Five defects in revocation state and credential semantics. Lands after the tracker-lifecycle batch on purpose: raising the over-cap TTL is only safe once the count feeding it is trustworthy. #13 -- a longer old revocation suppressed a newer cutoff. applyLocal kept or replaced the WHOLE record by expiry, so when the existing revocation expired later the new one was dropped entirely, including its newer RevokedAt. The durable upsert did the same, with a comment documenting it as intentional. RevokedAt is the user-kill CUTOFF, so this left a credential issued between the two cutoffs valid -- a second admin kill after a user re-authenticates silently failed to cut them. The two fields now merge independently: ExpiresAt stays monotonic, RevokedAt advances to the later value, and reason follows the newer cutoff. Both superseded comments are replaced rather than left contradicting the code. Session-kind revocation still ignores RevokedAt, so the enforcer's re-revoke cannot weaken a session kill. Also fixed while here: Redis received the merged record but pub/sub published the raw input one, so under pub/sub-only delivery (Redis down) an edge got the newer short record without the older long expiry and lost monotonicity. Both now carry the merged record. #7 + M1 -- Postgres could indefinitely block the urgent Redis kill. RevokeWithWarnings held the global opMu across all propagation, stripped the caller's deadline with WithoutCancel, and did the durable Postgres upsert BEFORE Redis, on a pool with no statement timeout. The local kill still applied, so playback on that process was fine -- but edge propagation, pub/sub, the admin response and every later revoke/unrevoke stalled behind the lock. Redis and pub/sub now go first, and the detached context is bounded. WithoutCancel is kept deliberately: propagation must outlive an aborted admin request. opMu scope is deliberately NOT narrowed. mirrorToRedis is an unconditional SET with no atomic merge, so same-process serialization is what stops an older value overwriting a newer one; narrowing the lock would also let an unrevoke interleave with a revoke's propagation. Bounding the context caps how long the lock can be held, which is the actual reported harm. The remaining cross-replica race -- two central replicas racing the same SET -- is documented, not half-fixed; it needs A6's shared picture. A2 / #6 -- a missed unrevoke got resurrected. In-memory tombstones already existed, but being process-local they did not survive a restart or reach a replica that missed the pub/sub event, so maintain's durable self-heal re-Upserted the surviving entry and the ban returned. Tombstones are now durable, via two nullable columns on stream_revocations rather than a second table: a tombstone is a state of the same key, and it needs its own expiry horizon separate from the revocation's. The upsert rejects a stale replica's write while a tombstone is live but lets a genuinely newer revocation clear it, and warm paths apply tombstones BEFORE revocations so an un-banned key cannot be restored as a live kill. Tombstones are pruned on the same sweep, so the table cannot grow without bound. A1 / #3 -- over-cap kills reopened after 5 minutes while the token stayed reconstructable for 24h. The TTL now derives from playback.MaxTokenTTL rather than duplicating 24h, behind a validated setting. Critically, the enforcer uses a revoke-if-absent path rather than re-revoking. Expiry is monotonic and the enforcer re-evaluates every 30s, so a plain long TTL would slide expiry forward by another full lifetime on every pass -- making a wrong kill effectively permanent for as long as any stale record persisted, with only an explicit unrevoke to recover it. Admin Revoke keeps its monotonic behaviour; only the enforcer's own repeat kill is non-extending. The setting is documented as affecting future revocations only, since monotonic expiry means it cannot shorten one already issued. A3 / #5 -- the user cutoff compared against a fresh time.Now() taken at request entry, so a request from a pre-cutoff login could look post-cutoff and escape the kill. The credential time is now the access token's iat. Two deliberate choices worth stating. API-key credentials carry no issue time, so they pass the zero time and, per IsRevoked's documented contract, are never matched by a user cutoff: a user kill provably cannot cut an API-key-owned pour. That is an accepted, logged, documented hole -- and strictly better than time.Now(), which actively defeats the cutoff. And jellycompat uses the compat session's CreatedAt rather than the bridged Silo token's iat, because that token refreshes without a new Jellyfin login, so its iat would advance on refresh and let a refreshed credential slip past a cutoff. Stream tokens are now bound to their route: a token whose SessionID does not match the URL's session_id is rejected with 403 instead of being silently ignored, matching the reconstruction helper that already refused a different session. Per-login logout cuts remain out of scope -- they need per-login identity in the stream credential. S5 (the (sessionID, userID, startedAt) clump) is rejected as ceremony now that A3 is the iat option rather than the generation model. #12 (closing an RSS feed does not cut its current pour) is deferred: it needs a namespaced revocation id that cannot collide with real session ids, that id threaded onto public feed requests, and protection against a new feed inheriting an old tombstone. Part of #305.
