Let an operator authenticate with a seeded key, and reset the schema

- TERDUT_OPERATOR_KEY creates or re-keys the instance-scoped service account
  "terdut-operator" at every start, so terdut-operator needs no bootstrap
  handshake. An instance-scoped account now acts as owner of every team's
  configuration, but is not a member of any team.
- POST /api/teams takes an external_id (instance service accounts only) and
  is idempotent on it, so automation finds its own team again after a crash
  instead of adopting by display name. GET /api/teams?name= is removed.
- Integration and dead man's switch names are unique per team (409). The
  escalation PUT accepts usernames and resolves them itself.
- The 18 migrations are squashed into 001_schema.sql, with no Default team.
  TERDUT_DEADMAN_* and the env seeding of switches are removed: teams carry
  their own. Existing development databases must be recreated.

Security and robustness:
- GET /api/users no longer returns other people's email or ntfy topic to
  non-admins.
- The access log records the route pattern, so integration keys and ack
  tokens in the path are not written to the log. Server errors are logged.
- Rate limits take the client address TERDUT_TRUSTED_PROXIES hops from the
  right of X-Forwarded-For instead of trusting the first, forgeable entry.
- /api/bootstrap runs in a transaction under an advisory lock, so two
  concurrent calls cannot both create an administrator.
- API key last_used_at is written at most every five minutes.

Cleanup: remove GET /api/incidents/{id}/alerts, unused exports, SQLite
remnants in comments and config.

Claude-Session: https://claude.ai/code/session_016mBLURvJoMuUEr9cB2RpUN

.gitea/workflows/ci.yaml

@@ -46,8 +46,8 @@ jobs:
         - go-build-cache:/root/.cache/go-build
         - gobin-cache:/go/bin
 
-    # The suite needs a real Postgres -- there is no in-memory Postgres the way there was
-    # an in-memory SQLite, so each test gets its own schema on a shared server instead.
+    # The suite needs a real Postgres -- there is no in-memory Postgres,
+    # so each test gets its own schema on a shared server instead.
     # The job and the service share the dind bridge, so the service is reachable by its
     # name rather than on localhost.
     services:

.gitignore

@@ -7,8 +7,6 @@
 # one (which has an unreachable entry) cannot abort a release
 /.helm-repos.yaml
 
-# SQLite database files
-*.db
 *.db-shm
 *.db-wal
 

Makefile

@@ -28,7 +28,7 @@ help: ## Show this help
 # -race below. Both need a Postgres to test against; see test-db.
 
 # The suite needs a Postgres, because the server does: there is no in-memory
-# Postgres the way there was an in-memory SQLite. TERDUT_TEST_DSN says where, and
+# Postgres. TERDUT_TEST_DSN says where, and
 # the tests fail rather than skip without it — a suite that quietly tests nothing
 # is worse than one that does not run. `make test-db` starts a local one;
 # ci.yaml runs the same thing as a service container.

charts/terdut-server/templates/deployment.yaml

@@ -95,12 +95,6 @@ spec:
               value: "{{ .Values.sweeper.staleAfter }}"
             - name: TERDUT_ARCHIVE_AFTER
               value: "{{ .Values.sweeper.archiveAfter }}"
-            - name: TERDUT_DEADMAN_MATCHERS
-              value: "{{ .Values.deadman.matchers }}"
-            - name: TERDUT_DEADMAN_TIMEOUT
-              value: "{{ .Values.deadman.timeout }}"
-            - name: TERDUT_DEADMAN_SEVERITY
-              value: "{{ .Values.deadman.severity }}"
             {{- if .Values.notify.ntfyUrl }}
             - name: TERDUT_NTFY_URL
               value: "{{ .Values.notify.ntfyUrl }}"
@@ -122,6 +116,8 @@ spec:
               value: "{{ .Values.notify.publicUrl | default (printf "https://%s" .Values.networking.hostname) }}"
             - name: TERDUT_PASSWORD_LOGIN
               value: {{ .Values.passwordLogin | quote }}
+            - name: TERDUT_TRUSTED_PROXIES
+              value: {{ .Values.trustedProxies | quote }}
             - name: TERDUT_OPERATOR_MODE
               value: {{ .Values.operatorMode | quote }}
             {{- if .Values.oidc.enabled }}

charts/terdut-server/values.yaml

@@ -72,43 +72,10 @@ sweeper:
   # How long a resolved alert stays in the default list before auto-archiving.
   archiveAfter: 168h
 
