Set the approval store tenant to exact tenant:platform

Operator decision 5ed3fb35-eca9-413a-82b9-95171ba85bf6 accepts tenant:platform
as the platform management, administration and services tenant, with no alias
to platform or tenant:coulomb and no implicit cross-tenant grant. This closes
the collision recorded in 5c87ba8, where the manifest served --tenant platform
while the requested registrations issued tenant:coulomb.

The store tenant is now exactly tenant:platform in the manifest, the CLI
default, and the Engine default, and the requested client registrations ask for
the same spelling. Exact JWT/store equality is retained: no mapping table, no
normalisation, no prefix handling.

Moving the defaults rather than only the manifest is deliberate. A default of
platform under a sanctioned value of tenant:platform is a trap, because a serve
that omits --tenant would come up healthy and then refuse every authenticated
call -- the exact failure this decision exists to prevent.

That default change broke ten tests whose identity fixtures hard-coded
platform. This is the hazard flex-auth reported as FLEX-DEC-2026-008: fixtures
that all carry one tenant prove nothing about the tenant field. Fixtures are
aligned to the exact spelling, and the field is now varied rather than merely
present. test_near_miss_tenant_spellings_are_forbidden refuses platform,
tenant:coulomb, case variants, whitespace variants and empty against a
tenant:platform store; test_exact_sanctioned_tenant_is_admitted pins the other
half so a reject-everything bug cannot pass it. 111 tests pass.

Also records the credential-independent half of the GLAS-WP-0015 image request:
the image builds non-root uid 10001 off the pinned base, carries schema v3 and
the new tenant default, migrates and verifies a fresh store to schema_version 3
with integrity ok, and refuses production without a persistent database or
authenticated audit delivery. No scan was run -- no scanner is installed here --
and no release digest exists, so T01 and T03 both stay open.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PM5HnEAhokxdfcPqBNpT7D

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 715850@bnt-lap001
Assistant-Session: eb557e93-7cb1-45d0-9e57-7d15b3edc60e
This commit is contained in:
tegwick 2026-09-06 22:33:50 +02:00
parent 5c87ba8610
commit 6d18f62a90
8 changed files with 182 additions and 38 deletions

View file

@ -37,7 +37,7 @@ def _parser() -> argparse.ArgumentParser:
serve.add_argument("--host", default="127.0.0.1") serve.add_argument("--host", default="127.0.0.1")
serve.add_argument("--port", type=int, default=8787) serve.add_argument("--port", type=int, default=8787)
serve.add_argument("--production", action="store_true") serve.add_argument("--production", action="store_true")
serve.add_argument("--tenant", default="platform") serve.add_argument("--tenant", default="tenant:platform")
serve.add_argument("--jwt-issuer") serve.add_argument("--jwt-issuer")
serve.add_argument("--jwt-audience") serve.add_argument("--jwt-audience")
serve.add_argument("--jwks-url") serve.add_argument("--jwks-url")

View file

@ -165,7 +165,7 @@ class Engine:
clock: Callable[[], datetime] | None = None, clock: Callable[[], datetime] | None = None,
freshness_ttl: int = DEFAULT_FRESHNESS_TTL, freshness_ttl: int = DEFAULT_FRESHNESS_TTL,
fail_outbox: bool = False, fail_outbox: bool = False,
tenant: str = "platform", tenant: str = "tenant:platform",
auto_migrate: bool = True, auto_migrate: bool = True,
) -> None: ) -> None:
self.path = str(path) self.path = str(path)

View file

@ -64,7 +64,7 @@ spec:
- --port - --port
- "8080" - "8080"
- --tenant - --tenant
- platform - tenant:platform
- --jwt-issuer - --jwt-issuer
- https://kc.coulomb.social - https://kc.coulomb.social
- --jwt-audience - --jwt-audience

