fix(scep-intune): close 11 audit gaps from 2026-04-29 pre-tag review

Closes the eleven gaps identified in the pre-v2.1.0 audit of the SCEP
RFC 8894 + Intune master bundle (cowork/scep-bundle-gap-closure-prompt.md).
Constitutional rule from cowork/CLAUDE.md::Operating Rules — 'Always
take the complete path, not the easy path' — drove this closure: each
gap was a load-bearing wire that crossed multiple layers (config →
validator → service wire-up → tests → docs) and shipping the bundle
without them would have produced lying-field footguns where operator-
visible config options stored values without affecting behavior.

WHAT LANDS:

Phase A — Clock-skew tolerance (master prompt §15 hazard closure)
  internal/scep/intune/challenge.go: ValidateChallenge migrated from
  positional args to ValidateOptions{} struct; new ClockSkewTolerance
  field with default 0 (strict). 24 call sites updated mechanically.
  Asymmetric application: now+tolerance >= iat AND now-tolerance < exp.
  internal/config/config.go: SCEPIntuneProfileConfig.ClockSkewTolerance
  default 60s + Validate() refusal when >= ChallengeValidity.
  cmd/server/main.go: SetIntuneIntegration signature extended;
  per-profile env-var loader honors CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CLOCK_SKEW_TOLERANCE.
  internal/service/scep.go: intuneClockSkew field + IntuneStatsSnapshot
  surfaces clock_skew_tolerance_ns. web/src/api/types.ts mirrors.
  4 new tests in challenge_test.go covering accept-within-tolerance,
  reject-beyond-tolerance, accept-expired-within-tolerance,
  negative-treated-as-zero defensive normalization.
  docs/scep-intune.md updated with the new env var + time-bounds rule.

Phase B — unknown-version-rejected golden test
  internal/scep/intune/golden_helper_test.go: goldenUnknownVersionPayload
  helper + signGoldenChallengeAny generic signer.
  challenge_golden_test.go: TestGoldenChallenge_UnknownVersionRejected
  uses an in-process ECDSA fixture (the on-disk PEM was generated with
  a Go-stdlib version that produces different ecdsa.GenerateKey bytes
  from the current call). TestRegenerateGoldenFixtures emits the new
  unknown_version fixture file too.

Phase C — Two named Intune e2e tests
  internal/api/handler/scep_intune_e2e_test.go:
    TestSCEPIntuneEnrollment_RateLimited_E2E (cap=2 + 3 attempts; 3rd
    returns FAILURE+badRequest with rate_limited counter ticked)
    TestSCEPIntuneEnrollment_TrustAnchorSIGHUPReload_E2E (rotate
    on-disk PEM + holder.Reload(); old-key challenge fails with
    badMessageCheck; signature_invalid counter ticked)
  intuneE2EFixture struct extended with trustHolder + trustPath fields
  so tests can rotate.

Phase D — Four new ChromeOS hermetic tests (10 total now)
  internal/api/handler/scep_chromeos_test.go:
    _RAKeyMismatch — PKIMessage encrypted to wrong RA cert; handler
      rejects without reaching service.
    _3DESBackwardCompat — RFC 8894 §3.5.2 legacy fallback verified.
    _RSACSR + _ECDSACSR — explicit matrix-pair pinning.
  buildTestECDSACSR helper for ECDSA P-256 CSR construction;
  tripleDESCBCEncrypt mirrors aesCBCEncrypt for 3DES-CBC;
  assertChromeOSPositiveCertRep shared assertion.

Phase E — Per-profile counter isolation test
  internal/api/handler/scep_profile_counter_isolation_test.go:
    TestSCEPHandler_PerProfileIntuneCountersIsolated wires two
    SCEPService instances + drives distinct PKIMessages + asserts
    counter isolation. Guards against a future cmd/server/main.go
    refactor that shares a *intuneCounterTab across profiles.
  buildPerProfileIntuneFixture parameterized helper.

Phase F — Server-boot regression tests
  cmd/server/preflight_scep_intune_test.go: 3 named tests covering
  disabled-backward-compat, broken-config-with-PathID, expired-cert
  refusal. preflightSCEPIntuneTrustAnchor signature extended with
  pathID arg so error messages carry PathID= for operator log-grep.