-# Alerts treated as dead man's switches: receiving one opens no incident, and
-# the absence of one does. The Watchdog alert kube-prometheus-stack ships is
-# exactly this — an always-firing alert whose only value is something noticing
-# when it stops.
-deadman:
-  # Which alerts to treat as heartbeats. ";" separates matchers, "," separates
-  # the label conditions within one, "=" is exact equality. Every matcher must
-  # name an alertname:
-  #   alertname=Watchdog,cluster=prod; alertname=EdgeHeartbeat
-  # Each distinct label set is watched independently, so two clusters sending
-  # the same alertname are two switches and a live one cannot mask a dead one.
-  matchers: "alertname=Watchdog"
-  # How long a heartbeat may go unheard before its switch is declared dead.
-  #
-  # This must be SHORTER than the Alertmanager repeat_interval of the route
-  # carrying the heartbeat — the opposite of sweeper.staleAfter. The default
-  # repeat_interval of 4h (12h in many setups) makes for a useless dead man's
-  # switch, so give the heartbeat a route of its own:
-  #
-  #   - matchers: [ 'alertname = "Watchdog"' ]
-  #     receiver: terdut
-  #     group_wait: 0s
-  #     group_interval: 1m
-  #     repeat_interval: 1m
-  #
-  # That delivers every 2m rather than every 1m: a group is only reconsidered
-  # each group_interval, and at exactly one elapsed interval repeat_interval has
-  # not quite passed, so equal values give 2x. Fine against 15m; use
-  # group_interval: 30s if you want a true 1m.
-  #
-  # Set to 0 to disable dead man's switch handling entirely.
-  timeout: 15m
-  # Severity a dead man's switch incident opens at. These incidents have no
-  # member alerts to derive one from, and the heartbeat's own severity label is
-  # meaningless — Watchdog ships as "none". Only "critical" maps to the ntfy
-  # priority that overrides a phone's quiet hours.
-  severity: critical
+# How many reverse proxies in front of the server append to X-Forwarded-For.
+# The per-address login/sign-up rate limits take the client address that many
+# entries from the right. 0 ignores the header.
+trustedProxies: 1
 
 notify:
   # ntfy server that push notifications are published to, e.g.
@@ -190,10 +157,8 @@ oidc:
   # Hard ceiling on a session made by a single sign-on login.
   sessionMaxAge: 12h
 
-# Backups are no longer this chart's business. The SQLite database lived on a PVC
-# beside the app, so it needed a sidecar with a sqlite3 module for k8up to exec a
-# dump in; Postgres is backed up where it runs, through a k8up.io/backupcommand
-# pg_dump annotation on the database pod itself.
+# Backups are not this chart's business: Postgres is backed up where it runs,
+# through a k8up.io/backupcommand pg_dump annotation on the database pod itself.
 
 bootstrap:
   enabled: true

cmd/terdut/main.go

@@ -39,21 +39,16 @@ func main() {
 		RepeatEvery:   cfg.NotifyRepeat,
 	}
 
-	// Dead man's switches live per team now. The environment variables are the
-	// defaults a team starts from: every team without a configuration of its
-	// own gets one from them here, and an owner's later edit is never
-	// overwritten by a redeploy.
-	deadman := api.ParseDeadmanConfig(cfg.DeadmanMatchers, cfg.DeadmanTimeout, cfg.DeadmanSeverity)
-	if err := api.SeedDeadmanConfigs(context.Background(), database, deadman); err != nil {
-		log.Fatalf("seed dead man's switch defaults: %v", err)
-	}
-
 	// The behaviour knobs move into the database on first start, after which an
 	// administrator owns them and a redeploy leaves them alone.
 	if err := api.SeedSettings(context.Background(), database, cfg); err != nil {
 		log.Fatalf("seed settings: %v", err)
 	}
 
