Files
OwnCord/Server/db/auth_queries.go
T
J3vbandClaude Fable 5 f5faf82a60 infra: observability, backups, guardrails, and deployment hardening (#1376)
* docs: add infrastructure roadmap plan

Records the verified recommendations from an infrastructure review in three
tracks: raising the single-instance ceiling, cheap seams for a possible
multi-instance future, and ops hygiene. Includes explicit anti-recommendations
and sequencing. Security-sensitive detail is intentionally excluded per
docs/security.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* feat(server): real health checks and saturation metrics

/api/v1/metrics now exposes signals that were already computed in memory but
never surfaced: reconnect replay tier hits, event-persister counters, SQLite
writer-pool wait stats, aggregate per-client backpressure counters (including
previously invisible low-priority drops), and permission-cache hit/miss.

/health now returns a real verdict: hub dispatch-loop liveness, a bounded
database ping, and a free-disk check, returning 503 with a subsystem reason
when degraded. Checks are cached so the unauthenticated endpoint cannot
amplify load. The hub's panic breaker now exits the process so a supervisor
can restart it, instead of leaving broadcast delivery silently dead while
clients still appear online.

OTel instruments that were declared but never recorded are now wired
(ws_active_connections, ws_broadcast_latency_seconds, ws_messages_total,
ws_events_dropped_total, voice gauges) or removed (db_query_duration_seconds).
Also corrects the docs/api.md description of broadcast_drops, which counts
hub-queue overflow, not client send-queue overflow.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* feat(server): implement scheduled backups, retention, and backup verification

The backup_schedule and backup_retention settings have existed in the admin
panel and API since the initial schema but were never read by any code. The
15-minute maintenance loop now enforces them: a scheduled backup is taken
when the newest backup on disk is older than the schedule interval (manual
backups reset the clock), and retention prunes backups older than the
configured days while always keeping the newest one.

Backups are now verified with PRAGMA integrity_check immediately after
VACUUM INTO (a failed backup is removed rather than listed as restorable)
and again before a restore may overwrite the live database. A failed VACUUM
INTO also cleans up its partial output file — but never a pre-existing one.

The backup directory is configurable via a new backup.dir key (default
data/backups) so operators can point backups at another disk or an off-host
mount, mirroring the SetDatabasePath plumb.

Restore-handler tests now use real SQLite fixtures (the integrity gate
correctly refuses text files) with the mid-copy failure injected through a
test-only copy hook. Also adds audited gosec suppressions to the Windows
disk-free syscall added in the previous commit, which the Windows lint leg
flagged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* feat(server): capacity and failure-mode guardrails

- server.max_ws_connections: optional cap on concurrent WebSocket clients,
  checked before the upgrade with a 503 + Retry-After; rejections are counted
  and exposed as ws_conn_rejects in /api/v1/metrics.
- Single-process database lock: an OS-level advisory lock (flock / exclusive
  handle) beside the SQLite file makes a second server process fail fast with
  a clear message instead of silently fighting the first over process-local
  state. A bounded retry covers the self-update/restore restart handoff, and
  the lock mechanism failing (e.g. network filesystems) only warns.
- Disk-space awareness: boot-time warnings for the data and backup volumes,
  plus a disk_free_mb metrics field, via a small cross-platform diskutil
  package (already used by /health).
- Upload storage failures: storage.Save now marks server-side filesystem
  failures with a sentinel (storage.ErrIO); handlers return 507 for those
  instead of blaming the client with a 400, and the emoji route stops echoing
  raw storage errors (which embed absolute paths) into responses.
- Unknown config keys now warn at startup — a typo like admin_alowed_cidrs
  previously kept the default silently while the operator believed the
  setting changed. Never fatal: newer servers tolerate older configs.
- Admin settings honesty: the three stored-but-inert settings (server_icon,
  max_upload_bytes, voice_quality) are shown read-only with a note pointing
  at the real config.yaml keys, instead of pretending to apply.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* perf(db): write-path efficiency and capacity knobs