Phase G — docs/connectors.md
  Four new subsections under §EST/SCEP Integration: multi-profile
  dispatch + mTLS sibling route + Intune Connector dispatcher + SCEP
  probe in network scanner. Each has a one-paragraph operator
  explanation + an env-var or endpoint table.

Phase H — Coverage uplift
  internal/service/scep_probe_persist_test.go: 5 unit tests on
  persistProbeResult (nil-safe + nil-repo-safe + repo-error swallow +
  nil-logger guard) + ListRecentSCEPProbes (empty-slice-not-nil + repo
  pass-through) + describeCertAlgorithm (RSA/ECDSA/QF1008-nil-curve
  defensive branch/Ed25519/DSA/empty). CI gates (service ≥70, handler
  ≥75) PASS at 70.9% / 79.3%.

Phase I — deploy/test integration variant
  deploy/test/scep_intune_e2e_test.go (//go:build integration):
    TestSCEPIntuneEnrollment_Integration + _RateLimited_Integration
    against the live docker-compose certctl container. Skip-when-
    stack-missing semantics so sandbox + CI both work.
  deploy/docker-compose.test.yml: new e2eintune SCEP profile env
  vars + bind-mount of deploy/test/fixtures/.
  deploy/test/fixtures/README.md: documents the deterministic trust
  anchor regeneration recipe.

VERIFICATION (sandbox):
  gofmt -d        — clean for all changed files
  staticcheck     — clean for intune + handler + config + service +
                    cmd/server packages
  go vet          — clean for the same packages
  go test -short  — green for intune (95.3% cov), service (70.9%),
                    handler (79.3%), config (94.0%), cmd/server (boot
                    path; my preflight tests cover the directly-
                    testable function), pkcs7 (80.5% informational)

DEFERRED (per closure prompt §7 out-of-scope):
  - V3-Pro Conditional Access gating + Microsoft Graph integration
  - Standalone certctl-scan CLI binary
  - OCSP rate-limiting, OCSP stapling, delta CRLs