+	if err := api.SeedOperatorKey(context.Background(), database, cfg.OperatorKey); err != nil {
+		log.Fatalf("%v", err)
+	}
+
 	router := api.NewRouter(database, notify, cfg, version)
 
 	srv := &http.Server{

internal/api/alertmanager.go

@@ -87,7 +87,7 @@ func handleIntegrationWebhook(db *sql.DB, notify NotifyConfig) http.HandlerFunc
 				respond(w, http.StatusUnauthorized, errResp("unknown integration key"))
 				return
 			}
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		receiveWebhook(w, r, db, notify, src)

internal/api/alerts.go

@@ -90,7 +90,7 @@ func handleListAlerts(db *sql.DB) http.HandlerFunc {
 			fmt.Sprintf("%s WHERE %s ORDER BY a.received_at DESC LIMIT %s", alertSelectFrom, clause, args.add(limit)),
 			args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -99,7 +99,7 @@ func handleListAlerts(db *sql.DB) http.HandlerFunc {
 		for rows.Next() {
 			a, err := scanAlert(rows)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			alerts = append(alerts, a)
@@ -121,7 +121,7 @@ func handleGetAlert(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, a)

internal/api/api_test.go

@@ -78,6 +78,10 @@ func newTSWith(t *testing.T, deadman api.DeadmanConfig, cfg api.NotifyConfig, co
 
 	s := &ts{Server: srv, key: key, db: database, notify: cfg, deadman: deadman}
 
+	// A fresh install has no team, so the tests that want "the" team make it
+	// here: it is id 1, owned by the admin, which is what defaultTeam names.
+	decode(t, s.req(t, http.MethodPost, "/api/teams", map[string]string{"name": "Default"}), &struct{}{})
+
 	var integration struct {
 		Key string `json:"key"`
 	}
@@ -768,7 +772,7 @@ func TestWebhook_IgnoresOutOfOrderRetry(t *testing.T) {
 // An expiry resolve writes ends_at as an upper bound, not an observed end: an
 // Alertmanager watermark already on the row is preserved, and a row that never
 // carried one is stamped at sweep time. Clients are told to read it that way —
-// see "resolution_source says how much to trust ends_at" in the README.
+// see "resolution_source says how much to trust ends_at" in docs/api.md.
 func TestExpiry_EndsAtIsUpperBound(t *testing.T) {
 	s := newTS(t)
 
@@ -813,7 +817,7 @@ func TestExpiry_EndsAtIsUpperBound(t *testing.T) {
 //
 // received_at is documented as a public liveness signal, so these lock the
 // behaviour clients are told they may rely on. See "received_at is a liveness
-// heartbeat" in the README and the comment on models.Alert.ReceivedAt.
+// heartbeat" in docs/api.md and the comment on models.Alert.ReceivedAt.
 // ---------------------------------------------------------------------------
 
