannot link `/style.css`, which is itself operator-gated). No hub
customer-login/session was built — the URL capability token IS the auth model; a cross-site POST
without both secrets only burns attempts (accepted + documented).
- **Tests + red-proofs (Part 4).** Scenarios AF + F1/F2 (9 tests). The **4 red-proofs** were each
applied and confirmed to turn exactly their scenario red, then reverted green: lockout removed → C1;
oracle introduced → B; `/bind/` prefix widened (drop the slash) → E (and D); single-active DELETE
dropped → C4. The passphrase is never logged/echoed/persisted; only attempt COUNTS are logged
(`self-bind attempt N/5 … token <8hex>…`); the raw link token never enters logs or events.
- **GC verdict (spec §3):** there is **no appliance-staleness GC** in the hub (`applianceStaleAfter`
is a DISPLAY badge only; `pruneAll`/`PurgeExpiredLogBundles` touch reports/log-bundles, not
appliances or selfbind tokens). The 7-day token TTL therefore stands alone and needs no reaper —
single-active-per-customer means at most one row per customer, superseded rows are deleted on
re-mint, and an expired row simply reads as expired (no security or storage pressure).
## v0.65.0 — PBS DR storage visibility (ep0 `usage` op) + Offsite tab split (Restic / PBS DR) + dual dashboard gauges (R-5) (2026-07-17) ## v0.65.0 — PBS DR storage visibility (ep0 `usage` op) + Offsite tab split (Restic / PBS DR) + dual dashboard gauges (R-5) (2026-07-17)
Makes the **PBS DR** storage visible like the restic pool box already is (v0.64.0), the two clearly Makes the **PBS DR** storage visible like the restic pool box already is (v0.64.0), the two clearly
+2 -1
View File
@@ -290,7 +290,8 @@ func main() {
webServer := web.New(dataStore, cfg.Auth.PasswordHash, cfg.API.ReportAPIKey, Version, staleThreshold, logger) webServer := web.New(dataStore, cfg.Auth.PasswordHash, cfg.API.ReportAPIKey, Version, staleThreshold, logger)
webServer.SetTemplateFetcher(templateFetcher) webServer.SetTemplateFetcher(templateFetcher)
webServer.SetAssetManager(assetsMgr) webServer.SetAssetManager(assetsMgr)
webServer.SetClaimEngine(claimEngine) // v0.50.0 — Setup-tab claim chip + resend button webServer.SetClaimEngine(claimEngine) // v0.50.0 — Setup-tab claim chip + resend button
webServer.SetSelfBindMailer(dispatcher) // v0.66.0 (R-27) — customer self-bind link button (sibling of claim mailer)
// Day-0 artifact version dropdowns: let the operator pick a version and have the hub derive the // Day-0 artifact version dropdowns: let the operator pick a version and have the hub derive the
// sha256 from Gitea (no hand-copied checksums). Reuses the registry creds; degrades to manual text // sha256 from Gitea (no hand-copied checksums). Reuses the registry creds; degrades to manual text
// entry when they're absent. // entry when they're absent.
+13 -2
View File
@@ -140,7 +140,15 @@ func (h *Handler) handleApplianceRegister(w http.ResponseWriter, r *http.Request
http.Error(w, "internal error", http.StatusInternalServerError) http.Error(w, "internal error", http.StatusInternalServerError)
return return
} }
isNew, err := h.store.RegisterAppliance(uuid, macSet, sshKeys, hwSummary, sha256hex(token)) // v0.66.0 (R-27): a stable 6-char pairing code the box prints on its console + the customer types
// into the self-bind page. The candidate is used only on first insert; a re-register keeps the code.
candidateCode, err := configgen.RandomPairingCode()
if err != nil {
h.logger.Printf("[ERROR] appliance register: pairing-code mint: %v", err)
http.Error(w, "internal error", http.StatusInternalServerError)
return
}
isNew, pairingCode, err := h.store.RegisterAppliance(uuid, macSet, sshKeys, hwSummary, sha256hex(token), candidateCode)
if err != nil { if err != nil {
h.logger.Printf("[ERROR] appliance register (uuid=%s): %v", uuid, err) h.logger.Printf("[ERROR] appliance register (uuid=%s): %v", uuid, err)
http.Error(w, "internal error", http.StatusInternalServerError) http.Error(w, "internal error", http.StatusInternalServerError)
@@ -153,7 +161,10 @@ func (h *Handler) handleApplianceRegister(w http.ResponseWriter, r *http.Request
} }
w.Header().Set("Content-Type", "application/json") w.Header().Set("Content-Type", "application/json")
w.WriteHeader(http.StatusOK) w.WriteHeader(http.StatusOK)
json.NewEncoder(w).Encode(map[string]any{"appliance_token": token, "poll_interval_sec": 30}) // pairing_code is additive — a pre-v0.66.0 bootstrap ignores it and stays operator-bind-only.
json.NewEncoder(w).Encode(map[string]any{
"appliance_token": token, "poll_interval_sec": 30, "pairing_code": configgen.FormatPairingCode(pairingCode),
})
} }
// handleAppliancePoll — GET /api/v1/appliance/poll (Bearer appliance-token). One-shot delivery: // handleAppliancePoll — GET /api/v1/appliance/poll (Bearer appliance-token). One-shot delivery:
+54
View File
@@ -4,6 +4,8 @@ import (
"crypto/rand" "crypto/rand"
"encoding/hex" "encoding/hex"
"fmt" "fmt"
"math/big"
"strings"
"time" "time"
"gitea.dooplex.hu/admin/felhom-hub/internal/store" "gitea.dooplex.hu/admin/felhom-hub/internal/store"
@@ -126,3 +128,55 @@ func RandomHex(n int) (string, error) {
} }
return hex.EncodeToString(b), nil return hex.EncodeToString(b), nil
} }
// pairingAlphabet excludes visually ambiguous characters (0/O, 1/I/L) so a customer can read the code
// off the box's console banner and type it into the self-bind page without confusion.
const pairingAlphabet = "ABCDEFGHJKMNPQRSTUVWXYZ23456789"
// RandomPairingCode returns a 6-char appliance pairing code from the unambiguous alphabet (rendered
// "ABC-DEF" for display; stored/compared without the dash). It is NOT a secret on its own — the
// self-bind flow also requires the customer's retrieval passphrase.
func RandomPairingCode() (string, error) {
b := make([]byte, 6)
max := big.NewInt(int64(len(pairingAlphabet)))
for i := range b {
idx, err := rand.Int(rand.Reader, max)
if err != nil {
return "", err
}
b[i] = pairingAlphabet[idx.Int64()]
}
return string(b), nil
}
// FormatPairingCode renders a stored 6-char code as "ABC-DEF" for the console banner + the operator UI.
func FormatPairingCode(code string) string {
if len(code) == 6 {
return code[:3] + "-" + code[3:]
}
return code
}
// NormalizePairingCode strips separators/whitespace and upper-cases (the customer may type "abc-def",
// "abc def", or "ABCDEF") so the compare is against a canonical form.
func NormalizePairingCode(s string) string {
var b strings.Builder
for _, r := range strings.ToUpper(s) {
if (r >= 'A' && r <= 'Z') || (r >= '0' && r <= '9') {
b.WriteRune(r)
}
}
return b.String()
}
// NormalizePassphrase canonicalizes a diceware passphrase for the constant-time compare: trim,
// lower-case, and collapse any run of dashes/whitespace to a single dash. The WORDS themselves are
// compared exactly (no accent-folding — the Hungarian wordlist is the source of truth), so a mistyped
// word fails.
func NormalizePassphrase(s string) string {
s = strings.ToLower(strings.TrimSpace(s))
fields := strings.FieldsFunc(s, func(r rune) bool {
return r == '-' || r == ' ' || r == '\t' || r == '\n' || r == '\r'
})
return strings.Join(fields, "-")
}
+21
View File
@@ -245,3 +245,24 @@ func (d *Dispatcher) SendClaimEmail(kind, customerID, email, domain, code string
d.store.LogNotification(customerID, eventType, "info", subject, "sent", "", "customer") d.store.LogNotification(customerID, eventType, "info", subject, "sent", "", "customer")
return nil return nil
} }
// SendSelfBindEmail delivers the customer self-bind capability link (v0.66.0, R-27 slice 1) to the
// REGISTERED customer address. Sibling of SendClaimEmail — NOT routed through the claim engine. The
// link is the capability; it is logged only via the notification_log subject (which carries no
// token), never the raw link. On failure the caller (web) invalidates the just-minted token so it is
// not left silently live.
func (d *Dispatcher) SendSelfBindEmail(customerID, email, link string) error {
if d.resendAPIKey == "" {
d.logger.Printf("[ERROR] self-bind link email for %s NOT sent: no Resend API key configured", customerID)
return fmt.Errorf("notify: no resend api key")
}
subject, body := FormatSelfBindEmail(customerID, link)
if err := d.sendEmailFn(email, subject, body); err != nil {
d.logger.Printf("[ERROR] self-bind link email to customer %s failed: %v", customerID, err)
d.store.LogNotification(customerID, "selfbind_link", "info", subject, "failed", err.Error(), "customer")
return err
}
d.logger.Printf("[INFO] self-bind link emailed to the registered address of %s", customerID)
d.store.LogNotification(customerID, "selfbind_link", "info", subject, "sent", "", "customer")
return nil
}
+29
View File
@@ -230,3 +230,32 @@ Felhom.eu`, code, dashboardURL)
return subject, body return subject, body
} }
} }
// FormatSelfBindEmail builds the customer-facing Hungarian email carrying the self-bind capability
// link (v0.66.0, R-27 slice 1). The link is the ONLY secret here — the passphrase is never in the
// mail (the customer already holds it), and the console pairing code is read off the box screen. The
// copy tells the customer they will need both factors on the page. Adult tone, no emoji.
func FormatSelfBindEmail(customerID, link string) (string, string) {
subject := "[Felhom] Kösd össze a Felhom dobozodat"
body := fmt.Sprintf(`Kedves Ügyfél!
Elkészült a Felhom dobozod, és készen áll az összekötésre. Az alábbi hivatkozáson
tudod te magad összekötni a fiókoddal nincs szükség bejelentkezésre:
%s
A hivatkozás megnyitása után két adatot kell megadnod:
1. A párosító kódot, amely a doboz képernyőjén (a monitoron) látható.
2. A visszaállító jelszavadat (az 5 szóból álló kifejezést), amelyet a
beállításkor kaptál.
A hivatkozás 7 napig érvényes. Biztonsági okból 5 sikertelen próbálkozás után
zárolódik ilyenkor vedd fel a kapcsolatot az ügyfélszolgálattal.
Ha nem te kérted ezt, hagyd figyelmen kívül ezt az e-mailt.