View file

@ -21,44 +21,57 @@ Required claims remain those in `docs/caller-authentication.md`: `iss`, `sub`,
| JWKS | `GET /jwks` on the KeyCape service | | JWKS | `GET /jwks` on the KeyCape service |
| Scopes | `approval:create`, `approval:read`, `approval:approve`, `approval:revoke`, `approval:supersede`, `approval:consume`, `approval:observe`, `approval:emit` | | Scopes | `approval:create`, `approval:read`, `approval:approve`, `approval:revoke`, `approval:supersede`, `approval:consume`, `approval:observe`, `approval:emit` |
## Tenant reconciliation — unresolved, blocks token issuance ## Tenant — resolved: exact `tenant:platform`
**This is a live collision, not a naming preference.** `ApiApplication.identity` The operator accepted `tenant:platform` as the platform management,
compares the verified JWT `tenant` claim to the engine's configured store tenant administration and services tenant (the landlord zone). Decision
with exact string equality and raises `Forbidden` before any object lookup `5ed3fb35-eca9-413a-82b9-95171ba85bf6`, recorded in
`glas-harness/docs/platform-tenant-decision.md` and relayed by `glas-harness`
2026-09-06.
**The spelling is the contract.** `ApiApplication.identity` compares the
verified JWT `tenant` claim to the engine's configured store tenant with exact
string equality and raises `Forbidden` before any object lookup
(`approval_engine/api.py:66`). There is no mapping table, no normalisation, and (`approval_engine/api.py:66`). There is no mapping table, no normalisation, and
no prefix handling anywhere in this engine. no prefix handling anywhere in this engine — deliberately. `platform` is **not**
an accepted alias for `tenant:platform`, and neither is `tenant:coulomb`.
The three values currently in play: Rendered values, all three now exactly `tenant:platform`:
| Value | Where it is set | Current content | | Value | Where it is set | Content |
| --- | --- | --- | | --- | --- | --- |
| Store tenant | `deploy/approval-engine.yaml` `--tenant` (CLI default `platform`) | `platform` | | Store tenant | `deploy/approval-engine.yaml` `--tenant` | `tenant:platform` |
| JWT `tenant` claim | the client registrations below | `tenant:coulomb` | | Store tenant default | `approval_engine/cli.py` `--tenant`, `Engine(tenant=…)` | `tenant:platform` |
| CheckRequest tenant | flex-auth policy subject | `tenant:platform`**never read by this engine** | | JWT `tenant` claim | the client registrations below | `tenant:platform` |
| CheckRequest tenant | flex-auth policy subject | `tenant:platform` |
`platform` != `tenant:coulomb`, so tokens issued under the registrations below The CLI and `Engine` defaults were moved off `platform` in the same change. A
would be denied `403` on every non-health route. Spelling similarity between default that differs from the sanctioned value is a trap: a `serve` invocation
`platform` and `tenant:platform` is not a mapping either; the flex-auth policy that omits `--tenant` would have come up healthy and then refused every
subject is a PDP input this engine never inspects, so it cannot participate in authenticated call, which is the failure this decision exists to prevent.
the comparison at all.
Denial evidence: `tests/test_auth.py::test_wrong_tenant_is_forbidden` — a The flex-auth CheckRequest tenant now matches by spelling, but note it still
signature-valid token whose tenant differs from the store tenant is refused does not participate in this comparison — it is a PDP policy subject this engine
without mutation. never reads. Alignment there is a property of the decision, not a mechanism
here.
**Resolution is an owner decision and is deliberately not taken here.** Either Denial evidence:
KeyCape issues `tenant: platform` to match the store, or this repo's manifest
sets `--tenant tenant:coulomb` to match the registration. Which is correct - `tests/test_auth.py::test_wrong_tenant_is_forbidden` — a signature-valid token
depends on whether `platform` and `tenant:coulomb` name the same layer — a whose tenant differs from the store tenant is refused `403` without mutation.
question this engine cannot answer, and answering it wrongly grants a token - `tests/test_auth.py::test_near_miss_tenant_spellings_are_forbidden` — pins the
cross-tenant access to the approval store. The values below are left as decision's "no alias" clause directly: `platform`, `tenant:coulomb`,
requested until that mapping is stated by an owner, so the mismatch stays `TENANT:PLATFORM`, `tenant:platform ` (trailing space) and `` (empty) are each
visible rather than being silently resolved by whichever document was edited refused against a `tenant:platform` store, while the exact spelling is
last. admitted. Without a test that *varies* the tenant, a suite of fixtures all
carrying the sanctioned value proves nothing about the field.
This resolves the choice of value only. Verification and the remaining admission
gates — KeyCape actually owning these registrations, credential materialization,