 // The heartbeat itself: an unchanged firing notification — what Alertmanager
@@ -875,3 +879,38 @@ func TestStats_ByDayReturnsSevenSlots(t *testing.T) {
 		t.Errorf("expected 7 day slots, got %d", len(slots))
 	}
 }
+
+// Two simultaneous bootstraps on an empty install must not both win.
+func TestBootstrap_ConcurrentCallsCreateOneAdmin(t *testing.T) {
+	database := newTestDB(t)
+	srv := httptest.NewServer(api.NewRouter(database, api.NotifyConfig{}, testConfig(), "test"))
+	t.Cleanup(srv.Close)
+
+	const n = 8
+	codes := make(chan int, n)
+	for i := 0; i < n; i++ {
+		go func(i int) {
+			body, _ := json.Marshal(map[string]string{"username": fmt.Sprintf("u%d", i), "email": fmt.Sprintf("u%d@x.com", i)})
+			resp, err := http.Post(srv.URL+"/api/bootstrap", "application/json", bytes.NewReader(body))
+			if err != nil {
+				codes <- 0
+				return
+			}
+			resp.Body.Close()
+			codes <- resp.StatusCode
+		}(i)
+	}
+	created := 0
+	for i := 0; i < n; i++ {
+		if <-codes == http.StatusCreated {
+			created++
+		}
+	}
+	var users int
+	if err := database.QueryRow("SELECT COUNT(*) FROM users").Scan(&users); err != nil {
+		t.Fatal(err)
+	}
+	if created != 1 || users != 1 {
+		t.Errorf("expected exactly one bootstrap to win, got %d created and %d users", created, users)
+	}
+}

internal/api/archiver.go

@@ -154,9 +154,7 @@ func expireStale(ctx context.Context, db *sql.DB, staleAfter time.Duration, skip
 }
 
 // staleAlertIDs reads the ids in one go and closes the cursor before the caller
-// writes. Under SQLite's single connection an open read would have blocked the
-// update outright; with a pool it is no longer a deadlock, but reading the set
-// first still keeps the write off a cursor the same transaction is walking.
+// writes, which keeps the write off a cursor the same transaction is walking.
 func staleAlertIDs(ctx context.Context, db *sql.DB, now time.Time, staleAfter time.Duration) ([]int64, error) {
 	rows, err := db.QueryContext(ctx, `
 		SELECT id FROM alerts

internal/api/auth.go

@@ -10,6 +10,7 @@ import (
 	"strconv"
 	"strings"
 	"sync"
+	"sync/atomic"
 	"time"
 
 	"github.com/go-chi/chi/v5"
@@ -28,6 +29,9 @@ const (
 	// sessionTouchEvery bounds how often a request may slide the expiry.
 	sessionTouchEvery = time.Hour
 
+	// keyTouchEvery is the same bound for an API key's last_used_at.
+	keyTouchEvery = 5 * time.Minute
+
 	minPasswordLen = 10
 	// maxPasswordLen is bcrypt's limit; it rejects longer input outright.
 	maxPasswordLen = 72
@@ -126,14 +130,32 @@ func purgeRateLimits(ctx context.Context, db *sql.DB) {
 	}
 }
 
+// trustedProxies is how many X-Forwarded-For hops clientAddr trusts. Set once
+// by NewRouter from config.
+var trustedProxies atomic.Int32
+
 // clientAddr is the address a login is counted against. Behind the gateway
-// RemoteAddr is the gateway itself, so the first X-Forwarded-For hop is used
-// when present. It can be forged, but only to dodge the address limit; the
-// per-username limit does not depend on it.
+// RemoteAddr is the gateway itself, so the client address is read from
+// X-Forwarded-For, counting trustedProxies entries from the right: each trusted
+// proxy appends the address it saw, so the entries to the left of those are
+// client-supplied and could be forged to dodge the limit.
 func clientAddr(r *http.Request) string {
-	if xff := r.Header.Get("X-Forwarded-For"); xff != "" {
-		first, _, _ := strings.Cut(xff, ",")
-		return strings.TrimSpace(first)
+	if n := int(trustedProxies.Load()); n > 0 {
+		var hops []string
+		for _, v := range r.Header.Values("X-Forwarded-For") {
+			for _, h := range strings.Split(v, ",") {
+				if h = strings.TrimSpace(h); h != "" {
+					hops = append(hops, h)
+				}
+			}
+		}
+		if len(hops) > 0 {
+			i := len(hops) - n
+			if i < 0 {
+				i = 0
+			}
+			return hops[i]
+		}
 	}
 	host, _, err := net.SplitHostPort(r.RemoteAddr)
 	if err != nil {
@@ -241,7 +263,7 @@ func handleLogin(db *sql.DB, limiter *loginLimiter, publicURL string) http.Handl
 			"SELECT id, password_hash FROM users WHERE username = $1", username,
 		).Scan(&userID, &hash)
 		if err != nil && !errors.Is(err, sql.ErrNoRows) {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -258,13 +280,13 @@ func handleLogin(db *sql.DB, limiter *loginLimiter, publicURL string) http.Handl
 		limiter.clear(r.Context(), userKey)
 