Üdvözlettel,
Felhom.eu`, link)
return subject, body
}
+65 -21
View File
@@ -32,6 +32,7 @@ type ApplianceRegistration struct {
CustomerID string // set at bind CustomerID string // set at bind
InstallMode string // staged at bind (appliance|byo) InstallMode string // staged at bind (appliance|byo)
ExtraArgs string // staged at bind ExtraArgs string // staged at bind
PairingCode string // v0.66.0 (R-27): stable 6-char code shown on the box console + the self-bind page
FirstSeen time.Time FirstSeen time.Time
LastSeen time.Time LastSeen time.Time
BoundAt *time.Time BoundAt *time.Time
@@ -46,57 +47,65 @@ type ApplianceRegistration struct {
// token yet, so it is genuinely starting over; the operator re-binds. isNew is true only on first // token yet, so it is genuinely starting over; the operator re-binds. isNew is true only on first
// insert (so the caller can log the first sighting). The token itself is never passed in — only its // insert (so the caller can log the first sighting). The token itself is never passed in — only its
// hash. // hash.
func (s *Store) RegisterAppliance(uuid, macSet, sshKeys, hwSummary, tokenHash string) (isNew bool, err error) { // v0.66.0 (R-27 slice 1): the caller passes a candidateCode (minted via configgen — store must not
// import configgen, an import cycle); it is used ONLY on the first insert and stays STABLE across the
// idempotent re-register. The effective stored code is returned so the register response can carry it.
func (s *Store) RegisterAppliance(uuid, macSet, sshKeys, hwSummary, tokenHash, candidateCode string) (isNew bool, pairingCode string, err error) {
tx, err := s.db.Begin() tx, err := s.db.Begin()
if err != nil { if err != nil {
return false, err return false, "", err
} }
defer tx.Rollback() defer tx.Rollback()
var id int64 var id int64
var status string var status, existingCode string
row := tx.QueryRow(`SELECT id, status FROM appliance_registrations WHERE uuid = ? AND mac_set = ?`, uuid, macSet) row := tx.QueryRow(`SELECT id, status, COALESCE(pairing_code,'') FROM appliance_registrations WHERE uuid = ? AND mac_set = ?`, uuid, macSet)
switch err := row.Scan(&id, &status); err { switch err := row.Scan(&id, &status, &existingCode); err {
case sql.ErrNoRows: case sql.ErrNoRows:
if _, err := tx.Exec(` if _, err := tx.Exec(`
INSERT INTO appliance_registrations (uuid, mac_set, ssh_host_pubkeys, hw_summary, token_hash, status) INSERT INTO appliance_registrations (uuid, mac_set, ssh_host_pubkeys, hw_summary, token_hash, status, pairing_code)
VALUES (?, ?, ?, ?, ?, 'registered')`, uuid, macSet, sshKeys, hwSummary, tokenHash); err != nil { VALUES (?, ?, ?, ?, ?, 'registered', ?)`, uuid, macSet, sshKeys, hwSummary, tokenHash, candidateCode); err != nil {
return false, fmt.Errorf("register appliance insert: %w", err) return false, "", fmt.Errorf("register appliance insert: %w", err)
} }
if err := tx.Commit(); err != nil { if err := tx.Commit(); err != nil {
return false, err return false, "", err
} }
return true, nil return true, candidateCode, nil
case nil: case nil:
// The code stays STABLE across re-register; a pre-v0.66.0 row (empty code) is backfilled once.
effectiveCode := existingCode
if effectiveCode == "" {
effectiveCode = candidateCode
}
// Existing record: refresh last_seen + identity + the token. Sticky-discard keeps its status; // Existing record: refresh last_seen + identity + the token. Sticky-discard keeps its status;
// everything else resets to registered (a re-registering box is starting over). // everything else resets to registered (a re-registering box is starting over).
if status == ApplianceDiscarded { if status == ApplianceDiscarded {
if _, err := tx.Exec(`UPDATE appliance_registrations if _, err := tx.Exec(`UPDATE appliance_registrations
SET token_hash = ?, ssh_host_pubkeys = ?, hw_summary = ?, last_seen = datetime('now') SET token_hash = ?, ssh_host_pubkeys = ?, hw_summary = ?, pairing_code = ?, last_seen = datetime('now')
WHERE id = ?`, tokenHash, sshKeys, hwSummary, id); err != nil { WHERE id = ?`, tokenHash, sshKeys, hwSummary, effectiveCode, id); err != nil {
return false, fmt.Errorf("register appliance (discarded) update: %w", err) return false, "", fmt.Errorf("register appliance (discarded) update: %w", err)
} }
} else { } else {
if _, err := tx.Exec(`UPDATE appliance_registrations if _, err := tx.Exec(`UPDATE appliance_registrations
SET token_hash = ?, ssh_host_pubkeys = ?, hw_summary = ?, status = 'registered', SET token_hash = ?, ssh_host_pubkeys = ?, hw_summary = ?, pairing_code = ?, status = 'registered',
customer_id = NULL, install_mode = NULL, extra_args = NULL, customer_id = NULL, install_mode = NULL, extra_args = NULL,
bound_at = NULL, delivered_at = NULL, last_seen = datetime('now') bound_at = NULL, delivered_at = NULL, last_seen = datetime('now')
WHERE id = ?`, tokenHash, sshKeys, hwSummary, id); err != nil { WHERE id = ?`, tokenHash, sshKeys, hwSummary, effectiveCode, id); err != nil {
return false, fmt.Errorf("register appliance update: %w", err) return false, "", fmt.Errorf("register appliance update: %w", err)
} }
} }
if err := tx.Commit(); err != nil { if err := tx.Commit(); err != nil {
return false, err return false, "", err
} }
return false, nil return false, effectiveCode, nil
default: default:
return false, fmt.Errorf("register appliance lookup: %w", err) return false, "", fmt.Errorf("register appliance lookup: %w", err)
} }
} }
// scanAppliance scans a full appliance row (column order fixed by applianceCols). // scanAppliance scans a full appliance row (column order fixed by applianceCols).
const applianceCols = `id, uuid, mac_set, ssh_host_pubkeys, hw_summary, status, const applianceCols = `id, uuid, mac_set, ssh_host_pubkeys, hw_summary, status,
COALESCE(customer_id,''), COALESCE(install_mode,''), COALESCE(extra_args,''), COALESCE(customer_id,''), COALESCE(install_mode,''), COALESCE(extra_args,''), COALESCE(pairing_code,''),
first_seen, last_seen, bound_at, delivered_at, discarded_at` first_seen, last_seen, bound_at, delivered_at, discarded_at`
func scanAppliance(sc interface{ Scan(...any) error }) (*ApplianceRegistration, error) { func scanAppliance(sc interface{ Scan(...any) error }) (*ApplianceRegistration, error) {
@@ -104,7 +113,7 @@ func scanAppliance(sc interface{ Scan(...any) error }) (*ApplianceRegistration,
var firstSeen, lastSeen string var firstSeen, lastSeen string
var boundAt, deliveredAt, discardedAt sql.NullString var boundAt, deliveredAt, discardedAt sql.NullString
if err := sc.Scan(&a.ID, &a.UUID, &a.MACSet, &a.SSHHostPubkeys, &a.HWSummary, &a.Status, if err := sc.Scan(&a.ID, &a.UUID, &a.MACSet, &a.SSHHostPubkeys, &a.HWSummary, &a.Status,
&a.CustomerID, &a.InstallMode, &a.ExtraArgs, &a.CustomerID, &a.InstallMode, &a.ExtraArgs, &a.PairingCode,
&firstSeen, &lastSeen, &boundAt, &deliveredAt, &discardedAt); err != nil { &firstSeen, &lastSeen, &boundAt, &deliveredAt, &discardedAt); err != nil {
return nil, err return nil, err
} }
@@ -138,6 +147,41 @@ func (s *Store) ApplianceByToken(tokenHash string) (*ApplianceRegistration, erro
return a, err return a, err
} }
// ApplianceByPairingCode resolves a console pairing code to the ONE bindable (registered, not yet
// bound/delivered/discarded) appliance carrying it — the customer-self-bind factor-1 lookup (v0.66.0,
// R-27). Returns (nil, nil) when zero OR more than one registered appliance matches (an ambiguous
// code binds nothing — the public handler maps that to the same generic failure as a wrong code, no
// oracle). code is the RAW stored form (uppercased, no separator); the caller normalizes first.
func (s *Store) ApplianceByPairingCode(code string) (*ApplianceRegistration, error) {
if code == "" {
return nil, nil
}
rows, err := s.db.Query(`SELECT `+applianceCols+` FROM appliance_registrations
WHERE pairing_code = ? AND status = 'registered'`, code)
if err != nil {
return nil, err
}
defer rows.Close()
var matches []*ApplianceRegistration
for rows.Next() {
a, err := scanAppliance(rows)
if err != nil {
return nil, err
}
matches = append(matches, a)
if len(matches) > 1 {
return nil, nil // ambiguous — bind nothing
}
}
if err := rows.Err(); err != nil {
return nil, err
}
if len(matches) != 1 {
return nil, nil
}
return matches[0], nil
}
// GetAppliance fetches by row id (operator UI actions). // GetAppliance fetches by row id (operator UI actions).
func (s *Store) GetAppliance(id int64) (*ApplianceRegistration, error) { func (s *Store) GetAppliance(id int64) (*ApplianceRegistration, error) {
a, err := scanAppliance(s.db.QueryRow(`SELECT `+applianceCols+` FROM appliance_registrations WHERE id = ?`, id)) a, err := scanAppliance(s.db.QueryRow(`SELECT `+applianceCols+` FROM appliance_registrations WHERE id = ?`, id))
+171
View File
@@ -0,0 +1,171 @@
package store
import (
"database/sql"
"fmt"
"time"
)
// selfbind.go — the customer self-bind capability-token store (v0.66.0, R-27 slice 1).
//
// The operator mints one token per customer and the hub emails a 7-day tokenized capability link.
// The customer opens it (public, no login — the token IS the capability), enters the box's console
// pairing code + their retrieval passphrase, and the hub stages the bind. Only sha256(token) is
// stored here (never the token); attempts lock at 5 (Viktor's ruling: "call support"); consumed_at
// is the one-shot flip. There is NO appliance data on this table — the two-factor check happens at
// bind time in web/selfbind.go, so a leaked token row reveals nothing about any box.
//
// Custody rules honoured here: single-active per customer (a re-mint deletes the prior row, so an old
// link dies the moment a new one is sent), and the token hash is the only representation at rest.