- channel_focus/mark_read now skip the read-state UPSERT when the stored row
  already matches (same last_message_id, no mentions) — refocus events fire
  at up to 10/s/user and every no-op write still occupied the single SQLite
  writer connection. The extra existence check runs on the reader pool, which
  doesn't serialize. Same shape as the session-touch throttle.
- DeleteExpiredSessions is now sargable: migration 031 normalizes legacy
  expiry formats to the RFC3339-Z layout the server writes and indexes
  expires_at, replacing the strftime full-table scan that ran on the writer
  every 15 minutes.
- Boot-time ANALYZE runs only when a migration actually applied; unchanged
  schemas get the cheap PRAGMA optimize instead (which also covers
  crash-restarts that never reached the shutdown optimize).
- The read/write SQL router gets a table-driven test with explicit expected
  values (INSERT ... RETURNING must hit the writer despite being :one).
- New knobs, all defaulting to current behavior: database.max_readers,
  security.auth_rate_limit_multiplier (for shared-NAT communities),
  event_persistence.replay_ring_size and replay_cold_limit.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* fix(server): shutdown lifecycle ordering

- The event pruner and maintenance loop are now joined (bounded) before the
  database closes: bgCtx cancellation used to run AFTER database.Close via
  LIFO defers, contradicting its own comment, and neither goroutine was ever
  waited on — a mid-tick scheduled backup or prune could still hold the
  writer while the pool tore down. StartEventPruner returns a done channel
  with the same join contract EventPersister.Stop already had.
- srv.Shutdown now runs before hub.GracefulStop, so in-flight HTTP handlers'
  broadcasts still reach a live hub and the event persister instead of
  vanishing from the replay/event store across a restart. Shutdown does not
  wait on hijacked WebSocket connections, so the swap adds no delay.
- GracefulStopContext threads the 30s shutdown budget into the hub: the 5s
  client-notice window (matching the countdown clients are shown) ends early
  when the budget expires, and is skipped entirely when nobody is connected —
  early-return startup paths and idle servers no longer sleep 5s for an
  audience of zero.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* build(deploy): systemd unit, compose hardening, boot-smoked releases, CI polish

- deploy/owncord.service: hardened systemd unit template with the two
  verified caveats encoded (install dir stays writable for self-update under
  ProtectSystem=strict; CAP_NET_BIND_SERVICE for ACME's :80), plus a
  'Linux (systemd)' deployment docs section — the Linux service story was
  previously 'Docker or nothing'.
- New 'Reverse Proxy Topology' docs section with a working nginx snippet and
  the correct signaling-vs-media distinction: /livekit/* is already proxied
  by the server, only WebRTC media ports must be directly reachable.
- docker-compose: log rotation, commented resource limits, and a healthcheck
  backed by a new 'chatserver healthcheck' subcommand (the distroless image
  has no shell) that probes /health without config side effects.
- release.yml: a concurrency group (queue, never cancel), and boot-smoke
  gates — the freshly built server binaries and the Docker image are cold
  booted and probed healthy BEFORE anything is signed or pushed. The release
  feed drives signed self-updates, so a binary that compiles but dies on
  boot previously would have shipped itself to every auto-updating instance.
- ci.yml: client-check/client-tests move to ubuntu with the reasoning
  recorded (no win32 code paths, LF enforced repo-wide); admin-e2e gets a
  written graduation criterion instead of an open-ended non-blocking status.
- docs: Tailscale guide notes the CGNAT range vs the default admin CIDRs;
  architecture overview records presence/voice state as the fifth
  single-instance blocker and the macOS client scope decision.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* perf(server): measured load tooling, narrowed invalidation, presence coalescing, storage and CIDR seams