 		if err := startSession(w, r, db, userID, publicURL); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		user, err := fetchUser(r.Context(), db, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, meResponse{User: user, HasPassword: true})
@@ -320,7 +342,7 @@ func handleMe(db *sql.DB) http.HandlerFunc {
 		}
 		user, err := fetchUser(r.Context(), db, caller.ID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		var hash sql.NullString
@@ -332,7 +354,7 @@ func handleMe(db *sql.DB) http.HandlerFunc {
 			// transient database problem, not a missing user — worth a 500
 			// rather than silently answering "no password, not dismissed",
 			// which a client would otherwise take at face value.
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, meResponse{
@@ -384,7 +406,7 @@ func handleSetPassword(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -397,30 +419,30 @@ func handleSetPassword(db *sql.DB) http.HandlerFunc {
 
 		hash, err := hashPassword(req.Password)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback()
 
 		if _, err := tx.ExecContext(r.Context(),
 			"UPDATE users SET password_hash = $1 WHERE id = $2", hash, id); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		keep, _ := sessionFromContext(r.Context()) // zero when changed with an API key
 		if _, err := tx.ExecContext(r.Context(),
 			"DELETE FROM sessions WHERE user_id = $1 AND id != $2", id, keep); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)

internal/api/caller.go

@@ -2,7 +2,6 @@ package api
 
 import (
 	"context"
-	"fmt"
 
 	"git.ryuvia.com/niklas/terdut-server/internal/models"
 )
@@ -61,7 +60,7 @@ func (c Caller) IsAdmin() bool {
 // by being a human (and becomes its owner as a side effect), an
 // instance-scoped service account creates one with no human owner at all;
 // the two paths are not interchangeable, so this predicate must not also
-// admit a human admin the way MayActAsInstanceAdmin deliberately does.
+// admit a human admin the way an administrator check would.
 func (c Caller) IsInstanceServiceAccount() bool {
 	return c.sa != nil && c.sa.scope == models.ServiceAccountScopeInstance
 }
@@ -110,24 +109,6 @@ func (c Caller) ServiceAccountName() (string, bool) {
 	return c.sa.name, true
 }
 
-// Identity is a stable, log/audit-facing string distinguishing a human
-// caller from a service account — "user:42" or "service-account:7". Not
-// wired into any database column — incidents.go's acknowledged_by/
-// incident_events.user_id use AsHuman()/ServiceAccountID() directly against
-// the parallel *_service_account_id columns (migration 015) instead, since a
-// column needs the id, not this rendered string. assigned_to stays
-// human-only and out of scope (terdut-server#25's follow-up).
-func (c Caller) Identity() string {
-	switch {
-	case c.user != nil:
-		return fmt.Sprintf("user:%d", c.user.ID)
-	case c.sa != nil:
-		return fmt.Sprintf("service-account:%d", c.sa.id)
-	default:
-		return "unknown"
-	}
-}
-
 func callerFromContext(ctx context.Context) (Caller, bool) {
 	c, ok := ctx.Value(ctxCaller).(Caller)
 	return c, ok

internal/api/deadman.go

@@ -65,28 +65,6 @@ func (m DeadmanMatcher) matches(labels map[string]string) bool {
 	return true
 }
 