// SelfBindMaxAttempts is the lockout threshold: the 5th failed factor-check locks the token and the
// customer is told to call support. Distinct from the controller's claim lockout — this is the
// hub-side public-surface guard against online guessing of the pairing code + passphrase.
const SelfBindMaxAttempts = 5
// SelfBindToken is a minted capability link's state.
type SelfBindToken struct {
ID int64
CustomerID string
Attempts int
Locked bool
CreatedAt time.Time
ExpiresAt time.Time
EmailedAt *time.Time
ConsumedAt *time.Time
}
// Expired reports whether the link is past its TTL (falls back to operator-bind, unchanged).
func (t *SelfBindToken) Expired(now time.Time) bool { return now.After(t.ExpiresAt) }
// Consumed reports whether the link was already used (one-shot).
func (t *SelfBindToken) Consumed() bool { return t.ConsumedAt != nil }
// Usable reports whether a POST may attempt a factor-check: not locked, not consumed, not expired.
func (t *SelfBindToken) Usable(now time.Time) bool {
return !t.Locked && !t.Consumed() && !t.Expired(now)
}
const selfBindCols = `id, customer_id, attempts, locked, created_at, expires_at, emailed_at, consumed_at`
func scanSelfBindToken(sc interface{ Scan(...any) error }) (*SelfBindToken, error) {
var t SelfBindToken
var locked int
var createdAt, expiresAt string
var emailedAt, consumedAt sql.NullString
if err := sc.Scan(&t.ID, &t.CustomerID, &t.Attempts, &locked, &createdAt, &expiresAt, &emailedAt, &consumedAt); err != nil {
return nil, err
}
t.Locked = locked != 0
t.CreatedAt = parseSQLiteTime(createdAt)
t.ExpiresAt = parseSQLiteTime(expiresAt)
if emailedAt.Valid && emailedAt.String != "" {
e := parseSQLiteTime(emailedAt.String)
t.EmailedAt = &e
}
if consumedAt.Valid && consumedAt.String != "" {
c := parseSQLiteTime(consumedAt.String)
t.ConsumedAt = &c
}
return &t, nil
}
// MintSelfBindToken creates a fresh capability token for the customer, single-active: any prior token
// for the same customer is DELETED in the same transaction, so an earlier link stops working the
// instant a new one is minted (no parallel-link enumeration). tokenHash is sha256(token) — the
// plaintext token exists only inside the email send. The token is valid for ttl from now.
func (s *Store) MintSelfBindToken(customerID, tokenHash string, ttl time.Duration) error {
if customerID == "" || tokenHash == "" {
return fmt.Errorf("selfbind: customer_id and token hash are required")
}
tx, err := s.db.Begin()
if err != nil {
return err
}
defer tx.Rollback()
if _, err := tx.Exec(`DELETE FROM selfbind_tokens WHERE customer_id = ?`, customerID); err != nil {
return err
}
expiresAt := time.Now().Add(ttl).UTC().Format("2006-01-02 15:04:05")
if _, err := tx.Exec(`INSERT INTO selfbind_tokens (customer_id, token_hash, expires_at)
VALUES (?, ?, ?)`, customerID, tokenHash, expiresAt); err != nil {
return err
}
return tx.Commit()
}
// DeleteSelfBindTokens removes every self-bind token for a customer. Used on the email-failure path
// (F2): a minted token whose delivery failed is not left silently live — the plaintext link is gone
// from memory, so the row is unusable dead weight; deleting it keeps the operator page honest (no
// stale "active link" with no delivery). Idempotent.
func (s *Store) DeleteSelfBindTokens(customerID string) error {
_, err := s.db.Exec(`DELETE FROM selfbind_tokens WHERE customer_id = ?`, customerID)
return err
}
// SelfBindTokenByHash resolves sha256(token) to its row, or (nil, nil) when unknown — the public GET
// maps that to the same generic "invalid link" as an expired one: no oracle for "was this ever real".
func (s *Store) SelfBindTokenByHash(tokenHash string) (*SelfBindToken, error) {
if tokenHash == "" {
return nil, nil
}
t, err := scanSelfBindToken(s.db.QueryRow(`SELECT `+selfBindCols+` FROM selfbind_tokens WHERE token_hash = ?`, tokenHash))
if err == sql.ErrNoRows {
return nil, nil
}
return t, err
}
// MarkSelfBindEmailed records that the capability link was delivered (operator "emailed_at" honesty
// on the customer page). Best-effort: a failure to record does not un-send the mail.
func (s *Store) MarkSelfBindEmailed(tokenHash string) error {
_, err := s.db.Exec(`UPDATE selfbind_tokens SET emailed_at = datetime('now') WHERE token_hash = ?`, tokenHash)
return err
}
// RecordSelfBindAttempt increments the failed-attempt counter for a token and locks it at the 5th
// failure. Called on EVERY failed factor-check (wrong code OR wrong passphrase — indistinguishable).
// Returns the post-increment attempt count and whether the token is now locked. A missing/consumed
// token is a no-op that reports locked=true (the caller already refuses those; belt-and-suspenders).
func (s *Store) RecordSelfBindAttempt(tokenHash string) (attempts int, locked bool, err error) {
tx, err := s.db.Begin()
if err != nil {
return 0, false, err
}
defer tx.Rollback()
t, err := scanSelfBindToken(tx.QueryRow(`SELECT `+selfBindCols+` FROM selfbind_tokens WHERE token_hash = ?`, tokenHash))
if err == sql.ErrNoRows {
return SelfBindMaxAttempts, true, nil
}
if err != nil {
return 0, false, err
}
attempts = t.Attempts + 1
locked = attempts >= SelfBindMaxAttempts
lockedInt := 0
if locked {
lockedInt = 1
}
if _, err := tx.Exec(`UPDATE selfbind_tokens SET attempts = ?, locked = ? WHERE id = ?`, attempts, lockedInt, t.ID); err != nil {
return 0, false, err
}
if err := tx.Commit(); err != nil {
return 0, false, err
}
return attempts, locked, nil
}
// ConsumeSelfBindToken flips consumed_at EXACTLY ONCE on a successful bind. ok=true for the single
// winning caller; ok=false for a replay or a lost race (already consumed). The two-factor check has
// already passed when this is called — this is the one-shot gate against a double-bind. It refuses a
// locked or expired token defensively (WHERE guards), though the handler checks Usable() first.
func (s *Store) ConsumeSelfBindToken(tokenHash string) (ok bool, err error) {
res, err := s.db.Exec(`UPDATE selfbind_tokens
SET consumed_at = datetime('now')
WHERE token_hash = ? AND consumed_at IS NULL AND locked = 0 AND expires_at > datetime('now')`, tokenHash)
if err != nil {
return false, err
}
n, _ := res.RowsAffected()
return n == 1, nil
}
+22
View File
@@ -637,15 +637,37 @@ func (s *Store) migrate() error {
bound_at DATETIME, bound_at DATETIME,
delivered_at DATETIME, delivered_at DATETIME,
discarded_at DATETIME, discarded_at DATETIME,
pairing_code TEXT NOT NULL DEFAULT '',
UNIQUE(uuid, mac_set) UNIQUE(uuid, mac_set)
); );
CREATE INDEX IF NOT EXISTS idx_appliance_status ON appliance_registrations(status, last_seen DESC); CREATE INDEX IF NOT EXISTS idx_appliance_status ON appliance_registrations(status, last_seen DESC);
CREATE INDEX IF NOT EXISTS idx_appliance_token ON appliance_registrations(token_hash); CREATE INDEX IF NOT EXISTS idx_appliance_token ON appliance_registrations(token_hash);
-- selfbind_tokens (v0.66.0, R-27 slice 1 customer self-bind): the 7-day tokenized capability
-- link the operator emails. token_hash is sha256 at rest (never the token). Single-active per
-- customer (delete-then-insert on re-mint). attempts lock at 5 (Viktor's ruling); consumed_at is
-- the one-shot flip. NO appliance data here the code+passphrase check happens at bind time.
CREATE TABLE IF NOT EXISTS selfbind_tokens (
id INTEGER PRIMARY KEY AUTOINCREMENT,
customer_id TEXT NOT NULL,
token_hash TEXT NOT NULL UNIQUE,
attempts INTEGER NOT NULL DEFAULT 0,
locked INTEGER NOT NULL DEFAULT 0,
created_at DATETIME NOT NULL DEFAULT (datetime('now')),
expires_at DATETIME NOT NULL,
emailed_at DATETIME,
consumed_at DATETIME
);
CREATE INDEX IF NOT EXISTS idx_selfbind_token ON selfbind_tokens(token_hash);
`) `)
if err != nil { if err != nil {
return err return err
} }
// v0.66.0 (R-27 slice 1): pairing_code on pre-existing appliance rows (idempotent — errors if the
// column already exists, which is fine on a fresh DB where the CREATE above already added it).
s.db.Exec("ALTER TABLE appliance_registrations ADD COLUMN pairing_code TEXT NOT NULL DEFAULT ''")
// v0.51.0 dr_tier one-time legacy backfill — see the ALTER above; runs last so every table // v0.51.0 dr_tier one-time legacy backfill — see the ALTER above; runs last so every table
// it touches (hosts, customer_configs) exists on a fresh DB too (where it finds nothing). // it touches (hosts, customer_configs) exists on a fresh DB too (where it finds nothing).
if drTierAlterErr == nil { if drTierAlterErr == nil {
+6 -3
View File
@@ -10,6 +10,7 @@ import (
"strings" "strings"
"time" "time"
"gitea.dooplex.hu/admin/felhom-hub/internal/configgen"
"gitea.dooplex.hu/admin/felhom-hub/internal/store" "gitea.dooplex.hu/admin/felhom-hub/internal/store"
) )
@@ -28,6 +29,7 @@ type applianceRow struct {
CPU string CPU string
MemGB string MemGB string
SSHFingerprints []string SSHFingerprints []string
PairingCode string // v0.66.0 (R-27): the code the customer reads off the box console + types on /bind
FirstSeen *time.Time FirstSeen *time.Time
LastSeen *time.Time LastSeen *time.Time
Stale bool Stale bool
@@ -61,9 +63,10 @@ func sshFingerprint(line string) string {
// applianceToRow builds the view model (parses hw_summary + computes SSH fingerprints). // applianceToRow builds the view model (parses hw_summary + computes SSH fingerprints).
func applianceToRow(a store.ApplianceRegistration, now time.Time, customerName func(string) string) applianceRow { func applianceToRow(a store.ApplianceRegistration, now time.Time, customerName func(string) string) applianceRow {
row := applianceRow{ row := applianceRow{
ID: a.ID, ID: a.ID,
UUID: a.UUID, UUID: a.UUID,
Bound: a.Status == store.ApplianceBound, PairingCode: configgen.FormatPairingCode(a.PairingCode),
Bound: a.Status == store.ApplianceBound,
} }
if a.MACSet != "" { if a.MACSet != "" {
row.MACs = strings.Split(a.MACSet, ",") row.MACs = strings.Split(a.MACSet, ",")
+1 -1
View File
@@ -17,7 +17,7 @@ func seedAppliance(t *testing.T, st *store.Store, uuid, macSet string) int64 {
// a real ed25519 host key line so the fingerprint helper has something to parse // a real ed25519 host key line so the fingerprint helper has something to parse
sshKey := "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIHVBv+9slP74+1/vNhiI0OJDrXQ2nvb8iwmIxMfUZn36 host" sshKey := "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIHVBv+9slP74+1/vNhiI0OJDrXQ2nvb8iwmIxMfUZn36 host"
hw := `{"product":"Intel N100 mini","cpu":"Intel(R) N100","mem_kb":16150372}` hw := `{"product":"Intel N100 mini","cpu":"Intel(R) N100","mem_kb":16150372}`
if _, err := st.RegisterAppliance(uuid, macSet, sshKey, hw, "hash-"+uuid); err != nil { if _, _, err := st.RegisterAppliance(uuid, macSet, sshKey, hw, "hash-"+uuid, "ABCDEF"); err != nil {
t.Fatalf("register appliance: %v", err) t.Fatalf("register appliance: %v", err)
} }
list, err := st.ListUnclaimedAppliances() list, err := st.ListUnclaimedAppliances()
+284
View File
@@ -0,0 +1,284 @@
package web
import (
"crypto/subtle"
"html/template"
"net/http"
"strings"
"sync"
"time"
"gitea.dooplex.hu/admin/felhom-hub/internal/configgen"
"gitea.dooplex.hu/admin/felhom-hub/internal/store"
)
// selfbind.go — the CUSTOMER side of self-bind (v0.66.0, R-27 slice 1): the PUBLIC /bind/<token> page.