- Fix scripts/k6/ws-load.js against the real wire protocol: envelope-wrapped
  frames, correct message types (typing_start, presence_update), the correct
  /api/v1/ws path, and thresholds that fail a run where nobody authenticated
  or went ready — the script had drifted to pre-envelope framing and reported
  100% green while every auth failed on the first frame. A new
  workflow_dispatch-only load-baseline workflow boots a real server, seeds
  users through the setup/invite APIs, runs the script, and uploads the k6
  summary plus a metrics snapshot for before/after comparison.
- Role-scoped channel-override changes now evict only the affected role's
  members from the permission cache (fail-safe: unreadable member list still
  flushes everything). InvalidateAll here repopulated every connected user —
  two reads each — synchronously inside the admin request via
  RefreshChannelVisibility, a stampede that scaled with total population
  rather than the role's size. Same pattern the per-user override endpoints
  already used.
- Connect/disconnect presence broadcasts now pass through a 300ms latest-wins
  coalescer (QueuePresence): each un-coalesced presence change is a sequenced
  global broadcast (an O(clients) fan-out under seqMu), so a reconnect storm
  fired O(users) of them from the connect critical path. A flap inside the
  window collapses to its final state; the wire format, seq ordering, and
  replay behaviour are unchanged, and the delivery path (BroadcastPresence)
  is untouched.
- Storage seam: api handlers now consume a FileStore interface (consumer-side,
  same pattern as service.Store) with Open returning a seekable storage.File —
  writing down the contract (range-request seeks included) an alternative
  backend would have to meet, without building one.
- The metrics surfaces and the LiveKit webhook/health endpoints get their own
  allowlist keys (metrics_allowed_cidrs, livekit_webhook_allowed_cidrs, both
  defaulting to admin_allowed_cidrs), so a central Prometheus scraper or an
  externally-hosted LiveKit no longer requires widening the admin panel's
  perimeter. Startup now also warns when admin_allowed_cidrs is customized
  while trusted_proxies is empty — behind a proxy or container network the
  check would otherwise compare the proxy's private address, not the client's.
- The container healthcheck probe now PINS the server's own certificate from
  disk (VerifyConnection, exact-match) instead of skipping TLS verification,
  addressing the CodeQL finding on the previous commit; WebPKI verification
  is used when no local cert exists (ACME).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* fix(server): address self-review findings on the hardening branch

Seven fixes from a high-effort review of the full branch diff:

- healthcheck CLI now works under tls.mode acme: it overrides ServerName
  with the configured domain for WebPKI verification instead of pinning a
  cert that doesn't exist (or is stale) in that mode. Previously an ACME
  deployment's container healthcheck failed forever.
- /health pings the READER pool (new db.PingRead): the writer ping queued
  behind a scheduled backup's VACUUM INTO and reported the server degraded
  for the whole backup — which an autoheal watchdog would turn into a
  nightly mid-backup restart.
- /health runs its cached checks under context.WithoutCancel so a probe
  that disconnects mid-request cannot poison the shared cache with a false
  degraded verdict for the next 5 seconds.
- The token CLI uses a new db.OpenShared that skips the single-process
  lock: minting a token against a running server is safe under WAL and was
  a documented workflow the lock had broken.
- The per-user TOTP failure cap is no longer scaled by
  security.auth_rate_limit_multiplier — that knob exists for per-IP limits;
  scaling the only cross-IP brute-force defence multiplied an attacker's
  distributed guess budget. Mirrors the unscaled per-user login threshold.
- A direct presence_update now drops the user's queued entry in the
  connect/disconnect coalescer, so a stale connect-time presence can no
  longer flush 300ms later over the user's fresher chosen status.
- The scheduled-backup filename collision loop breaks on any stat error
  and bounds its suffix probing, instead of spinning the maintenance
  goroutine forever on a persistent EACCES.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

* test(admin): real SQLite fixture for the merged Close-failure restore test

TestHandleRestoreBackup_RestartsWhenCloseFails arrived from main (#1375)
with a plain-text backup fixture; this branch's restore handler verifies
backups with integrity_check before touching the live database, so the text
fixture was (correctly) refused with 400 before the Close-failure branch
under test was reached. Use a real backup via BackupToSafe, matching the
other restore tests.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RtDNHSYWwPKArL8MsRdbj

