fix: distinguish identity session failures

Principal resolution could not tell a broken session store from a genuinely
anonymous request, and did unnecessary database work for callers that never
presented a session at all. Both are corrected here.

Any cookie previously triggered the ambiguity check, so `Cookie: x=y` on an
otherwise anonymous request caused an identity database round trip an
unauthenticated caller could drive with a header they chose themselves.
`hasConfiguredSessionCookie` now parses the Cookie header per RFC 6265 (split
on `;`, name is the text before the first `=`, trimmed) and compares the whole
name exactly — never `includes`, a substring test, or an unanchored regular
expression — so `better-auth.session_token_x`, `x_better-auth.session_token`, a
differently-cased spelling, and a percent-encoded lookalike are all correctly
not the session cookie. A duplicated name still reports true: ambiguity fails
safe towards checking, and reporting true never grants access on its own.

The cookie name is not a second hard-coded copy. One `betterAuthOptions` object
now feeds both `betterAuth()` and Better Auth's own supported `getCookies`
helper, so the detected name — including the `__Secure-` prefix an https baseURL
implies — is derived from the exact configuration Better Auth itself uses, and
drift is structurally impossible rather than merely tested.

The readiness check itself proved almost nothing: `select 1` still succeeded
against a dropped session table, a drifted or removed column, or a revoked
SELECT privilege, so Better Auth's ambiguous null was reported as a genuinely
anonymous request. `probeSessionStore` now plans a real read of the session
relation and the exact columns the adapter depends on, under `where false`:
PostgreSQL resolves the relation, every column, and the caller's privilege at
plan time while matching zero rows, so no session row, token, or user id can be
read, logged, or enumerated by the check. It uses the injected pool, never a
second connection, and stays bounded by the existing timeout.

A missing relation, dropped column, revoked privilege, unreachable connection,
and query timeout now all produce the same sanitized 503 with no relation name,
column name, SQL, driver text, host, or stack — proven against real PostgreSQL,
since that is the only place a broken store looks exactly like an anonymous
request.

Co-Authored-By: Claude Opus 5 <[email protected]>
This commit is contained in:
ekklesia
2026-07-26 17:02:42 -05:00
co-authored by Claude Opus 5
parent 3d6c85b0f8
commit 7a0f624fc3
6 changed files with 654 additions and 132 deletions
+31 -1
View File
@@ -144,7 +144,8 @@ ADR 0010.
`handleAuthRequest(request): Promise<Response>` (delegates directly to Better Auth's handler; its
normal responses, including its own error responses, pass through unmodified) and
`resolvePrincipal(headers): Promise<EkklesiaPrincipal>` (never silently downgrades a database
outage to an anonymous result — see ADR 0011 for exactly how). Neither method, nor anything else
outage to an anonymous result — see ADR 0011 and [session resolution](#how-a-principal-is-resolved)
below). Neither method, nor anything else
exported from `@ekklesia/core/identity`, exposes `auth.api`, Better Auth's session/user types, the
Drizzle adapter, the raw database, a session token, or Better Auth configuration. `apps/api` is the
only current consumer of this boundary; see [API](api.md).
@@ -166,6 +167,35 @@ hung or slow query, which `connectionTimeoutMs` alone cannot do since it only bo
the connection. `apps/api` sets both timeouts explicitly (see [API](api.md)); an invalid value is
rejected the same sanitized way an invalid connection string already is.
## How a principal is resolved
`resolvePrincipal(headers)` returns the anonymous principal for a missing, expired, revoked, or
malformed session, and throws a sanitized `PRINCIPAL_RESOLUTION_FAILED` for an infrastructure
failure. Telling those apart takes explicit work, because Better Auth's `getSession` swallows
internal failures and returns `null` for them exactly as it does for a genuinely absent session.
1. **No cookies, or only unrelated cookies → no identity database work at all.** `Cookie: x=y` is
not evidence that Better Auth ever attempted to resolve an Ekklesia session, so its `null` is
unambiguous and the anonymous principal is returned immediately. An unauthenticated caller
cannot drive identity database load with a cookie header they chose themselves.
2. **The configured session cookie is present, and Better Auth returned `null` → the ambiguous
case, and only then is the session store checked.** Detection is an exact, whole-name cookie
comparison (RFC 6265 parsing; names are case-sensitive and never percent-decoded), so
`better-auth.session_token_x` and `x_better-auth.session_token` are correctly *not* the session
cookie. The name itself is derived from the same options object passed to Better Auth, via its
own exported `getCookies` helper — including the `__Secure-` prefix an `https` base URL implies
— so there is no second copy of it to drift.
3. **The check reads the real session relation**, not `select 1`: it plans a read of the exact
columns the Better Auth adapter depends on under a `where false` predicate. PostgreSQL resolves
the relation, every named column, and the caller's `SELECT` privilege while matching zero rows,
so no session row, token, or user id can be read, logged, or enumerated by the check. It runs on
the injected pool — never a second connection — under a bounded 3-second timeout.
A missing session table, a dropped or drifted column, a revoked `SELECT` privilege, an unreachable
database, and a query timeout therefore all produce a sanitized
`503 {"code":"PRINCIPAL_RESOLUTION_FAILED"}` — never a silent anonymous result. The response body,
logs, and error carry no relation name, column name, SQL, driver message, host, or stack.
## What is not operational yet
No email verification, password recovery, invitation flow, email delivery, social or OAuth