// A logged-out customer opens the emailed capability link and binds their own freshly-installed
// appliance by proving TWO factors — the console pairing code (physical possession of the box) and
// their retrieval passphrase (customer identity). No hub login exists; the URL token IS the auth.
//
// THE TRAP (spec §9.2): /bind/ is the ONE new public prefix, exempted from operator auth + CSRF at
// the two gate sites the /login exemption occupies. isPublicBindPath is the SINGLE definition of that
// prefix — both gate sites and the route dispatch call it, so widening it is one visible change (and
// the red-proof targets exactly this function). It is matched TIGHTLY (trailing slash → no sibling
// prefix like /bindsecret; the ServeMux cleans .. before we see the path → no traversal reach).
//
// No-oracle rules on this surface: the page NEVER renders/enumerates any appliance data; a wrong code
// and a wrong passphrase produce ONE identical generic failure; both factors are compared
// unconditionally before the decision; and only attempt COUNTS are logged (never the secrets, never
// the raw token — a hash prefix at most).
// isPublicBindPath reports whether a path is the public customer self-bind surface. THE single
// definition of the /bind/ public prefix (THE TRAP §9.2) — do not inline a second copy anywhere.
func isPublicBindPath(path string) bool {
return strings.HasPrefix(path, "/bind/")
}
// --- per-IP rate limiter (web-package sibling of api.ipRateLimiter; the type there is unexported) ---
type bindBucket struct {
tokens float64
last time.Time
}
type bindRateLimiter struct {
mu sync.Mutex
perMinute float64
buckets map[string]*bindBucket
now func() time.Time
}
func newBindRateLimiter(perMinute int) *bindRateLimiter {
if perMinute <= 0 {
perMinute = 30
}
return &bindRateLimiter{perMinute: float64(perMinute), buckets: make(map[string]*bindBucket), now: time.Now}
}
func (rl *bindRateLimiter) allow(ip string) bool {
rl.mu.Lock()
defer rl.mu.Unlock()
now := rl.now()
b, ok := rl.buckets[ip]
if !ok {
rl.buckets[ip] = &bindBucket{tokens: rl.perMinute - 1, last: now}
return true
}
b.tokens += now.Sub(b.last).Seconds() * (rl.perMinute / 60.0)
if b.tokens > rl.perMinute {
b.tokens = rl.perMinute
}
b.last = now
if b.tokens < 1 {
return false
}
b.tokens--
return true
}
// bindClientIP extracts the client IP behind the ingress (first XFF hop, else RemoteAddr) — buckets
// only; the ingress geo-gate is the real access control.
func bindClientIP(r *http.Request) string {
if xff := r.Header.Get("X-Forwarded-For"); xff != "" {
if i := strings.IndexByte(xff, ','); i > 0 {
return strings.TrimSpace(xff[:i])
}
return strings.TrimSpace(xff)
}
if i := strings.LastIndexByte(r.RemoteAddr, ':'); i > 0 {
return r.RemoteAddr[:i]
}
return r.RemoteAddr
}
// --- the page ---
type bindPageData struct {
State string // "form" | "success" | "expired" | "consumed" | "locked"
Token string // echoed into the form action (the capability itself; already in the URL)
Failed bool // generic factor-check failure (form state only)
}
var bindTemplate = template.Must(template.New("bind").Parse(bindPageHTML))
func (s *Server) renderBind(w http.ResponseWriter, status int, data bindPageData) {
w.Header().Set("Content-Type", "text/html; charset=utf-8")
w.WriteHeader(status)
if err := bindTemplate.Execute(w, data); err != nil {
s.logger.Printf("[ERROR] rendering /bind page: %v", err)
}
}
// handleBind serves the public self-bind page. GET renders the current link state; POST validates
// both factors and, on success, stages the bind (same BindAppliance the operator uses; provenance =
// customer self-bind). Reached only via isPublicBindPath — auth + CSRF exempt at the gate sites.
func (s *Server) handleBind(w http.ResponseWriter, r *http.Request) {
if s.bindLimiter != nil && !s.bindLimiter.allow(bindClientIP(r)) {
s.renderBind(w, http.StatusTooManyRequests, bindPageData{State: "expired"})
return
}
token := strings.TrimPrefix(r.URL.Path, "/bind/")
// A trailing segment only — reject anything with further path structure (defence in depth atop
// the ServeMux path-clean; the token is a flat hex string).
if token == "" || strings.Contains(token, "/") {
s.renderBind(w, http.StatusNotFound, bindPageData{State: "expired"})
return
}
hash := selfBindHash(token)
tok, err := s.store.SelfBindTokenByHash(hash)
if err != nil {
s.logger.Printf("[ERROR] /bind lookup failed: %v", err)
http.Error(w, "Internal error", http.StatusInternalServerError)
return
}
now := time.Now()
// Terminal link states — identical for GET and POST, no factor check attempted. An unknown token
// (nil) is folded into "expired": no oracle for "was this link ever real".
switch {
case tok == nil || tok.Expired(now):
s.renderBind(w, http.StatusOK, bindPageData{State: "expired"})
return
case tok.Consumed():
s.renderBind(w, http.StatusOK, bindPageData{State: "consumed"})
return
case tok.Locked:
s.renderBind(w, http.StatusOK, bindPageData{State: "locked"})
return
}
if r.Method != http.MethodPost {
s.renderBind(w, http.StatusOK, bindPageData{State: "form", Token: token})
return
}
// --- POST: validate BOTH factors unconditionally, then decide (no oracle) ---
normCode := configgen.NormalizePairingCode(r.FormValue("pairing_code"))
normPass := configgen.NormalizePassphrase(r.FormValue("passphrase"))
// Factor 2 (passphrase) — the customer's retrieval passphrase, constant-time compared. Computed
// even when the customer/appliance is absent so the two paths are indistinguishable by timing.
var storedPass string
if cc, cerr := s.store.GetCustomerConfig(tok.CustomerID); cerr == nil && cc != nil {
storedPass = configgen.NormalizePassphrase(cc.RetrievalPassword)
}
passOK := storedPass != "" && subtle.ConstantTimeCompare([]byte(normPass), []byte(storedPass)) == 1
// Factor 1 (pairing code) — the ONE bindable appliance carrying that console code.
appliance, aerr := s.store.ApplianceByPairingCode(normCode)
if aerr != nil {
s.logger.Printf("[ERROR] /bind appliance lookup failed: %v", aerr)
http.Error(w, "Internal error", http.StatusInternalServerError)
return
}
codeOK := appliance != nil
if !codeOK || !passOK {
attempts, locked, rerr := s.store.RecordSelfBindAttempt(hash)
if rerr != nil {
s.logger.Printf("[ERROR] /bind recording attempt: %v", rerr)
}
// COUNTS only — never which factor failed, never the secrets, never the raw token.
s.logger.Printf("[WARN] self-bind attempt %d/%d failed for token %s… (customer %s)", attempts, store.SelfBindMaxAttempts, hash[:8], tok.CustomerID)
if locked {
s.renderBind(w, http.StatusOK, bindPageData{State: "locked"})
return
}
s.renderBind(w, http.StatusOK, bindPageData{State: "form", Token: token, Failed: true})
return
}
// Both factors passed. Consume one-shot FIRST (atomic gate against a double-bind race).
consumed, cerr := s.store.ConsumeSelfBindToken(hash)
if cerr != nil {
s.logger.Printf("[ERROR] /bind consuming token: %v", cerr)
http.Error(w, "Internal error", http.StatusInternalServerError)
return
}
if !consumed {
// Lost the race (a concurrent request consumed it) — it is already being bound.
s.renderBind(w, http.StatusOK, bindPageData{State: "consumed"})
return
}
if err := s.store.BindAppliance(appliance.ID, tok.CustomerID, "appliance", ""); err != nil {
// Rare: the appliance became unbindable (operator discarded it) between lookup and bind. The
// token is spent; surface a neutral generic failure rather than an appliance-state oracle.
s.logger.Printf("[WARN] self-bind: BindAppliance %d → %s failed after factor match: %v", appliance.ID, tok.CustomerID, err)
s.renderBind(w, http.StatusOK, bindPageData{State: "form", Failed: true})
return
}
if _, err := s.store.SaveEvent(tok.CustomerID, "appliance_bound", "info",
"Az ügyfél saját maga kötötte össze az új eszközt (bare-metal telepítés); a hozzáférést a doboz a következő lekérdezéskor megkapja.", "", "customer_selfbind"); err != nil {
s.logger.Printf("[WARN] self-bind: save event for %s: %v", tok.CustomerID, err)
}
s.logger.Printf("[INFO] self-bind SUCCESS: appliance %d bound to customer %s by customer self-service (token %s…)", appliance.ID, tok.CustomerID, hash[:8])
s.renderBind(w, http.StatusOK, bindPageData{State: "success"})
}
// bindPageHTML is the self-contained public page. It CANNOT link /style.css (that route is
// operator-auth gated), so all styling is inline — mirroring the login page. Design tokens: navy
// surface, 2px radius, hairline rules, exception color for the failure banner. Hungarian, adult tone,
// no emoji. It renders NO appliance data in any state.