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-08-15 20:50:47 +02:00

578 lines
19 KiB
Go

package db
import (
"context"
"crypto/rand"
"database/sql"
"encoding/hex"
"errors"
"fmt"
"strings"
"time"
"github.com/owncord/server/db/dbgen"
)
// ─── User Operations ──────────────────────────────────────────────────────────
// CreateUser inserts a new user record and returns the assigned ID.
func (d *DB) CreateUser(ctx context.Context, username, passwordHash string, roleID int) (int64, error) {
res, err := d.writer.ExecContext(ctx,
`INSERT INTO users (username, password, role_id) VALUES (?, ?, ?)`,
username, passwordHash, roleID,
)
if err != nil {
return 0, fmt.Errorf("CreateUser: %w", err)
}
return res.LastInsertId()
}
// CreateOwnerIfEmpty atomically checks that no users exist and inserts the
// first owner in a single transaction. Returns ErrConflict if any user already
// exists, closing the TOCTOU race in the setup endpoint (BUG-119).
func (d *DB) CreateOwnerIfEmpty(ctx context.Context, username, passwordHash string, roleID int) (int64, error) {
tx, err := d.writer.BeginTx(ctx, nil)
if err != nil {
return 0, fmt.Errorf("CreateOwnerIfEmpty begin: %w", err)
}
committed := false
defer func() {
if !committed {
_ = tx.Rollback()
}
}()
var count int64
if err := tx.QueryRow(`SELECT COUNT(*) FROM users`).Scan(&count); err != nil {
return 0, fmt.Errorf("CreateOwnerIfEmpty count: %w", err)
}
if count > 0 {
return 0, ErrConflict
}
res, err := tx.Exec(
`INSERT INTO users (username, password, role_id) VALUES (?, ?, ?)`,
username, passwordHash, roleID,
)
if err != nil {
return 0, fmt.Errorf("CreateOwnerIfEmpty insert: %w", err)
}
uid, err := res.LastInsertId()
if err != nil {
return 0, fmt.Errorf("CreateOwnerIfEmpty last_id: %w", err)
}
if err := tx.Commit(); err != nil {
return 0, fmt.Errorf("CreateOwnerIfEmpty commit: %w", err)
}
committed = true
return uid, nil
}
// CreateUserWithInvite atomically consumes an invite and creates the user in
// the same transaction so a failed registration does not burn the invite.
func (d *DB) CreateUserWithInvite(ctx context.Context, username, passwordHash string, roleID int, inviteCode string) (int64, error) {
tx, err := d.writer.BeginTx(ctx, nil)
if err != nil {
return 0, fmt.Errorf("CreateUserWithInvite begin: %w", err)
}
committed := false
defer func() {
if !committed {
_ = tx.Rollback()
}
}()
result, err := tx.Exec(
`UPDATE invites SET use_count = use_count + 1
WHERE code = ? AND revoked = 0
AND (max_uses IS NULL OR use_count < max_uses)
AND (expires_at IS NULL OR strftime('%s', expires_at) > strftime('%s', 'now'))`,
inviteCode,
)
if err != nil {
return 0, fmt.Errorf("CreateUserWithInvite use invite: %w", err)
}
rows, err := result.RowsAffected()
if err != nil {
return 0, fmt.Errorf("CreateUserWithInvite invite rows: %w", err)
}
if rows == 0 {
return 0, fmt.Errorf("CreateUserWithInvite invite unavailable: %w", ErrNotFound)
}
result, err = tx.Exec(
`INSERT INTO users (username, password, role_id) VALUES (?, ?, ?)`,
username, passwordHash, roleID,
)
if err != nil {
return 0, fmt.Errorf("CreateUserWithInvite create user: %w", err)
}
uid, err := result.LastInsertId()
if err != nil {
return 0, fmt.Errorf("CreateUserWithInvite last insert id: %w", err)
}
if err := tx.Commit(); err != nil {
return 0, fmt.Errorf("CreateUserWithInvite commit: %w", err)
}
committed = true
return uid, nil
}
// GetUserByUsername returns the user with the given username (case-insensitive),
// or nil if not found.
func (d *DB) GetUserByUsername(ctx context.Context, username string) (*User, error) {
u, err := d.q.GetUserByUsername(ctx, username)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
if err != nil {