157 lines
6.0 KiB
Go
157 lines
6.0 KiB
Go
package streamrevoke
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"time"
|
|
|
|
"github.com/jackc/pgx/v5/pgxpool"
|
|
)
|
|
|
|
// permanentExpiry is the far-future sentinel written when a Revocation has a
|
|
// zero-value ExpiresAt. The hot path treats a zero ExpiresAt as "never expires"
|
|
// (a permanent kill), but the DB column is NOT NULL and Prune/List compare
|
|
// expires_at <= now(): a literal zero time (0001-01-01) would be excluded by
|
|
// List and deleted by the very next Prune, silently evaporating a
|
|
// permanent kill. Writing a year-2999 sentinel preserves the intent durably.
|
|
var permanentExpiry = time.Date(2999, 1, 1, 0, 0, 0, 0, time.UTC)
|
|
|
|
// PostgresDurableStore is the Postgres-backed DurableStore: a durable mirror of
|
|
// the kill list so revocations survive a Redis flush or a server restart. It is
|
|
// never on the hot path — Store consults it only on write (Upsert), on
|
|
// warm/reconcile (List), and on trim (Prune).
|
|
//
|
|
// Rows are keyed by (kind, id) so re-revoking the same session/user (the async
|
|
// over-cap enforcer does this every pass) UPSERTs the same row rather than
|
|
// accumulating duplicates; physical growth is reclaimed by Prune.
|
|
type PostgresDurableStore struct {
|
|
pool *pgxpool.Pool
|
|
}
|
|
|
|
const durableUpsertSQL = `
|
|
INSERT INTO stream_revocations (kind, id, reason, revoked_at, expires_at)
|
|
VALUES ($1, $2, $3, $4, $5)
|
|
ON CONFLICT (kind, id) DO UPDATE SET
|
|
reason = CASE WHEN EXCLUDED.revoked_at > stream_revocations.revoked_at
|
|
THEN EXCLUDED.reason ELSE stream_revocations.reason END,
|
|
revoked_at = GREATEST(stream_revocations.revoked_at, EXCLUDED.revoked_at),
|
|
expires_at = CASE WHEN stream_revocations.unrevoked_at IS NOT NULL
|
|
THEN EXCLUDED.expires_at
|
|
ELSE GREATEST(stream_revocations.expires_at, EXCLUDED.expires_at) END,
|
|
unrevoked_at = NULL,
|
|
tombstone_expires_at = NULL
|
|
WHERE stream_revocations.unrevoked_at IS NULL
|
|
OR stream_revocations.tombstone_expires_at <= now()
|
|
OR EXCLUDED.revoked_at > stream_revocations.unrevoked_at`
|
|
|
|
const durableTombstoneUpsertSQL = `
|
|
INSERT INTO stream_revocations (
|
|
kind, id, reason, revoked_at, expires_at, unrevoked_at, tombstone_expires_at
|
|
)
|
|
VALUES ($1, $2, '', $3, $4, $3, $4)
|
|
ON CONFLICT (kind, id) DO UPDATE SET
|
|
unrevoked_at = EXCLUDED.unrevoked_at,
|
|
tombstone_expires_at = EXCLUDED.tombstone_expires_at
|
|
WHERE EXCLUDED.unrevoked_at >= stream_revocations.revoked_at
|
|
AND (
|
|
stream_revocations.unrevoked_at IS NULL
|
|
OR EXCLUDED.unrevoked_at >= stream_revocations.unrevoked_at
|
|
)`
|
|
|
|
// NewPostgresDurableStore builds a DurableStore from a pgx pool. It returns a
|
|
// nil DurableStore interface when pool is nil so callers can pass the result
|
|
// straight into Options.Durable and a Redis-less/DB-less mode degrades to a
|
|
// true nil interface (avoiding the "non-nil interface wrapping a nil pointer"
|
|
// trap that would make Store.durable != nil erroneously true).
|
|
func NewPostgresDurableStore(pool *pgxpool.Pool) DurableStore {
|
|
if pool == nil {
|
|
return nil
|
|
}
|
|
return &PostgresDurableStore{pool: pool}
|
|
}
|
|
|
|
// Upsert writes or refreshes a revocation, keyed by (kind, id).