const bindPageHTML = `<!DOCTYPE html>
<html lang="hu">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta name="robots" content="noindex, nofollow">
<title>Felhom Doboz összekötése</title>
<style>
:root { --navy:#0A2540; --ink:#0A2540; --muted:#5b6b7d; --line:#d8e0e8; --brand:#0083D8; --exc:#c0392b; --bg:#f4f7fa; }
* { box-sizing: border-box; }
body { margin:0; font-family:-apple-system,BlinkMacSystemFont,"Segoe UI",Roboto,sans-serif; background:var(--bg); color:var(--ink); line-height:1.5; }
.wrap { max-width:460px; margin:3rem auto; padding:0 1rem; }
.card { background:#fff; border:1px solid var(--line); border-radius:2px; padding:1.75rem; }
h1 { font-size:1.25rem; margin:0 0 .25rem; color:var(--navy); }
h1 span { color:var(--brand); }
.lead { color:var(--muted); font-size:.92rem; margin:.25rem 0 1.25rem; }
label { display:block; font-weight:600; font-size:.9rem; margin:1rem 0 .35rem; }
.hint { color:var(--muted); font-size:.8rem; margin:.15rem 0 0; }
input[type=text] { width:100%; padding:.6rem .7rem; border:1px solid var(--line); border-radius:2px; font-size:1rem; }
input.code { text-transform:uppercase; letter-spacing:.12em; font-family:ui-monospace,SFMono-Regular,Menlo,monospace; }
button { margin-top:1.5rem; width:100%; padding:.7rem; background:var(--brand); color:#fff; border:none; border-radius:2px; font-size:1rem; cursor:pointer; }
button:hover { background:#006cb0; }
.banner { border:1px solid var(--exc); color:var(--exc); background:#fbeae8; border-radius:2px; padding:.6rem .75rem; font-size:.88rem; margin:0 0 1rem; }
.note { border-top:1px solid var(--line); margin-top:1.5rem; padding-top:1rem; color:var(--muted); font-size:.82rem; }
.foot { text-align:center; color:var(--muted); font-size:.75rem; margin-top:1.25rem; }
.ok { color:#1e7e34; }
</style>
</head>
<body>
<div class="wrap">
<div class="card">
<h1>Felhom <span>doboz</span> összekötése</h1>
{{if eq .State "form"}}
<p class="lead">Kösd össze a most telepített Felhom dobozodat a fiókoddal. Add meg a doboz képernyőjén látható párosító kódot és a visszaállító jelszavadat.</p>
{{if .Failed}}<div class="banner">A megadott adatok nem megfelelőek. Ellenőrizd a párosító kódot és a jelszót, majd próbáld újra.</div>{{end}}
<form method="POST" action="/bind/{{.Token}}">
<label for="pairing_code">Párosító kód</label>
<input class="code" type="text" id="pairing_code" name="pairing_code" autocomplete="off" autocapitalize="characters" spellcheck="false" required autofocus placeholder="ABC-234">
<p class="hint">A doboz monitorán jelenik meg, a telepítés után.</p>
<label for="passphrase">Visszaállító jelszó</label>
<input type="text" id="passphrase" name="passphrase" autocomplete="off" spellcheck="false" required placeholder="öt szó, kötőjellel vagy szóközzel">
<p class="hint">Az öt szóból álló kifejezés, amelyet a beállításkor kaptál.</p>
<button type="submit">Összekötés</button>
</form>
<p class="note">Biztonsági okból 5 sikertelen próbálkozás után a hivatkozás zárolódik. Ilyenkor vedd fel a kapcsolatot az ügyfélszolgálattal.</p>
{{else if eq .State "success"}}
<p class="lead ok">Sikeres összekötés.</p>
<p>A doboz kb. egy percen belül folytatja a telepítést. Ezt az oldalt bezárhatod a beállítás a háttérben befejeződik, és a vezérlőpultod hamarosan elérhető lesz.</p>
{{else if eq .State "consumed"}}
<p class="lead">Ez a hivatkozás már fel lett használva.</p>
<p>A doboz összekötése megtörtént. Ha úgy gondolod, hogy ez tévedés, vedd fel a kapcsolatot az ügyfélszolgálattal.</p>
{{else if eq .State "locked"}}
<p class="lead">Ez a hivatkozás zárolva van.</p>
<p>Túl sok sikertelen próbálkozás történt. Biztonsági okból a hivatkozás zárolódott kérjük, vedd fel a kapcsolatot az ügyfélszolgálattal a doboz összekötéséhez.</p>
{{else}}
<p class="lead">Ez a hivatkozás érvénytelen vagy lejárt.</p>
<p>A hivatkozás 7 napig érvényes. Ha lejárt, kérj újat az ügyfélszolgálattól, vagy az összekötést az üzemeltető is elvégezheti.</p>
{{end}}
</div>
<p class="foot">Felhom.eu</p>
</div>
</body>
</html>`
+94
View File
@@ -0,0 +1,94 @@
package web
import (
"crypto/sha256"
"encoding/hex"
"net/http"
"time"
"gitea.dooplex.hu/admin/felhom-hub/internal/configgen"
)
// selfbind_mint.go — the OPERATOR side of customer self-bind (v0.66.0, R-27 slice 1): the "Send
// self-bind link" button on the customer page mints a 7-day capability token and emails the customer
// a public bind link. The customer side (the public /bind/ page) lives in selfbind.go. No hub
// customer-login exists — the emailed link IS the auth model.
// selfBindTTL is the capability link's lifetime. After it, self-bind falls back to operator-bind
// unchanged (the token simply reads as expired; nothing else regresses).
const selfBindTTL = 7 * 24 * time.Hour
// selfBindBaseURL is the hub's public origin, matching the hardcoded origin used elsewhere
// (configgen). The link is https://hub.felhom.eu/bind/<token>.
const selfBindBaseURL = "https://hub.felhom.eu"
// SelfBindMailer delivers the self-bind capability link to the registered customer address. The
// notify.Dispatcher implements it (a sibling of the claim mailer — NOT routed through claim). The
// signature takes plain strings so the implementation needs no import of this package.
type SelfBindMailer interface {
SendSelfBindEmail(customerID, email, link string) error
}
// SetSelfBindMailer wires the self-bind link sender (v0.66.0). Absent → the button returns 502.
func (s *Server) SetSelfBindMailer(m SelfBindMailer) { s.selfBindMailer = m }
func selfBindHash(token string) string {
sum := sha256.Sum256([]byte(token))
return hex.EncodeToString(sum[:])
}
// handleSelfBindLinkSend — POST /customers/{id}/selfbind-link. Mints a single-active capability token
// for the customer and emails the public bind link. Honesty rules:
// - F1: no registered email → nothing is minted, LOUD flash (a link no one can receive is useless).
// - F2: email send fails → the just-minted token is deleted (not left silently live), LOUD flash.
//
// The plaintext token exists only between minting and the send; it is never logged (only an 8-char
// hash prefix) and never persisted (only its sha256).
func (s *Server) handleSelfBindLinkSend(w http.ResponseWriter, r *http.Request, customerID string) {
if s.selfBindMailer == nil {
http.Error(w, "Self-bind mailer is not configured on this hub", http.StatusBadGateway)
return
}
cfg, err := s.store.GetCustomerConfig(customerID)
if err != nil || cfg == nil {
http.NotFound(w, r)
return
}
// F1: refuse to mint a link that cannot be delivered.
if cfg.Email == "" {
s.logger.Printf("[WARN] self-bind link for %s NOT sent: customer has no registered email", customerID)
http.Redirect(w, r, "/customers/"+customerID+"?flash=selfbind-no-email#tab=setup", http.StatusSeeOther)
return
}
token, err := configgen.RandomHex(32) // 256-bit capability token
if err != nil {
s.logger.Printf("[ERROR] self-bind link for %s: token generation: %v", customerID, err)
http.Error(w, "Internal error", http.StatusInternalServerError)
return
}
hash := selfBindHash(token)
if err := s.store.MintSelfBindToken(customerID, hash, selfBindTTL); err != nil {
s.logger.Printf("[ERROR] self-bind link for %s: minting token: %v", customerID, err)
http.Error(w, "Internal error", http.StatusInternalServerError)
return
}
link := selfBindBaseURL + "/bind/" + token
if err := s.selfBindMailer.SendSelfBindEmail(customerID, cfg.Email, link); err != nil {
// F2: delivery failed — do not leave a live capability token behind (the plaintext link is
// already gone from memory, so nobody could ever use it; delete it and surface the failure).
if derr := s.store.DeleteSelfBindTokens(customerID); derr != nil {
s.logger.Printf("[ERROR] self-bind link for %s: send failed AND cleanup failed: send=%v cleanup=%v", customerID, err, derr)
} else {
s.logger.Printf("[ERROR] self-bind link for %s: email send failed, token invalidated (hash %s…): %v", customerID, hash[:8], err)
}
http.Redirect(w, r, "/customers/"+customerID+"?flash=selfbind-send-failed#tab=setup", http.StatusSeeOther)
return
}
if err := s.store.MarkSelfBindEmailed(hash); err != nil {
s.logger.Printf("[WARN] self-bind link sent to %s but emailed_at not recorded: %v", customerID, err)
}
s.logger.Printf("[INFO] self-bind link (hash %s…, valid 7 days) emailed to the registered address of %s", hash[:8], customerID)
http.Redirect(w, r, "/customers/"+customerID+"?flash=selfbind-sent#tab=setup", http.StatusSeeOther)
}
+325
View File
@@ -0,0 +1,325 @@
package web
import (
"net/http"
"net/http/httptest"
"net/url"
"strconv"
"strings"
"testing"
"time"
"golang.org/x/crypto/bcrypt"
"gitea.dooplex.hu/admin/felhom-hub/internal/store"
)
// selfbind_test.go — customer self-bind (v0.66.0, R-27 slice 1), Scenarios AF.
//
// Each red-proof below names the ONE line to break to turn a scenario red — the guard that the test
// actually pins. If a red-proof does NOT turn its scenario red, the test is hollow.
const (
testPass = "alpha beta gamma delta epsilon" // the customer retrieval passphrase (5 words)
testCode = "ABC234" // an appliance console pairing code (raw stored form)
testCodeFmt = "abc-234" // as a human might type it (lowercased, separated)
)
// selfBindSetup seeds a customer (with passphrase + email) and one registered appliance carrying the
// given console pairing code, and returns the appliance id. Distinct customers must use distinct
// codes (a code shared by two registered appliances is ambiguous → binds nothing).
func selfBindSetup(t *testing.T, st *store.Store, customerID, code string) int64 {
t.Helper()
if err := st.SaveCustomerConfig(&store.CustomerConfig{
CustomerID: customerID, CustomerName: customerID, APIKey: "k",
RetrievalPassword: testPass, Email: customerID + "@example.test",
}); err != nil {
t.Fatal(err)
}
sshKey := "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIHVBv+9slP74+1/vNhiI0OJDrXQ2nvb8iwmIxMfUZn36 host"
if _, _, err := st.RegisterAppliance("uuid-"+customerID, "bc:24:11:98:10:0e", sshKey, `{"product":"N100"}`, "aphash-"+customerID, code); err != nil {
t.Fatal(err)
}
list, _ := st.ListUnclaimedAppliances()
for _, a := range list {
if a.UUID == "uuid-"+customerID {
return a.ID
}
}
t.Fatal("seeded appliance not found")
return 0
}
// mintLink mints a self-bind token for the customer and returns the plaintext token (URL segment).