return nil, fmt.Errorf("GetUserByUsername: %w", err)
}
return userFromGen(u), nil
}
// GetUserByID returns the user with the given ID, or nil if not found.
func (d *DB) GetUserByID(ctx context.Context, id int64) (*User, error) {
u, err := d.q.GetUserByID(ctx, id)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
if err != nil {
return nil, fmt.Errorf("GetUserByID: %w", err)
}
return userFromGen(u), nil
}
// UpdateUserStatus sets the status column for the given user ID.
func (d *DB) UpdateUserStatus(ctx context.Context, id int64, status string) error {
if err := d.q.UpdateUserStatus(ctx, dbgen.UpdateUserStatusParams{
Status: status,
ID: id,
}); err != nil {
return fmt.Errorf("UpdateUserStatus: %w", err)
}
return nil
}
// UpdateUserTOTPSecret sets or clears the TOTP secret for a user.
func (d *DB) UpdateUserTOTPSecret(ctx context.Context, id int64, secret *string) error {
if err := d.q.UpdateUserTOTPSecret(ctx, dbgen.UpdateUserTOTPSecretParams{
TotpSecret: secret,
ID: id,
}); err != nil {
return fmt.Errorf("UpdateUserTOTPSecret: %w", err)
}
return nil
}
// UpdateUserIdentityKey sets or clears the E2EE identity public key for a user
// (F3 voice E2EE TOFU). Last write wins; key changes are audited at the
// service layer so peers can detect a rotation.
func (d *DB) UpdateUserIdentityKey(ctx context.Context, id int64, key *string) error {
if err := d.q.UpdateUserIdentityKey(ctx, dbgen.UpdateUserIdentityKeyParams{
IdentityPublicKey: key,
ID: id,
}); err != nil {
return fmt.Errorf("UpdateUserIdentityKey: %w", err)
}
return nil
}
// ResetAllUserStatuses clears the "online" status left behind by a previous
// run or crash. Called on server startup. Chosen statuses (idle/dnd/invisible)
// are left standing — they are what the user picked, not evidence of a session,
// and the read path already renders a user with no live connection as offline.
func (d *DB) ResetAllUserStatuses(ctx context.Context) error {
if err := d.q.ResetAllUserStatuses(ctx); err != nil {
return fmt.Errorf("ResetAllUserStatuses: %w", err)
}
return nil
}
// MarkUserDisconnected records that a user's last session went away: last_seen
// is refreshed and "online" falls back to "offline", while a chosen
// idle/dnd/invisible is preserved for the next connect to honour.
func (d *DB) MarkUserDisconnected(ctx context.Context, userID int64) error {
if err := d.q.MarkUserDisconnected(ctx, userID); err != nil {
return fmt.Errorf("MarkUserDisconnected: %w", err)
}
return nil
}
// BanUser marks a user as banned with an optional expiry. Pass nil for a
// permanent ban.
func (d *DB) BanUser(ctx context.Context, id int64, reason string, expires *time.Time) error {
var expiresStr *string
if expires != nil {
s := expires.UTC().Format("2006-01-02T15:04:05Z")
expiresStr = &s
}
reasonCopy := reason
if err := d.q.BanUser(ctx, dbgen.BanUserParams{
BanReason: &reasonCopy,
BanExpires: expiresStr,
ID: id,
}); err != nil {
return fmt.Errorf("BanUser: %w", err)
}
return nil
}
// UnbanUser removes the ban from a user.
func (d *DB) UnbanUser(ctx context.Context, id int64) error {
if err := d.q.UnbanUser(ctx, id); err != nil {
return fmt.Errorf("UnbanUser: %w", err)
}
return nil
}
// ─── Session Operations ───────────────────────────────────────────────────────
// maxSessionsPerUser is the maximum number of concurrent sessions allowed per
// user. When exceeded, the oldest session is evicted. This prevents unbounded
// session accumulation from credential stuffing or token theft (H-6).
const maxSessionsPerUser = 25
// CreateSession inserts a new session and returns the session ID.
// tokenHash must already be hashed (never store plaintext tokens).
// H-6: Enforces a per-user session cap by evicting the oldest session when