-// DeadmanConfig is the server-wide default a team's switches are seeded from:
-// the environment's matchers, timeout and severity. Switches themselves are rows
-// of a team's own — see DeadmanSwitch — and this is only how a fresh install
-// starts out.
-type DeadmanConfig struct {
-	Matchers []DeadmanMatcher
-
-	// Timeout is how long a matched alert may go without a refreshing webhook
-	// before it is declared dead. It must be shorter than Alertmanager's
-	// repeat_interval for the heartbeat's route, which is what refreshes it.
-	// Zero disables dead man's switch handling entirely.
-	Timeout time.Duration
-
-	// Severity is the severity every dead man's switch incident opens at. These
-	// incidents have no member alerts to derive one from, and the heartbeat's
-	// own severity label is meaningless — Watchdog ships as "none".
-	Severity string
-}
-
-// enabled reports whether there is anything to watch.
-func (c DeadmanConfig) enabled() bool { return c.Timeout > 0 && len(c.Matchers) > 0 }
-
 // DeadmanSwitch inverts the handling of the alerts it matches: receiving one
 // opens nothing, and the absence of one opens an incident.
 //
@@ -160,46 +138,6 @@ func parseDeadmanMatcher(entry string) (DeadmanMatcher, error) {
 	return m, nil
 }
 
-// ParseDeadmanConfig reads the matcher list from its configured form:
-// ";" separates matchers, and each is parsed as parseDeadmanMatcher does.
-//
-// A malformed or alertname-less entry is dropped rather than fatal, following
-// config.duration's rule that one bad tuning knob should not take the server
-// down. Silence would be worse here than elsewhere, though — a typo that
-// disarms the switch is exactly the failure this feature exists to catch — so
-// the matchers that survived are logged.
-func ParseDeadmanConfig(matchers string, timeout time.Duration, severity string) DeadmanConfig {
-	cfg := DeadmanConfig{Timeout: timeout, Severity: severity}
-
-	for _, entry := range strings.Split(matchers, ";") {
-		entry = strings.TrimSpace(entry)
-		if entry == "" {
-			continue
-		}
-		m, err := parseDeadmanMatcher(entry)
-		if err != nil {
-			log.Printf("deadman: ignoring matcher %q: %v", entry, err)
-			continue
-		}
-		cfg.Matchers = append(cfg.Matchers, m)
-	}
-
-	switch {
-	case timeout <= 0:
-		log.Print("deadman: disabled (timeout is zero)")
-	case len(cfg.Matchers) == 0:
-		log.Print("deadman: disabled (no usable matchers)")
-	default:
-		rendered := make([]string, 0, len(cfg.Matchers))
-		for _, m := range cfg.Matchers {
-			rendered = append(rendered, m.String())
-		}
-		log.Printf("deadman: default for new teams: %s, timeout %s, severity %s",
-			strings.Join(rendered, "; "), timeout, severity)
-	}
-	return cfg
-}
-
 // deadmanAlert is one heartbeat: the alert row carrying its last sighting, and
 // the switch that claimed it.
 type deadmanAlert struct {
@@ -490,56 +428,6 @@ func deadmanSets(ctx context.Context, db *sql.DB) (map[int64]deadmanSet, error)
 	return scanDeadmanSwitches(rows)
 }
 