// Each call uses a fresh nonce so distinct calls yield distinct tokens (single-active still applies —
// the store deletes the prior row for the customer on each mint).
var mintNonce int
func mintLink(t *testing.T, st *store.Store, customerID string, ttl time.Duration) string {
t.Helper()
mintNonce++
token := "tok-" + customerID + "-" + string(rune('a'+mintNonce%26)) + strconv.Itoa(mintNonce)
if err := st.MintSelfBindToken(customerID, selfBindHash(token), ttl); err != nil {
t.Fatal(err)
}
return token
}
func bindGET(t *testing.T, s *Server, token string) *httptest.ResponseRecorder {
t.Helper()
rr := httptest.NewRecorder()
s.handleBind(rr, httptest.NewRequest("GET", "/bind/"+token, nil))
return rr
}
func bindPOST(t *testing.T, s *Server, token, code, pass string) *httptest.ResponseRecorder {
t.Helper()
form := url.Values{"pairing_code": {code}, "passphrase": {pass}}
req := httptest.NewRequest("POST", "/bind/"+token, strings.NewReader(form.Encode()))
req.Header.Set("Content-Type", "application/x-www-form-urlencoded")
rr := httptest.NewRecorder()
s.handleBind(rr, req)
return rr
}
// --- Scenario A: happy path (GET renders form; POST with both correct factors stages the bind) ---
func TestSelfBind_A_HappyPath(t *testing.T) {
s, st := newTestServer(t)
id := selfBindSetup(t, st, "acme", testCode)
token := mintLink(t, st, "acme", selfBindTTL)
if body := bindGET(t, s, token).Body.String(); !strings.Contains(body, "Párosító kód") || !strings.Contains(body, "action=\"/bind/"+token+"\"") {
t.Fatalf("GET did not render the entry form")
}
// A human types the code lowercased + separated and the passphrase with odd spacing — normalization
// must accept both.
rr := bindPOST(t, s, token, testCodeFmt, " Alpha Beta gamma-delta epsilon ")
if rr.Code != http.StatusOK || !strings.Contains(rr.Body.String(), "egy percen belül") {
t.Fatalf("POST success page not rendered: code=%d body=%q", rr.Code, rr.Body.String())
}
// The appliance is bound to the customer (same effect as an operator bind).
if a, _ := st.GetAppliance(id); a == nil || a.Status != store.ApplianceBound || a.CustomerID != "acme" {
t.Fatalf("appliance not bound: %+v", a)
}
// Provenance event is customer self-bind, not operator/hub.
ev, _ := st.GetLatestEventByType("acme", "appliance_bound")
if ev == nil || ev.Source != "customer_selfbind" {
t.Fatalf("expected customer_selfbind provenance event, got %+v", ev)
}
// One-shot: the token is now consumed; a second open shows the consumed state.
if body := bindGET(t, s, token).Body.String(); !strings.Contains(body, "már fel lett használva") {
t.Fatalf("token not consumed after success")
}
// Red-proof: drop the ConsumeSelfBindToken call (or make BindAppliance the only effect) → the
// token stays live and this consumed-state assertion goes red.
}
// --- Scenario B: NO ORACLE — wrong code and wrong passphrase yield the SAME generic failure ---
func TestSelfBind_B_NoOracle(t *testing.T) {
s, st := newTestServer(t)
// Two customers with DISTINCT pairing codes so the single-active mint does not cross-invalidate,
// and each has one fresh (attempt-count 1) token — the only difference between the two failure
// pages is then the token in the form action, which we normalize out.
selfBindSetup(t, st, "acme", testCode)
selfBindSetup(t, st, "acme2", "XYZ789")
t1 := mintLink(t, st, "acme", selfBindTTL) // wrong-code attempt (right passphrase)
t2 := mintLink(t, st, "acme2", selfBindTTL) // wrong-passphrase attempt (right code)
wrongCode := strings.ReplaceAll(bindPOST(t, s, t1, "ZZZ999", testPass).Body.String(), t1, "TOKEN")
wrongPass := strings.ReplaceAll(bindPOST(t, s, t2, "xyz-789", "wrong words here now").Body.String(), t2, "TOKEN")
if wrongCode != wrongPass {
t.Fatalf("failure pages differ between wrong-code and wrong-passphrase — that is an oracle")
}
// The generic failure must not leak any appliance data.
if !strings.Contains(wrongCode, "nem megfelelőek") {
t.Fatalf("failure page missing the generic banner: %q", wrongCode)
}
if strings.Contains(wrongCode, "uuid-") || strings.Contains(wrongCode, "N100") {
t.Fatalf("failure page leaked appliance data")
}
// Red-proof: give wrong-code and wrong-passphrase distinct messages/states (an oracle) → the
// wrongCode == wrongPass assertion goes red.
}
// --- Scenario C1: LOCKOUT after 5 failed attempts; a subsequent CORRECT attempt cannot bind ---
func TestSelfBind_C1_Lockout(t *testing.T) {
s, st := newTestServer(t)
id := selfBindSetup(t, st, "acme", testCode)
token := mintLink(t, st, "acme", selfBindTTL)
for i := 1; i <= store.SelfBindMaxAttempts; i++ {
rr := bindPOST(t, s, token, "ZZZ999", "definitely wrong words indeed")
if i < store.SelfBindMaxAttempts && !strings.Contains(rr.Body.String(), "nem megfelelőek") {
t.Fatalf("attempt %d should re-render the form with a failure, got %q", i, rr.Body.String())
}
}
// The 5th failure locked it: even the CORRECT secrets now bind nothing.
rr := bindPOST(t, s, token, testCodeFmt, testPass)
if !strings.Contains(rr.Body.String(), "zárolva") {
t.Fatalf("token not locked after %d failures: %q", store.SelfBindMaxAttempts, rr.Body.String())
}
if a, _ := st.GetAppliance(id); a.Status != store.ApplianceRegistered {
t.Fatalf("a locked link still bound the appliance: %+v", a)
}
// Red-proof: remove the `locked = attempts >= SelfBindMaxAttempts` lock (never lock) → the correct
// post-lockout POST binds and both the "zárolva" and still-registered assertions go red.
}
// --- Scenario C4: SINGLE-ACTIVE per customer — re-minting kills the prior link ---
func TestSelfBind_C4_SingleActive(t *testing.T) {
s, st := newTestServer(t)
selfBindSetup(t, st, "acme", testCode)
first := mintLink(t, st, "acme", selfBindTTL)
second := mintLink(t, st, "acme", selfBindTTL) // re-mint for the SAME customer
if body := bindGET(t, s, first).Body.String(); !strings.Contains(body, "érvénytelen vagy lejárt") {
t.Fatalf("the first link still resolves after a re-mint (not single-active): %q", body)
}
if body := bindGET(t, s, second).Body.String(); !strings.Contains(body, "Párosító kód") {
t.Fatalf("the freshly-minted link does not render the form: %q", body)
}
// Red-proof: drop the `DELETE FROM selfbind_tokens WHERE customer_id` in MintSelfBindToken → the
// first link still resolves and the érvénytelen assertion goes red.
}
// --- Scenario D: the operator auth gate is intact — self-bind's exemption did NOT open other routes ---
func TestSelfBind_D_AuthGateIntact(t *testing.T) {
s, st := newAuthServer(t)
selfBindSetup(t, st, "acme", testCode)
h := s.RequireAuth(http.HandlerFunc(s.ServeHTTP))
for _, path := range []string{"/", "/hosts", "/customers/acme", "/configuration", "/offsite"} {
rr := httptest.NewRecorder()
h.ServeHTTP(rr, httptest.NewRequest("GET", path, nil))
if rr.Code != http.StatusFound || rr.Header().Get("Location") != "/login" {
t.Fatalf("gated route %s not redirected to /login: code=%d loc=%q", path, rr.Code, rr.Header().Get("Location"))
}
}
// Red-proof: this is the companion to Scenario E — widening isPublicBindPath turns THIS red too.
}
// --- Scenario E: THE TRAP — /bind/ is exempt from operator auth, matched TIGHTLY (no leak/traversal) ---
func TestSelfBind_E_TheTrap(t *testing.T) {
s, st := newAuthServer(t)
selfBindSetup(t, st, "acme", testCode)
token := mintLink(t, st, "acme", selfBindTTL)
h := s.RequireAuth(http.HandlerFunc(s.ServeHTTP))
// The public bind link renders WITHOUT a login (the exemption works).
rr := httptest.NewRecorder()
h.ServeHTTP(rr, httptest.NewRequest("GET", "/bind/"+token, nil))
if rr.Code != http.StatusOK || !strings.Contains(rr.Body.String(), "Párosító kód") {
t.Fatalf("public /bind/ link was gated or not rendered: code=%d", rr.Code)
}
// A sibling prefix must NOT be exempt: /bindsecret is still gated (tight trailing-slash match).
rr = httptest.NewRecorder()
h.ServeHTTP(rr, httptest.NewRequest("GET", "/bindsecret", nil))
if rr.Code != http.StatusFound {
t.Fatalf("/bindsecret leaked through the exemption (prefix not tight): code=%d", rr.Code)
}
// Traversal through the exempt prefix must not reach a gated handler: it stays inside handleBind,
// which rejects a token containing '/', never touching /hosts.
rr = httptest.NewRecorder()
h.ServeHTTP(rr, httptest.NewRequest("GET", "/bind/../hosts", nil))
if strings.Contains(rr.Body.String(), "Unclaimed appliances") || strings.Contains(rr.Body.String(), "No hosts enrolled") {
t.Fatalf("path traversal through /bind/ reached the hosts page")
}
// Red-proof: widen isPublicBindPath to strings.HasPrefix(path, "/bind") (drop the slash) → the
// /bindsecret gated assertion goes red; broaden it further and Scenario D goes red too.