// the limit is reached.
func (d *DB) CreateSession(ctx context.Context, userID int64, tokenHash, device, ip string) (int64, error) {
// Evict oldest sessions if at or above the cap.
_ = d.q.EvictOldestSessions(ctx, dbgen.EvictOldestSessionsParams{
UserID: userID,
Offset: maxSessionsPerUser - 1,
})
expiresAt := time.Now().Add(sessionTTL).UTC().Format(sessionTimeLayout)
deviceCopy, ipCopy := device, ip
res, err := d.q.InsertSession(ctx, dbgen.InsertSessionParams{
UserID: userID,
Token: tokenHash,
Device: &deviceCopy,
IpAddress: &ipCopy,
ExpiresAt: expiresAt,
})
if err != nil {
return 0, fmt.Errorf("CreateSession: %w", err)
}
return res.LastInsertId()
}
// GetSessionByTokenHash retrieves a session by its hashed token, or nil if
// not found.
func (d *DB) GetSessionByTokenHash(ctx context.Context, tokenHash string) (*Session, error) {
s, err := d.q.GetSessionByTokenHash(ctx, tokenHash)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
if err != nil {
return nil, fmt.Errorf("GetSessionByTokenHash: %w", err)
}
sess := sessionFromGen(s)
return &sess, nil
}
// SessionWithBanStatus combines session data with user ban fields
// in a single query, avoiding two sequential DB round-trips.
type SessionWithBanStatus struct {
Session
Banned bool
BanReason *string
BanExpires *string
}
// GetSessionWithBanStatus returns the session joined with the user's ban
// status in a single query. Returns nil, nil when not found.
func (d *DB) GetSessionWithBanStatus(ctx context.Context, tokenHash string) (*SessionWithBanStatus, error) {
row, err := d.q.GetSessionWithBanStatus(ctx, tokenHash)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
if err != nil {
return nil, fmt.Errorf("GetSessionWithBanStatus: %w", err)
}
return &SessionWithBanStatus{
Session: Session{
ID: row.ID,
UserID: row.UserID,
TokenHash: row.Token,
Device: derefString(row.Device),
IP: derefString(row.IpAddress),
CreatedAt: row.CreatedAt,
LastUsed: row.LastUsed,
ExpiresAt: row.ExpiresAt,
},
Banned: row.Banned != 0,
BanReason: row.BanReason,
BanExpires: row.BanExpires,
}, nil
}
// GetSessionsWithBanStatusBatch returns session+ban rows for every token hash
// in tokenHashes, keyed by token hash. Hashes with no session row are simply
// absent from the map. Used by the ws revoked-session sweep so N connected
// clients cost one query per sweep instead of N.
func (d *DB) GetSessionsWithBanStatusBatch(ctx context.Context, tokenHashes []string) (map[string]*SessionWithBanStatus, error) {
result := make(map[string]*SessionWithBanStatus, len(tokenHashes))
// Chunk the IN list to stay far below SQLite's bound-parameter limit.
const chunkSize = 500
for start := 0; start < len(tokenHashes); start += chunkSize {
chunk := tokenHashes[start:min(start+chunkSize, len(tokenHashes))]
placeholders := make([]string, len(chunk))
args := make([]any, len(chunk))
for i, hash := range chunk {
placeholders[i] = "?"
args[i] = hash
}
query := fmt.Sprintf( //nolint:gosec // G201: placeholder interpolation, not user input
`SELECT s.id, s.user_id, s.token, s.device, s.ip_address,
s.created_at, s.last_used, s.expires_at,
u.banned, u.ban_reason, u.ban_expires
FROM sessions s
JOIN users u ON s.user_id = u.id
WHERE s.token IN (%s)`,
strings.Join(placeholders, ","),
)
rows, err := d.reader.QueryContext(ctx, query, args...)
if err != nil {
return nil, fmt.Errorf("GetSessionsWithBanStatusBatch: %w", err)
}
if err := scanSessionsWithBanStatus(rows, result); err != nil {
return nil, err
}
}
return result, nil
}
// scanSessionsWithBanStatus scans batch rows into result and closes rows.
func scanSessionsWithBanStatus(rows *sql.Rows, result map[string]*SessionWithBanStatus) error {