-// deadmanSeededKey is the settings row that records the environment defaults
-// were handed out. Without it, a team that deleted its last switch would get
-// the default back on the next restart.
-const deadmanSeededKey = "deadman_seeded"
-
-// SeedDeadmanConfigs gives every team the server's environment defaults as
-// switches, exactly once per install, so a fresh install watches Watchdog
-// without anybody setting it up.
-//
-// Once seeded it never runs again: a team's switches are its own, and a redeploy
-// must not quietly put the environment's value back over an owner's edit or
-// deletion. Installs that upgraded from per-team configuration were already
-// seeded, which migration 009 records.
-//
-// A team created after that gets none and watches nothing until its owner says
-// otherwise. That is deliberate: inheriting an install-wide heartbeat would page
-// a new team about a source it has never heard of, and a switch nobody chose is
-// the kind that gets muted rather than fixed.
-func SeedDeadmanConfigs(ctx context.Context, db *sql.DB, cfg DeadmanConfig) error {
-	if !cfg.enabled() {
-		return nil
-	}
-
-	tx, err := db.BeginTx(ctx, nil)
-	if err != nil {
-		return err
-	}
-	defer tx.Rollback() //nolint:errcheck
-
-	res, err := tx.ExecContext(ctx,
-		"INSERT INTO settings (key, value) VALUES ($1, '1') ON CONFLICT (key) DO NOTHING",
-		deadmanSeededKey)
-	if err != nil {
-		return err
-	}
-	if n, _ := res.RowsAffected(); n == 0 {
-		return nil
-	}
-
-	for _, m := range cfg.Matchers {
-		if _, err := tx.ExecContext(ctx, `
-			INSERT INTO deadman_switches (team_id, name, matcher, timeout_seconds, severity)
-			SELECT id, $1, $1, $2, $3 FROM teams`,
-			m.config(), int64(cfg.Timeout.Seconds()), cfg.Severity); err != nil {
-			return err
-		}
-	}
-	return tx.Commit()
-}
-
 // ---------------------------------------------------------------------------
 // Status
 // ---------------------------------------------------------------------------

internal/api/deadman_test.go

@@ -1,7 +1,6 @@
 package api_test
 
 import (
-	"context"
 	"net/http"
 	"strings"
 	"testing"
@@ -122,8 +121,11 @@ func TestDeadman_MixedGroupExcludesHeartbeat(t *testing.T) {
 		t.Fatalf("expected 1 incident for the real alert, got %d", got)
 	}
 
-	var alerts []map[string]any
-	decode(t, s.req(t, http.MethodGet, "/api/incidents/1/alerts", nil), &alerts)
+	var incident struct {
+		Alerts []map[string]any `json:"alerts"`
+	}
+	decode(t, s.req(t, http.MethodGet, "/api/incidents/1", nil), &incident)
+	alerts := incident.Alerts
 	if len(alerts) != 1 {
 		t.Fatalf("expected 1 member alert, got %d", len(alerts))
 	}
@@ -717,26 +719,3 @@ func TestDeadman_DeleteIsScopedToTheTeam(t *testing.T) {
 		t.Errorf("expected no switches, got %d", got)
 	}
 }
-
-// The environment's defaults are handed out once and then belong to the teams.
-func TestDeadman_SeedRunsOnce(t *testing.T) {
-	s := newTS(t)
-	cfg := api.ParseDeadmanConfig("alertname=Watchdog", time.Hour, "critical")
-
-	if err := api.SeedDeadmanConfigs(context.Background(), s.db, cfg); err != nil {
-		t.Fatalf("seed: %v", err)
-	}
-	if got := len(listSwitches(t, s)); got != 1 {
-		t.Fatalf("the first seed should add the default, got %d switches", got)
-	}
-
-	// The owner deletes it; a restart must not put it back.
-	id := int64(listSwitches(t, s)[0]["id"].(float64))
-	s.req(t, http.MethodDelete, "/api/teams/"+defaultTeam+"/deadman/switches/"+id64(id), nil).Body.Close()
-	if err := api.SeedDeadmanConfigs(context.Background(), s.db, cfg); err != nil {
-		t.Fatalf("seed again: %v", err)
-	}
-	if got := len(listSwitches(t, s)); got != 0 {
-		t.Errorf("a second seed resurrected %d switch(es)", got)
-	}
-}

internal/api/device.go

@@ -81,7 +81,7 @@ func handleDeviceStart(db *sql.DB, limiter *loginLimiter, publicURL string) http
 