}
// --- Scenario F: EXPIRY falls back — an expired link binds nothing, even with correct factors ---
func TestSelfBind_F_ExpiryFallsBack(t *testing.T) {
s, st := newTestServer(t)
id := selfBindSetup(t, st, "acme", testCode)
token := mintLink(t, st, "acme", -1*time.Hour) // already expired
if body := bindGET(t, s, token).Body.String(); !strings.Contains(body, "érvénytelen vagy lejárt") {
t.Fatalf("expired link did not render the expired state")
}
// Even the CORRECT secrets on an expired link bind nothing (operator-bind fallback is unchanged).
bindPOST(t, s, token, testCodeFmt, testPass)
if a, _ := st.GetAppliance(id); a.Status != store.ApplianceRegistered {
t.Fatalf("an expired link still bound the appliance: %+v", a)
}
// Red-proof: drop the `expires_at > datetime('now')` guard in ConsumeSelfBindToken AND the
// tok.Expired() gate → the expired POST binds and the still-registered assertion goes red.
}
// --- Scenario (F1/F2): the operator MINT honesty paths ---
// F1: a customer with no registered email → nothing minted, LOUD flash.
func TestSelfBind_MintNoEmail(t *testing.T) {
s, st := newTestServer(t)
if err := st.SaveCustomerConfig(&store.CustomerConfig{CustomerID: "noemail", APIKey: "k", RetrievalPassword: testPass}); err != nil {
t.Fatal(err)
}
s.SetSelfBindMailer(&stubMailer{})
rr := httptest.NewRecorder()
req := httptest.NewRequest("POST", "/customers/noemail/selfbind-link", nil)
s.handleSelfBindLinkSend(rr, req, "noemail")
if rr.Code != http.StatusSeeOther || !strings.Contains(rr.Header().Get("Location"), "selfbind-no-email") {
t.Fatalf("no-email mint should redirect with selfbind-no-email: code=%d loc=%q", rr.Code, rr.Header().Get("Location"))
}
if tok, _ := st.SelfBindTokenByHash(selfBindHash("x")); tok != nil {
t.Fatal("a token was minted despite no email")
}
}
// F2: the email send fails → the just-minted token is deleted (not left silently live).
func TestSelfBind_MintSendFailsCleansUp(t *testing.T) {
s, st := newTestServer(t)
selfBindSetup(t, st, "acme", testCode)
stub := &stubMailer{fail: true}
s.SetSelfBindMailer(stub)
rr := httptest.NewRecorder()
s.handleSelfBindLinkSend(rr, httptest.NewRequest("POST", "/customers/acme/selfbind-link", nil), "acme")
if !strings.Contains(rr.Header().Get("Location"), "selfbind-send-failed") {
t.Fatalf("send failure should redirect with selfbind-send-failed: %q", rr.Header().Get("Location"))
}
// The token that was minted for the (failed) send is gone — not left silently live (F2 cleanup).
// Recover the token from the link the mailer was handed, and confirm it no longer resolves.
tokenFromLink := stub.link[strings.LastIndexByte(stub.link, '/')+1:]
if tokenFromLink == "" {
t.Fatal("mailer was never handed a link")
}
if tok, _ := st.SelfBindTokenByHash(selfBindHash(tokenFromLink)); tok != nil {
t.Fatal("a failed send left a live token behind")
}
}
// stubMailer records the last send and can be told to fail.
type stubMailer struct {
fail bool
link string
}
func (m *stubMailer) SendSelfBindEmail(customerID, email, link string) error {
m.link = link
if m.fail {
return errStubSend
}
return nil
}
var errStubSend = &stubErr{}
type stubErr struct{}
func (*stubErr) Error() string { return "stub send failure" }
// newAuthServer is newTestServer with an operator password configured, so RequireAuth is live.
func newAuthServer(t *testing.T) (*Server, *store.Store) {
t.Helper()
s, st := newTestServer(t)
h, err := bcrypt.GenerateFromPassword([]byte("operator-pw"), bcrypt.MinCost)
if err != nil {
t.Fatal(err)
}
s.configPasswordHash = string(h)
return s, st
}
+26 -3
View File
@@ -71,6 +71,8 @@ type Server struct {
pbsdrBox func() (monitor.PBSBoxSnapshot, bool) // optional (v0.65.0, R-5); the PBS-DR datastore fill snapshot accessor pbsdrBox func() (monitor.PBSBoxSnapshot, bool) // optional (v0.65.0, R-5); the PBS-DR datastore fill snapshot accessor
tenantsync tenancyProvisioner // optional; enables PBS DR tier provisioning (web/pbsdr.go) tenantsync tenancyProvisioner // optional; enables PBS DR tier provisioning (web/pbsdr.go)
claimEngine *claim.Engine // optional; enables the customer-claim resend button (v0.50.0) claimEngine *claim.Engine // optional; enables the customer-claim resend button (v0.50.0)
selfBindMailer SelfBindMailer // optional; enables the customer self-bind link button (v0.66.0, R-27)
bindLimiter *bindRateLimiter // per-IP throttle for the PUBLIC /bind/ surface (v0.66.0, R-27)
// intentHub (v0.58.0, Direction-2 immediate-sync) is Bumped by every operator-intent handler // intentHub (v0.58.0, Direction-2 immediate-sync) is Bumped by every operator-intent handler
// (config save/delete, claim resend, offsite re-issue/freeze, floor, block/unblock, log pull) // (config save/delete, claim resend, offsite re-issue/freeze, floor, block/unblock, log pull)
// so a box long-polling GET /api/v1/wait wakes in seconds. Shared with the API handler. nil = // so a box long-polling GET /api/v1/wait wakes in seconds. Shared with the API handler. nil =
@@ -129,6 +131,7 @@ func New(store *store.Store, passwordHash, apiKey, version string, staleThreshol
templates: tmpl, templates: tmpl,
staleThreshold: staleThreshold, staleThreshold: staleThreshold,
sessions: make(map[string]*hubSession), sessions: make(map[string]*hubSession),
bindLimiter: newBindRateLimiter(30), // public /bind/ surface: 30 req/min/IP burst (R-27)
} }
} }
@@ -253,7 +256,10 @@ func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
// CSRF protection for all state-changing requests (web routes only). // CSRF protection for all state-changing requests (web routes only).
// API routes (/api/v1/) are Bearer-token authenticated and exempt. // API routes (/api/v1/) are Bearer-token authenticated and exempt.
if r.Method != http.MethodGet && r.Method != http.MethodHead && r.Method != http.MethodOptions { if r.Method != http.MethodGet && r.Method != http.MethodHead && r.Method != http.MethodOptions {
if path != "/login" && s.effectivePasswordHash() != "" { // /bind/ is the public customer self-bind surface (THE TRAP §9.2): no operator session to
// ride, so CSRF is exempt here exactly as it is for /login. The URL capability token is the
// authorization boundary; a cross-site POST without both secrets only burns attempts.
if path != "/login" && !isPublicBindPath(path) && s.effectivePasswordHash() != "" {
if !s.validateCSRF(r) { if !s.validateCSRF(r) {
s.logger.Printf("[WARN] CSRF rejected: %s %s from %s", r.Method, path, r.RemoteAddr) s.logger.Printf("[WARN] CSRF rejected: %s %s from %s", r.Method, path, r.RemoteAddr)
http.Error(w, "CSRF token missing or invalid. Please reload the page.", http.StatusForbidden) http.Error(w, "CSRF token missing or invalid. Please reload the page.", http.StatusForbidden)
@@ -377,6 +383,10 @@ func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
s.handleHostDetail(w, r, hostID) s.handleHostDetail(w, r, hostID)
case path == "/login": case path == "/login":
s.handleLogin(w, r) s.handleLogin(w, r)
case isPublicBindPath(path):
// PUBLIC customer self-bind (R-27 slice 1) — GET renders the form/state, POST validates the
// two factors. Auth + CSRF exempt above via the SAME isPublicBindPath predicate.
s.handleBind(w, r)
case strings.HasPrefix(path, "/customers/") && strings.HasSuffix(path, "/block"): case strings.HasPrefix(path, "/customers/") && strings.HasSuffix(path, "/block"):
customerID := strings.TrimPrefix(path, "/customers/") customerID := strings.TrimPrefix(path, "/customers/")
customerID = strings.TrimSuffix(customerID, "/block") customerID = strings.TrimSuffix(customerID, "/block")
@@ -385,6 +395,15 @@ func (s *Server) ServeHTTP(w http.ResponseWriter, r *http.Request) {
} else { } else {
http.Error(w, "Method not allowed", http.StatusMethodNotAllowed) http.Error(w, "Method not allowed", http.StatusMethodNotAllowed)
} }
case strings.HasPrefix(path, "/customers/") && strings.HasSuffix(path, "/selfbind-link"):
// R-27 slice 1: mint + email a customer self-bind capability link. POST only.
customerID := strings.TrimPrefix(path, "/customers/")
customerID = strings.TrimSuffix(customerID, "/selfbind-link")
if r.Method == http.MethodPost {
s.handleSelfBindLinkSend(w, r, customerID)
} else {
http.Error(w, "Method not allowed", http.StatusMethodNotAllowed)
}
case strings.HasPrefix(path, "/customers/") && strings.HasSuffix(path, "/unblock"): case strings.HasPrefix(path, "/customers/") && strings.HasSuffix(path, "/unblock"):
customerID := strings.TrimPrefix(path, "/customers/") customerID := strings.TrimPrefix(path, "/customers/")
customerID = strings.TrimSuffix(customerID, "/unblock") customerID = strings.TrimSuffix(customerID, "/unblock")
@@ -561,8 +580,12 @@ func (s *Server) RequireAuth(next http.Handler) http.Handler {
return return
} }
// Always allow the login page through (GET and POST) // Always allow the login page through (GET and POST), and the PUBLIC customer self-bind
if r.URL.Path == "/login" { // surface (THE TRAP §9.2): /bind/ is exempt from operator auth exactly as /login is — the
// emailed URL capability token IS the auth model there. isPublicBindPath is the SINGLE
// definition of the prefix (matched tightly: trailing slash, path already .. -cleaned by the
// ServeMux) so this exemption cannot reach any operator-gated route.
if r.URL.Path == "/login" || isPublicBindPath(r.URL.Path) {
next.ServeHTTP(w, r) next.ServeHTTP(w, r)
return return
} }
@@ -55,6 +55,9 @@
{{else if eq .Flash "log_tail_requested"}}Log tail requested — the controller delivers it on its next report cycle (a few minutes). A customer-visible event line was recorded. {{else if eq .Flash "log_tail_requested"}}Log tail requested — the controller delivers it on its next report cycle (a few minutes). A customer-visible event line was recorded.
{{else if eq .Flash "claim-resent"}}Code re-sent to the registered address. A kód a doboz következő jelentésekor (~15 percen belül) aktiválódik. {{else if eq .Flash "claim-resent"}}Code re-sent to the registered address. A kód a doboz következő jelentésekor (~15 percen belül) aktiválódik.
{{else if eq .Flash "claim-resend-failed"}}Claim code resend FAILED — check the hub log (email delivery / send error). {{else if eq .Flash "claim-resend-failed"}}Claim code resend FAILED — check the hub log (email delivery / send error).