Spec preserved at cowork/scep-bundle-gap-closure-prompt.md;
journal at cowork/scep-rfc8894-intune/progress.md (audit-closure
section appended).
This commit is contained in:
Shankar
2026-04-29 20:28:53 +00:00
parent 9fcea95708
commit 444942eab8
20 changed files with 2143 additions and 74 deletions
+52
View File
@@ -331,6 +331,58 @@ Note: EST and SCEP are not connectors — they are protocol handlers (`internal/
**SCEP RA cert + key (post-2026-04-29):** the SCEP server's RFC 8894 path requires an RA cert/key pair (`CERTCTL_SCEP_RA_CERT_PATH` + `CERTCTL_SCEP_RA_KEY_PATH`, mode 0600) — clients encrypt their CSR to the RA cert's public key per RFC 8894 §3.2.2. Multi-profile deployments configure per-profile pairs via `CERTCTL_SCEP_PROFILES=corp,iot` + `CERTCTL_SCEP_PROFILE_<NAME>_RA_*_PATH`. See [`legacy-est-scep.md`](legacy-est-scep.md#scep-rfc-8894-native-implementation-post-2026-04-29) for the openssl recipe + ChromeOS Admin Console pointer + must-staple per-profile policy.
#### Multi-profile SCEP dispatch
A single certctl deploy can publish multiple SCEP endpoints — one per fleet, one per device class, or one per Connector — by setting `CERTCTL_SCEP_PROFILES=<comma-separated>` and a matching set of `CERTCTL_SCEP_PROFILE_<NAME>_*` environment variables. The router publishes `/scep/<pathID>?operation=...` for every profile whose `<NAME>` appears in the list (or `/scep` for the legacy single-profile shape when `CERTCTL_SCEP_PROFILES` is unset). Each profile carries its OWN issuer binding, RA cert/key pair, challenge password, must-staple policy, optional mTLS sibling route, and optional Microsoft Intune Connector trust anchor — heterogeneous fleets share one server, distinct credentials.
| Variable | Required | Default | Description |
|----------|----------|---------|-------------|
| `CERTCTL_SCEP_PROFILES` | No | — | Comma-separated profile names (e.g. `corp,iot`). When unset, the legacy single-profile config (`CERTCTL_SCEP_*` without the `_PROFILE_<NAME>_` infix) is used. |
| `CERTCTL_SCEP_PROFILE_<NAME>_ISSUER_ID` | Yes | — | Issuer connector ID this profile dispatches to (e.g. `iss-local`, `iss-ejbca-corp`). |
| `CERTCTL_SCEP_PROFILE_<NAME>_PROFILE_ID` | No | — | Optional certificate profile ID for fine-grained issuance policy. |
| `CERTCTL_SCEP_PROFILE_<NAME>_CHALLENGE_PASSWORD` | No | — | Static challenge password for the legacy SCEP auth path. Set to "" when only Intune dynamic challenges are expected. |
| `CERTCTL_SCEP_PROFILE_<NAME>_RA_CERT_PATH` | Yes | — | RA cert PEM path (mode 0600 enforced). |
| `CERTCTL_SCEP_PROFILE_<NAME>_RA_KEY_PATH` | Yes | — | RA private key PEM path (mode 0600 enforced). |
See [`legacy-est-scep.md`](legacy-est-scep.md#scep-rfc-8894-native-implementation-post-2026-04-29) for the full per-profile env-var list and the mTLS / Intune extensions.
#### SCEP mTLS sibling route (opt-in)
For deploys that already have a previously-issued certctl client cert and want a stronger renewal binding than the static challenge password, certctl exposes an opt-in mTLS sibling route at `/scep-mtls/<pathID>`. The TLS handshake is configured with `tls.VerifyClientCertIfGiven` against an operator-supplied trust bundle; presented client certs are validated against the bundle before the SCEP handler runs. The standard `/scep/<pathID>` route stays open for new-enrollment devices that don't yet have a client cert.
| Variable | Required | Default | Description |
|----------|----------|---------|-------------|
| `CERTCTL_SCEP_PROFILE_<NAME>_MTLS_ENABLED` | No | `false` | Set `true` to publish `/scep-mtls/<pathID>` alongside `/scep/<pathID>`. |
| `CERTCTL_SCEP_PROFILE_<NAME>_MTLS_CLIENT_CA_TRUST_BUNDLE_PATH` | When MTLS enabled | — | PEM bundle of CAs that may sign client certs. Preflight refuses a missing/empty bundle. |
See [`legacy-est-scep.md`](legacy-est-scep.md#scep-mtls-sibling-route-phase-65) for the operator recipe + threat-model rationale.
#### Microsoft Intune Certificate Connector dispatcher
When a profile has `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_ENABLED=true`, certctl validates the Microsoft Intune Certificate Connector's signed-challenge JWS natively as a drop-in NDES replacement (the Intune Connector documents itself as RFC 8894-compliant and works against any RFC 8894 SCEP server). The dispatcher walks parse → JWS signature verify (RS256 + ES256, alg=none rejected) → version dispatch → time bounds with ±tolerance → audience pin → CSR ↔ claim binding → replay cache → per-device rate limit → optional V3-Pro compliance hook. The trust anchor file is reloaded on `SIGHUP` (operator rotates the on-disk PEM, then `kill -HUP <certctl-pid>`); a parse failure during reload keeps the OLD pool so a half-rotation doesn't take Intune down.
| Variable | Required | Default | Description |