defer rows.Close() //nolint:errcheck
for rows.Next() {
var s SessionWithBanStatus
var device, ip *string
var banned int
if err := rows.Scan(
&s.ID, &s.UserID, &s.TokenHash, &device, &ip,
&s.CreatedAt, &s.LastUsed, &s.ExpiresAt,
&banned, &s.BanReason, &s.BanExpires,
); err != nil {
return fmt.Errorf("GetSessionsWithBanStatusBatch scan: %w", err)
}
s.Device = derefString(device)
s.IP = derefString(ip)
s.Banned = banned != 0
result[s.TokenHash] = &s
}
if err := rows.Err(); err != nil {
return fmt.Errorf("GetSessionsWithBanStatusBatch rows: %w", err)
}
return nil
}
// DeleteSession removes the session with the given token hash.
func (d *DB) DeleteSession(ctx context.Context, tokenHash string) error {
if err := d.q.DeleteSessionByToken(ctx, tokenHash); err != nil {
return fmt.Errorf("DeleteSession: %w", err)
}
return nil
}
// DeleteOtherSessions removes all sessions for the given user except the one
// with keepSessionID. Used after password change or 2FA state change to
// invalidate all other sessions (BUG-108).
func (d *DB) DeleteOtherSessions(ctx context.Context, userID, keepSessionID int64) (int64, error) {
result, err := d.q.DeleteOtherSessions(ctx, dbgen.DeleteOtherSessionsParams{
UserID: userID,
ID: keepSessionID,
})
if err != nil {
return 0, fmt.Errorf("DeleteOtherSessions: %w", err)
}
n, _ := result.RowsAffected()
return n, nil
}
// DeleteExpiredSessions removes all sessions whose expires_at is in the past.
// The comparison is plain text against idx_sessions_expires_at, so the cutoff
// MUST use sessionTimeLayout — the exact stored format (migration 031
// normalized legacy rows). A space-separated cutoff would compare wrong and
// silently delete nothing (' ' sorts before 'T').
func (d *DB) DeleteExpiredSessions(ctx context.Context) error {
cutoff := time.Now().UTC().Format(sessionTimeLayout)
if err := d.q.DeleteExpiredSessions(ctx, cutoff); err != nil {
return fmt.Errorf("DeleteExpiredSessions: %w", err)
}
return nil
}
// TouchSession updates last_used for the session with the given token hash.
func (d *DB) TouchSession(ctx context.Context, tokenHash string) error {
if err := d.q.TouchSession(ctx, tokenHash); err != nil {
return fmt.Errorf("TouchSession: %w", err)
}
return nil
}
// ─── Invite Operations ────────────────────────────────────────────────────────
// CreateInvite generates a random invite code, persists it, and returns the
// code. maxUses=0 means unlimited. expiresAt=nil means never expires.
func (d *DB) CreateInvite(ctx context.Context, createdBy int64, maxUses int, expiresAt *time.Time) (string, error) {
code, err := generateInviteCode()
if err != nil {
return "", fmt.Errorf("CreateInvite generate code: %w", err)
}
var maxUsesVal *int
if maxUses > 0 {
maxUsesVal = &maxUses
}
var expiresStr *string
if expiresAt != nil {
s := expiresAt.UTC().Format("2006-01-02T15:04:05Z")
expiresStr = &s
}
if err := d.q.CreateInvite(ctx, dbgen.CreateInviteParams{
Code: code,
CreatedBy: createdBy,
MaxUses: ptrItoI64(maxUsesVal),
ExpiresAt: expiresStr,
}); err != nil {
return "", fmt.Errorf("CreateInvite insert: %w", err)
}
return code, nil
}
// GetInvite returns the invite for the given code, or nil if not found.
func (d *DB) GetInvite(ctx context.Context, code string) (*Invite, error) {
r, err := d.q.GetInvite(ctx, code)
if errors.Is(err, sql.ErrNoRows) {
return nil, nil
}
if err != nil {
return nil, fmt.Errorf("GetInvite: %w", err)
}
return &Invite{
ID: r.ID,
Code: r.Code,
CreatedBy: r.CreatedBy,
Uses: int(r.UseCount),
MaxUses: ptrI64toI(r.MaxUses),
ExpiresAt: r.ExpiresAt,
Revoked: r.Revoked != 0,
CreatedAt: r.CreatedAt,
}, nil
}
// UseInviteAtomic validates and increments the use_count in a single SQL