|
|
func (s *PostgresDurableStore) Upsert(ctx context.Context, r Revocation) error {
|
|
// A zero ExpiresAt means "permanent" on the hot path; persist it as a
|
|
// far-future sentinel so Prune/List don't immediately reap the row.
|
|
expiresAt := r.ExpiresAt
|
|
if expiresAt.IsZero() {
|
|
expiresAt = permanentExpiry
|
|
}
|
|
// Merge cutoff and expiry independently, matching applyLocal: the later
|
|
// RevokedAt (and its reason) wins while expiry remains monotonic. A live
|
|
// tombstone rejects stale revocations; a genuinely newer revocation replaces
|
|
// the tombstone and starts a fresh expiry horizon.
|
|
_, err := s.pool.Exec(ctx, durableUpsertSQL,
|
|
string(r.Kind), r.ID, r.Reason, r.RevokedAt, expiresAt)
|
|
if err != nil {
|
|
return fmt.Errorf("streamrevoke upsert: %w", err)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// UpsertTombstone durably records an explicit unrevoke. A stale tombstone
|
|
// cannot erase a newer revocation, and a stale replica cannot overwrite a live
|
|
// tombstone through Upsert.
|
|
func (s *PostgresDurableStore) UpsertTombstone(ctx context.Context, t Tombstone) error {
|
|
if _, err := s.pool.Exec(ctx, durableTombstoneUpsertSQL,
|
|
string(t.Kind), t.ID, t.UnrevokedAt, t.ExpiresAt); err != nil {
|
|
return fmt.Errorf("streamrevoke tombstone upsert: %w", err)
|
|
}
|
|
return nil
|
|
}
|
|
|
|
// List returns one snapshot of every active revocation and live tombstone.
|
|
// Tombstoned rows are never surfaced as revocations.
|
|
func (s *PostgresDurableStore) List(ctx context.Context) (DurableState, error) {
|
|
rows, err := s.pool.Query(ctx, `
|
|
SELECT kind, id, reason, revoked_at, expires_at,
|
|
unrevoked_at, tombstone_expires_at
|
|
FROM stream_revocations
|
|
WHERE (unrevoked_at IS NULL AND expires_at > now())
|
|
OR (unrevoked_at IS NOT NULL AND tombstone_expires_at > now())`)
|
|
if err != nil {
|
|
return DurableState{}, fmt.Errorf("streamrevoke list: %w", err)
|
|
}
|
|
defer rows.Close()
|
|
|
|
var state DurableState
|
|
for rows.Next() {
|
|
var r Revocation
|
|
var kind string
|
|
var unrevokedAt, tombstoneExpiresAt *time.Time
|
|
if err := rows.Scan(
|
|
&kind, &r.ID, &r.Reason, &r.RevokedAt, &r.ExpiresAt,
|
|
&unrevokedAt, &tombstoneExpiresAt,
|
|
); err != nil {
|
|
return DurableState{}, fmt.Errorf("streamrevoke scan: %w", err)
|
|
}
|
|
r.Kind = Kind(kind)
|
|
if unrevokedAt != nil {
|
|
if tombstoneExpiresAt != nil {
|
|
state.Tombstones = append(state.Tombstones, Tombstone{
|
|
Kind: r.Kind, ID: r.ID, UnrevokedAt: *unrevokedAt, ExpiresAt: *tombstoneExpiresAt,
|
|
})
|
|
}
|
|
continue
|
|
}
|
|
state.Revocations = append(state.Revocations, r)
|
|
}
|
|
if err := rows.Err(); err != nil {
|
|
return DurableState{}, fmt.Errorf("streamrevoke list rows: %w", err)
|
|
}
|
|
return state, nil
|
|
}
|
|
|
|
// Prune physically deletes expired revocations and expired tombstones so
|
|
// neither state grows without bound.
|
|
func (s *PostgresDurableStore) Prune(ctx context.Context) error {
|
|
if _, err := s.pool.Exec(ctx, `
|
|
DELETE FROM stream_revocations
|
|
WHERE (unrevoked_at IS NULL AND expires_at <= now())
|
|
OR (unrevoked_at IS NOT NULL AND tombstone_expires_at <= now())`); err != nil {
|
|
return fmt.Errorf("streamrevoke prune: %w", err)
|
|
}
|
|
return nil
|
|
}
|