|----------|----------|---------|-------------|
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_ENABLED` | No | `false` | Gate the dispatcher. |
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CONNECTOR_CERT_PATH` | When enabled | — | PEM bundle of the Connector's signing certs. Preflight refuses a missing/expired bundle. |
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_AUDIENCE` | No | — | Expected `aud` claim (typically the public SCEP URL the Connector calls). Empty disables the audience check. |
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CHALLENGE_VALIDITY` | No | `60m` | Defense-in-depth cap on top of the challenge's own `exp`. |
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CLOCK_SKEW_TOLERANCE` | No | `60s` | ±tolerance on iat/exp checks. Raise on poorly-NTP-synced fleets, lower to enforce strict time. Refused at boot when ≥ `INTUNE_CHALLENGE_VALIDITY`. |
| `CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_PER_DEVICE_RATE_LIMIT_24H` | No | `3` | Max enrollments per `(claim.Subject, claim.Issuer)` in any rolling 24h window. Zero disables. |
See [`scep-intune.md`](scep-intune.md) for the full deployment guide — NDES + EJBCA migration playbook, Intune SCEP profile field mapping, trust-anchor extraction recipe, monitoring + Prometheus alert thresholds, and the Microsoft Learn citations operators paste into procurement-team requests.
#### SCEP probe in network scanner
The Network Scans GUI surface includes a one-click "Probe SCEP" form that runs a capability + posture check against any reachable SCEP server URL — `GetCACaps` + `GetCACert` (NEVER `PKCSReq`) so the probe is read-only and safe to run against production endpoints. Result fields surface advertised caps (POSTPKIOperation, SHA-256, SHA-512, AES, SCEPStandard, Renewal), CA cert subject + issuer + algorithm + days-to-expiry + chain length, and a probe duration. Results persist to `scep_probe_results` (migration `000021`) and the probe history is paginated under `GET /api/v1/network-scan/scep-probes`. Useful for pre-migration assessment ("what does the existing NDES advertise?") and compliance-posture audits.
| Endpoint | Auth | Description |
|----------|------|-------------|
| `POST /api/v1/network-scan/scep-probe` | Bearer | Body `{"url":"https://..."}`. Synchronous probe; returns `SCEPProbeResult`. |
| `GET /api/v1/network-scan/scep-probes` | Bearer | Recent probe history, paginated `[1, 200]`. |
The probe goes through the same dual-layer SSRF defense (`validation.ValidateSafeURL` up-front + `SafeHTTPDialContext` at dial time) as the rest of the network scanner. Standalone CLI binary is explicitly deferred — the in-tree network scanner is the only entrypoint today.
### Built-in: Vault PKI
The Vault PKI connector integrates with HashiCorp Vault's PKI secrets engine using its native `/sign` API with token-based authentication. This is ideal for organizations using Vault as their internal certificate authority — synchronous issuance without the complexity of ACME or challenge solving.
+10 -2
View File
@@ -84,9 +84,14 @@ PKIMessage with the documented `pkiStatus`/`failInfo` codes (per RFC
payload prelude. v1 (current Connector format, no `version` key)
routes to `unmarshalChallengeV1`. Future v2 plugs in a sibling
parser without touching the validator.
4. **Time bounds**`now ≥ iat AND now < exp`. Configurable cap on
4. **Time bounds**`now+tolerance ≥ iat AND now-tolerance < exp`.
The `±tolerance` window is configurable per profile via
`INTUNE_CLOCK_SKEW_TOLERANCE` (default 60s, covers modest clock
drift between the Connector host and certctl). Configurable cap on
top via `INTUNE_CHALLENGE_VALIDITY` (defense-in-depth against a
Connector that mints long-validity challenges).
Connector that mints long-validity challenges). The validator
refuses `tolerance ≥ ChallengeValidity` at startup-validation time
to keep the cap meaningful.
5. **Audience pin**`claim.aud == INTUNE_AUDIENCE` (skipped when
`INTUNE_AUDIENCE` is empty for proxy/load-balancer scenarios).
6. **CSR binding**`claim.DeviceMatchesCSR(csr)` checks
@@ -156,6 +161,9 @@ so the production validation step is a manual run-book item.
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_ENABLED=true
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CONNECTOR_CERT_PATH=/etc/certctl/intune-corp.pem
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_AUDIENCE=https://certctl.example.com/scep/corp
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CHALLENGE_VALIDITY=60m
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_CLOCK_SKEW_TOLERANCE=60s # ±tolerance on iat/exp; raise on poorly-NTP-synced fleets, lower to enforce strict time
CERTCTL_SCEP_PROFILE_<NAME>_INTUNE_PER_DEVICE_RATE_LIMIT_24H=3
```
Restart certctl. The startup preflight refuses to boot if the