and the `T03` rollout evidence — are unchanged and still open.
Tracked against `APPROVAL-WP-0002-T01`; related `KEY-WP-0013-T02`, Tracked against `APPROVAL-WP-0002-T01`; related `KEY-WP-0013-T02`,
`SECRETS-WP-0009-T03`. `SECRETS-WP-0009-T03`, `GLAS-WP-0015`.
## Clients ## Clients
@ -76,7 +89,7 @@ clients:
secretRef: env:KEYCAPE_SECRETS_ENGINE_APPROVAL_CLIENT_SECRET secretRef: env:KEYCAPE_SECRETS_ENGINE_APPROVAL_CLIENT_SECRET
serviceSubject: service:secrets-engine serviceSubject: service:secrets-engine
principal_type: service principal_type: service
tenant: tenant:coulomb tenant: tenant:platform
roles: [secrets-engine] roles: [secrets-engine]
tokenLifetime: 15m tokenLifetime: 15m
@ -96,7 +109,7 @@ clients:
secretRef: env:KEYCAPE_APPROVAL_ENGINE_OPERATOR_CLIENT_SECRET secretRef: env:KEYCAPE_APPROVAL_ENGINE_OPERATOR_CLIENT_SECRET
serviceSubject: service:approval-engine-operator serviceSubject: service:approval-engine-operator
principal_type: service principal_type: service
tenant: tenant:coulomb tenant: tenant:platform
roles: [approval-operator] roles: [approval-operator]
tokenLifetime: 15m tokenLifetime: 15m
``` ```

View file

@ -28,7 +28,7 @@ def app(engine):
issuer="https://keycape.example", issuer="https://keycape.example",
audiences=("approval-engine",), audiences=("approval-engine",),
principal_type="service", principal_type="service",
tenant="platform", tenant="tenant:platform",
roles=frozenset({"secrets-engine"}), roles=frozenset({"secrets-engine"}),
scopes=frozenset( scopes=frozenset(
{ {

View file

@ -118,7 +118,7 @@ def test_api_requires_scope_and_binds_create_actor(engine):
issuer="test", issuer="test",
audiences=("approval-engine",), audiences=("approval-engine",),
principal_type="service", principal_type="service",
tenant="platform", tenant="tenant:platform",
roles=frozenset(), roles=frozenset(),
scopes=frozenset({"approval:create"}), scopes=frozenset({"approval:create"}),
assurance={"level": "aal1"}, assurance={"level": "aal1"},
@ -201,6 +201,72 @@ def test_wrong_tenant_is_forbidden(engine):
assert body["error"] == "forbidden" assert body["error"] == "forbidden"
@pytest.mark.parametrize(
"tenant",
[
"platform",
"tenant:coulomb",
"TENANT:PLATFORM",
"Tenant:Platform",
"tenant:platform ",
" tenant:platform",
"tenant:platform:",
"",
],
)
def test_near_miss_tenant_spellings_are_forbidden(engine, tenant):
"""Decision 5ed3fb35 accepted exactly `tenant:platform` with no alias.
The store comparison is exact string equality, so every near miss below must
be refused: the bare `platform` this repo used to serve, the `tenant:coulomb`
the registrations used to request, case variants, and whitespace. Varying the
field is the point a suite whose fixtures all carry the sanctioned value
proves nothing about the tenant check.
"""
identity = Identity(
subject="agt-secrets-engine",
issuer="test",
audiences=("approval-engine",),
principal_type="service",
tenant=tenant,
roles=frozenset(),
scopes=frozenset({"approval:observe"}),
assurance={},
evidence_ref="test",
)
from approval_engine.api import App
app = App(engine, StaticTokenAuthenticator({"near": identity}))
status, body = call(app, "GET", "/v1/cadence", authorization="Bearer near")
assert status == 403, f"{tenant!r} was admitted as an alias"
assert body["error"] == "forbidden"
def test_exact_sanctioned_tenant_is_admitted(engine):
"""The other half of the pin: the exact spelling must actually work.
Without this, a bug that rejected every tenant would pass the near-miss test
above while denying the sanctioned caller too.
"""
assert engine.tenant == "tenant:platform"
identity = Identity(
subject="agt-secrets-engine",
issuer="test",
audiences=("approval-engine",),
principal_type="service",
tenant="tenant:platform",
roles=frozenset(),
scopes=frozenset({"approval:observe"}),
assurance={},
evidence_ref="test",
)
from approval_engine.api import App
app = App(engine, StaticTokenAuthenticator({"exact": identity}))
status, _ = call(app, "GET", "/v1/cadence", authorization="Bearer exact")
assert status == 200
def test_deny_all_default_does_not_mutate(engine): def test_deny_all_default_does_not_mutate(engine):
from approval_engine.api import App from approval_engine.api import App
@ -225,7 +291,7 @@ def test_human_principal_cannot_consume(engine):
issuer="test", issuer="test",
audiences=("approval-engine",), audiences=("approval-engine",),
principal_type="service", principal_type="service",
tenant="platform", tenant="tenant:platform",
roles=frozenset(), roles=frozenset(),
scopes=frozenset( scopes=frozenset(
{"approval:create", "approval:approve", "approval:read", "approval:consume"} {"approval:create", "approval:approve", "approval:read", "approval:consume"}
@ -238,7 +304,7 @@ def test_human_principal_cannot_consume(engine):
issuer="test", issuer="test",
audiences=("approval-engine",), audiences=("approval-engine",),
principal_type="human", principal_type="human",
tenant="platform", tenant="tenant:platform",
roles=frozenset(), roles=frozenset(),
scopes=frozenset({"approval:consume"}), scopes=frozenset({"approval:consume"}),