{{else if eq .Flash "selfbind-sent"}}Self-bind link sent to the registered address — valid for 7 days. The customer enters the box's console pairing code + their retrieval passphrase; no operator bind needed.
{{else if eq .Flash "selfbind-no-email"}}Self-bind link NOT sent — this customer has no registered email address. Set one first, or bind the appliance manually from the Hosts page.
{{else if eq .Flash "selfbind-send-failed"}}Self-bind link send FAILED — the link was invalidated (not left live). Check the hub log (email delivery / send error).
{{else if eq .Flash "reset_done"}}Customer RESET complete — every operational trace was destroyed (offsite repo, PBS namespace, DR recipe, claim state, retained escrow custody). Identity and basic config survive; the audit event stream records it. {{else if eq .Flash "reset_done"}}Customer RESET complete — every operational trace was destroyed (offsite repo, PBS namespace, DR recipe, claim state, retained escrow custody). Identity and basic config survive; the audit event stream records it.
{{end}} {{end}}
</div> </div>
@@ -454,6 +457,15 @@
<button type="submit" class="btn btn-outline btn-sm" data-confirm="Send a fresh code to the registered address? The previous code stops working immediately (the box activates it on its next report, ~15 min).">{{if .Claim.ClaimedAt}}Visszaállító kód küldése{{else}}Kód újraküldése{{end}}</button> <button type="submit" class="btn btn-outline btn-sm" data-confirm="Send a fresh code to the registered address? The previous code stops working immediately (the box activates it on its next report, ~15 min).">{{if .Claim.ClaimedAt}}Visszaállító kód küldése{{else}}Kód újraküldése{{end}}</button>
</form> </form>
{{end}} {{end}}
<div class="form-group" style="margin-top: 1rem; padding-top: 1rem; border-top: 1px solid var(--border);">
<label class="form-label">Customer self-bind (R-27)</label>
<span class="form-hint">Let the customer bind their own freshly-installed appliance — no operator bind needed. Sends a 7-day capability link to the registered address ({{.Email}}); the customer opens it and enters the box's <strong>console pairing code</strong> + their <strong>retrieval passphrase</strong>. Wrong entries lock the link after 5 attempts. If the link expires, bind the appliance manually from the Hosts page.</span>
<form method="POST" action="/customers/{{.CustomerID}}/selfbind-link" style="margin-top: 0.5rem;">
{{.CSRFField}}
<button type="submit" class="btn btn-outline btn-sm" data-confirm="Email a self-bind link to the registered address? Any previous self-bind link for this customer stops working immediately.">Send self-bind link</button>
</form>
</div>
</div> </div>
</section> </section>
+2 -1
View File
@@ -39,7 +39,7 @@
<div style="overflow-x: auto;"> <div style="overflow-x: auto;">
<table class="data-table"> <table class="data-table">
<thead> <thead>
<tr><th>Appliance</th><th>MACs</th><th>Hardware</th><th>SSH host keys</th><th>Seen</th><th>Bind to customer</th><th></th></tr> <tr><th>Appliance</th><th>Pairing code</th><th>MACs</th><th>Hardware</th><th>SSH host keys</th><th>Seen</th><th>Bind to customer</th><th></th></tr>
</thead> </thead>
<tbody> <tbody>
{{range .Unclaimed}} {{range .Unclaimed}}
@@ -48,6 +48,7 @@
{{if .Stale}}<br><span class="status-badge status-warn" title="No poll in over 7 days">stale</span>{{end}} {{if .Stale}}<br><span class="status-badge status-warn" title="No poll in over 7 days">stale</span>{{end}}
{{if .Bound}}<br><span class="status-badge status-ok" title="Bound — awaiting the box's next poll">bound → {{.BoundCustomer}}</span>{{end}} {{if .Bound}}<br><span class="status-badge status-ok" title="Bound — awaiting the box's next poll">bound → {{.BoundCustomer}}</span>{{end}}
</td> </td>
<td style="font-family: var(--font-mono); letter-spacing: 0.05em;">{{if .PairingCode}}<strong>{{.PairingCode}}</strong>{{else}}<span class="text-muted"></span>{{end}}</td>
<td style="font-size: 0.78em; font-family: var(--font-mono)">{{range .MACs}}{{.}}<br>{{end}}</td> <td style="font-size: 0.78em; font-family: var(--font-mono)">{{range .MACs}}{{.}}<br>{{end}}</td>
<td style="font-size: 0.8em;">{{if .Product}}{{.Product}}<br>{{end}}{{if .CPU}}<span class="text-muted">{{.CPU}}</span><br>{{end}}{{if .MemGB}}<span class="text-muted">{{.MemGB}}</span>{{end}}</td> <td style="font-size: 0.8em;">{{if .Product}}{{.Product}}<br>{{end}}{{if .CPU}}<span class="text-muted">{{.CPU}}</span><br>{{end}}{{if .MemGB}}<span class="text-muted">{{.MemGB}}</span>{{end}}</td>
<td style="font-size: 0.72em; font-family: var(--font-mono)">{{range .SSHFingerprints}}{{.}}<br>{{end}}</td> <td style="font-size: 0.72em; font-family: var(--font-mono)">{{range .SSHFingerprints}}{{.}}<br>{{end}}</td>
+12
View File
@@ -1,5 +1,17 @@
# Felhom scripts — Changelog # Felhom scripts — Changelog
## build-felhom-iso.sh v1.20.0 — console pairing-code banner for customer self-bind (R-27 slice 1) (2026-07-17)
Supports the hub's customer self-bind flow (hub v0.66.0). In PAIRING mode, `felhom-bootstrap.sh` now
reads the additive `pairing_code` from the `POST /api/v1/appliance/register` response, persists it at
`/etc/felhom/appliance-pairing-code`, and prints a Hungarian **console banner** (to `/dev/console`,
stdout fallback) each pairing cycle so the customer can read the code off the physical screen and type
it — together with their retrieval passphrase — on the hub's public `/bind/<token>` page. The code is
**non-secret** (possession proof only; the passphrase is the second factor), so it is safe on the
console. **Graceful degradation both ways:** a hub older than v0.66.0 omits `pairing_code` → the banner
prints nothing and register/poll are unchanged; an old ISO against a v0.66.0 hub simply ignores the new
field. No change to DIRECT mode. Green: `bash -n` clean on both scripts.
## build-felhom-iso.sh v1.19.0 — the universal secret-free ISO: `--pairing` mode (R-21 slice C) (2026-07-17) ## build-felhom-iso.sh v1.19.0 — the universal secret-free ISO: `--pairing` mode (R-21 slice C) (2026-07-17)
The scripts half of the universal ISO. `felhom-bootstrap.sh` gains a PAIRING mode — **one unit, two The scripts half of the universal ISO. `felhom-bootstrap.sh` gains a PAIRING mode — **one unit, two
+1 -1
View File
@@ -32,7 +32,7 @@
#=============================================================================== #===============================================================================
set -euo pipefail set -euo pipefail
ISO_VERSION="1.19.0" # Felhom release the ISO is tagged to (aligns with felhom-host-install SCRIPT_VERSION). ISO_VERSION="1.20.0" # Felhom release the ISO is tagged to (aligns with felhom-host-install SCRIPT_VERSION).
IMAGE="${FELHOM_ISO_ASSISTANT_IMAGE:-felhom-iso-assistant:trixie}" IMAGE="${FELHOM_ISO_ASSISTANT_IMAGE:-felhom-iso-assistant:trixie}"
HERE="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)" HERE="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+24 -1
View File
@@ -33,9 +33,26 @@ STATE_FILE=/var/lib/felhom-install/state.json
PASS_FILE=/run/felhom-bootstrap-pass PASS_FILE=/run/felhom-bootstrap-pass
SCRIPT_TMP=/run/felhom-host-install.sh SCRIPT_TMP=/run/felhom-host-install.sh
TOKEN_FILE=/etc/felhom/appliance-token # PAIRING: the box's only pre-day-0 credential (0600, persists reboots) TOKEN_FILE=/etc/felhom/appliance-token # PAIRING: the box's only pre-day-0 credential (0600, persists reboots)
PAIRING_CODE_FILE=/etc/felhom/appliance-pairing-code # R-27: non-secret pairing code shown on the console
log() { echo "felhom-bootstrap: $*"; } log() { echo "felhom-bootstrap: $*"; }
# print_pairing_banner (R-27, v0.66.0) — show the pairing code prominently on the physical console while
# the box waits to be bound, so the customer can read it into the self-bind page. Non-secret (the bind
# still requires the customer's retrieval passphrase). A hub older than v0.66.0 sends no code → no banner
# (the box stays operator-bind-only — graceful, no behavior change).
print_pairing_banner() {
local code; code=$(cat "$PAIRING_CODE_FILE" 2>/dev/null)
[[ -n "$code" ]] || return 0
{ printf '\n================================================\n'
printf ' Felhom — a doboz parositasra var / párosításra vár\n\n'
printf ' Párosító kód: %s\n\n' "$code"
printf ' Nyisd meg az e-mailben kapott self-bind linket,\n'
printf ' és add meg ezt a kódot + a jelszavadat.\n'
printf '================================================\n\n'
} > /dev/console 2>/dev/null || printf 'Párosító kód: %s\n' "$code"
}
cleanup_pass() { [[ -e "$PASS_FILE" ]] && { shred -u "$PASS_FILE" 2>/dev/null || rm -f "$PASS_FILE"; }; return 0; } cleanup_pass() { [[ -e "$PASS_FILE" ]] && { shred -u "$PASS_FILE" 2>/dev/null || rm -f "$PASS_FILE"; }; return 0; }
trap cleanup_pass EXIT trap cleanup_pass EXIT
@@ -175,8 +192,14 @@ run_pairing() {
exit 1 exit 1
fi fi
( umask 077; printf '%s' "$token" > "$TOKEN_FILE" ) ( umask 077; printf '%s' "$token" > "$TOKEN_FILE" )
log "registered — appliance token stored (0600); waiting for the operator to bind this box" # R-27 (v0.66.0): persist the non-secret pairing code (absent on a pre-v0.66.0 hub — tolerated).
local pcode; pcode=$(printf '%s' "$resp" | python3 -c 'import json,sys; print(json.load(sys.stdin).get("pairing_code",""))' 2>/dev/null)
if [[ -n "$pcode" ]]; then
printf '%s' "$pcode" > "$PAIRING_CODE_FILE"
fi
log "registered — appliance token stored (0600); waiting for the operator or a customer self-bind"
fi fi
print_pairing_banner # show the code on the console each pairing cycle
# 2. ONE poll. RestartSec=30 is the poll interval. # 2. ONE poll. RestartSec=30 is the poll interval.
local token; token=$(cat "$TOKEN_FILE") local token; token=$(cat "$TOKEN_FILE")