// statement, eliminating the TOCTOU race that exists when GetInvite and
// UseInvite are called as separate operations.
//
// The UPDATE only matches rows where:
// - the code exists
// - revoked = 0
// - max_uses IS NULL (unlimited) OR uses < max_uses
// - expires_at IS NULL (never) OR expires_at > now
//
// If zero rows are affected the invite is missing, revoked, expired, or
// exhausted — an error is returned in all such cases.
func (d *DB) UseInviteAtomic(ctx context.Context, code string) error {
result, err := d.q.UseInviteAtomic(ctx, code)
if err != nil {
return fmt.Errorf("UseInviteAtomic: %w", err)
}
rows, err := result.RowsAffected()
if err != nil {
return fmt.Errorf("UseInviteAtomic rows: %w", err)
}
if rows == 0 {
return fmt.Errorf("UseInviteAtomic: invite not found, revoked, expired, or exhausted: %w", ErrNotFound)
}
return nil
}
// RevokeInvite marks an invite as revoked.
func (d *DB) RevokeInvite(ctx context.Context, code string) error {
if err := d.q.RevokeInvite(ctx, code); err != nil {
return fmt.Errorf("RevokeInvite: %w", err)
}
return nil
}
// ─── Helpers ──────────────────────────────────────────────────────────────────
// MemberSummary is a lightweight user shape for the ready payload.
type MemberSummary struct {
ID int64 `json:"id"`
Username string `json:"username"`
Avatar *string `json:"avatar"`
Status string `json:"status"`
Role string `json:"role"`
// IdentityPublicKey is the user's long-term E2EE identity public key
// (base64), pinned by peers on first sight (F3 TOFU). Omitted when the
// user has not published one.
IdentityPublicKey *string `json:"identity_public_key,omitempty"`
// DisplayName is the nickname to render instead of Username. Null when
// unset; clients fall back to Username.
DisplayName *string `json:"display_name"`
// CustomStatus is the free-text status line shown under the name. Null
// when unset.
CustomStatus *string `json:"custom_status"`
}
// ForViewer returns a copy of the summary as viewerID may see it: an invisible
// member is offline to everyone but themselves. Ready payloads go through this
// so "who is invisible" is decided in exactly one place.
func (m MemberSummary) ForViewer(viewerID int64) MemberSummary {
m.Status = StatusForViewer(m.Status, m.ID, viewerID)
// A connected-but-invisible member collapses to "offline" above, but their
// custom_status is a separate column BroadcastStatus never touches. Left
// alone, a viewer sees {status:"offline", custom_status:"<text>"} for that
// member while every genuinely disconnected member is
// {status:"offline", custom_status:null} — the surviving text is a tell
// that the member is actually online. Blank it at the same choke point
// that collapses the status so no other caller has to remember to.
if m.ID != viewerID && m.Status == StatusOffline {
m.CustomStatus = nil
}
return m
}
// ListMembers returns non-banned users as lightweight summaries.
// M-12: Limited to 1000 rows to prevent unbounded result sets on large servers.
func (d *DB) ListMembers(ctx context.Context) ([]MemberSummary, error) {
rows, err := d.q.ListMembers(ctx)
if err != nil {
return nil, fmt.Errorf("ListMembers: %w", err)
}
members := make([]MemberSummary, 0, len(rows))
for _, r := range rows {
members = append(members, MemberSummary{
ID: r.ID,
Username: r.Username,
Avatar: r.Avatar,
Status: r.Status,
Role: r.Lower,
IdentityPublicKey: r.IdentityPublicKey,
DisplayName: r.DisplayName,
CustomStatus: r.CustomStatus,
})
}
return members, nil
}
// generateInviteCode produces a random 8-byte (16-char hex) code.
func generateInviteCode() (string, error) {
b := make([]byte, 8)
if _, err := rand.Read(b); err != nil {
return "", err
}
return hex.EncodeToString(b), nil
}