assurance={"level": "aal2"}, assurance={"level": "aal2"},

View file

@ -185,7 +185,7 @@ def live_http(tmp_path):
issuer="https://keycape.example", issuer="https://keycape.example",
audiences=("approval-engine",), audiences=("approval-engine",),
principal_type="service", principal_type="service",
tenant="platform", tenant="tenant:platform",
roles=frozenset({"secrets-engine"}), roles=frozenset({"secrets-engine"}),
scopes=frozenset( scopes=frozenset(
{ {

View file

@ -90,6 +90,37 @@ the approval store. The registrations doc's stale issuer
(`https://auth.netkingdom.local`) is corrected to the live (`https://auth.netkingdom.local`) is corrected to the live
`https://kc.coulomb.social` from `06544b0`. T01 stays `progress`. `https://kc.coulomb.social` from `06544b0`. T01 stays `progress`.
2026-09-06 resolution: the operator accepted `tenant:platform` as the platform
management/administration/services tenant (landlord zone) — decision
`5ed3fb35-eca9-413a-82b9-95171ba85bf6`,
`glas-harness/docs/platform-tenant-decision.md`, relayed by `glas-harness`. The
collision above is closed by setting the store tenant to exactly
`tenant:platform`: `deploy/approval-engine.yaml` `--tenant`, the
`approval_engine/cli.py` `--tenant` default, and the `Engine(tenant=…)` default
all move off bare `platform`, and the requested client registrations now ask for
`tenant: tenant:platform`. Exact JWT/store equality is retained — no alias, no
normalisation, no prefix handling, and no implicit cross-tenant grant.
Moving the *defaults* rather than only the manifest is deliberate: a default of
`platform` under a sanctioned value of `tenant:platform` is a trap, because a
`serve` that omits `--tenant` would come up healthy and then refuse every
authenticated call.
The default change broke ten tests whose identity fixtures hard-coded
`platform`, which is the fixture-consistency hazard flex-auth reported as
`FLEX-DEC-2026-008` (29 fixtures all carrying one tenant proved nothing about
the field). Fixtures are aligned to the exact spelling, and the field is now
*varied* rather than merely present:
`tests/test_auth.py::test_near_miss_tenant_spellings_are_forbidden` refuses
`platform`, `tenant:coulomb`, case variants, whitespace variants and empty
against a `tenant:platform` store, and
`test_exact_sanctioned_tenant_is_admitted` pins the other half so a
reject-everything bug cannot pass. 111 tests pass.
This resolves the choice of value only. T01 stays `progress`: KeyCape still has
to own and prove these registrations, and credential materialization is
unchanged.
## Harden durable storage and migrations ## Harden durable storage and migrations
```task ```task
@ -138,6 +169,40 @@ read-only root, resources, probes, and default-deny policies pass client dry-run
Waiting on release digest, KeyCape/audit registrations and credentials, rollout, Waiting on release digest, KeyCape/audit registrations and credentials, rollout,
restart, and restore evidence. restart, and restore evidence.
2026-09-06 image preparation (GLAS-WP-0015 request; credential-independent half):
built and validated locally against the agreed tenant and current schema.
- Build: `make image-build` off the digest-pinned base
`python:3.12-slim@sha256:d764629c…`. Local manifest-list digest
`sha256:85e46ddf47b3ac0cdab620163a7b034027451dff286b63ecff6a2493d7a57ecc`.
**This is a local build digest, not a release digest** — the manifest still
carries `REPLACE_WITH_RELEASE_DIGEST` because pinning requires a push to
`forgejo.coulomb.social`, which needs registry credentials this session does
not hold.
- Runtime identity: `uid=10001(approval) gid=10001(approval)`, non-root as
required.
- Schema: image carries `LATEST_SCHEMA_VERSION = 3`, matching the migrated
store, and `Engine` tenant default `tenant:platform`.
- First-install migration (no prior DB): `migrate` then `verify` on a fresh
volume both report `schema_version: 3`, `schema_current: true`,
`integrity: ["ok"]`, `foreign_key_violations: 0`, `persistent: true`.
- Fail-closed configuration proven in the image, not only in tests:
`serve --production --db :memory:` refuses with "production requires a
persistent database"; `serve --production` on a real DB without audit
configuration refuses with "production requires authenticated audit
delivery".
- Manifest inputs: `kubectl apply --dry-run=client` passes for the namespace,
service, and StatefulSet with the new `--tenant tenant:platform`.
Scan gate NOT met: no scanner (`trivy`, `grype`, `docker scout`) is installed on
this workstation, so no vulnerability scan was run and none is claimed. The
inventory a scanner needs is the pinned base above plus
`cryptography==50.0.1`, `PyJWT==2.13.0`, `waitress==3.0.2`, `cffi==2.1.1`,
`pycparser==3.0`, `pip==25.0.1`.
T03 stays `wait`: still no release digest, no KeyCape/audit credentials, no
rollout, and no restart/restore evidence. Nothing here is a deploy.
## Wire outbox delivery and reconciliation ## Wire outbox delivery and reconciliation
```task ```task