 		deviceCode, deviceHash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -105,7 +105,7 @@ func handleDeviceStart(db *sql.DB, limiter *loginLimiter, publicURL string) http
 		}
 		if err != nil {
 			log.Printf("device login: start: %v", err)
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -160,7 +160,7 @@ func handleDeviceDecision(db *sql.DB, approve bool) http.HandlerFunc {
 			 WHERE user_code = $3 AND status = 'pending' AND expires_at > $4`,
 			status, caller.ID, code, time.Now().Unix())
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -190,7 +190,7 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback() //nolint:errcheck
@@ -206,7 +206,7 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -226,11 +226,11 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 			}
 			if _, err := tx.ExecContext(r.Context(),
 				"UPDATE device_logins SET last_polled_at = $1 WHERE device_hash = $2", now.Unix(), hash); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if err := tx.Commit(); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			respond(w, http.StatusAccepted, map[string]string{"status": "pending"})
@@ -240,7 +240,7 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 		// Approved. Single use: the row goes before the session is made, so two
 		// racing polls cannot both be given one.
 		if _, err := tx.ExecContext(r.Context(), "DELETE FROM device_logins WHERE device_hash = $1", hash); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		var disabled, sso bool
@@ -252,7 +252,7 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 			return
 		}
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if disabled {
@@ -269,12 +269,12 @@ func handleDeviceToken(db *sql.DB, ssoMaxAge time.Duration, publicURL string) ht
 		}
 		if err := startSessionCapped(w, r, db, userID.Int64, publicURL, maxAge); err != nil {
 			log.Printf("device login: start session: %v", err)
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		user, err := fetchUser(r.Context(), db, userID.Int64)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, meResponse{User: user, HasPassword: false})

internal/api/escalation.go

@@ -3,6 +3,7 @@ package api
 import (
 	"context"
 	"database/sql"
+	"errors"
 	"log"
 	"net/http"
 	"strconv"
@@ -343,12 +344,12 @@ func handleGetEscalation(db *sql.DB) http.HandlerFunc {
 
 		policy, err := loadEscalationPolicy(r.Context(), db, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		view, err := escalationStatus(r.Context(), db, teamID, policy)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, view)
@@ -543,6 +544,10 @@ type escalationLevelJSON struct {
 type escalationTargetJSON struct {
 	Kind   string `json:"kind"`
 	UserID *int64 `json:"user_id,omitempty"`
+	// Username is accepted in place of user_id on a PUT, and resolved to the
+	// id before anything is stored. It is never returned: the stored form is
+	// the id, which survives a rename.
+	Username string `json:"username,omitempty"`
 }
 
 type escalationJSON struct {
@@ -600,6 +605,27 @@ func handleSetEscalation(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		for i, l := range req.Levels {
+			for j, t := range l.Targets {
+				if t.Username == "" {
+					continue
+				}
+				if t.Kind != "user" || t.UserID != nil {
+					respond(w, http.StatusBadRequest, errResp("username belongs on a user target, instead of user_id"))
+					return
+				}
+				var id int64
+				err := db.QueryRowContext(r.Context(), "SELECT id FROM users WHERE username = $1", t.Username).Scan(&id)
+				if errors.Is(err, sql.ErrNoRows) {
+					respond(w, http.StatusBadRequest, errResp("unknown user "+strconv.Quote(t.Username)))
+					return
+				}
+				if err != nil {
+					serverError(w, r, err)
+					return
+				}
+				req.Levels[i].Targets[j].UserID = &id
+				req.Levels[i].Targets[j].Username = ""
+			}
 			if l.TimeoutSeconds <= 0 {
 				respond(w, http.StatusBadRequest, errResp("every level needs a timeout"))
 				return
@@ -632,7 +658,7 @@ func handleSetEscalation(db *sql.DB) http.HandlerFunc {
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback() //nolint:errcheck
@@ -645,13 +671,13 @@ func handleSetEscalation(db *sql.DB) http.HandlerFunc {
 				fallback_topic = excluded.fallback_topic,
 				updated_at = excluded.updated_at`,
 			teamID, req.RepeatCount, req.FallbackTopic); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		// The levels are replaced, not merged; the cascade takes the targets.
 		if _, err := tx.ExecContext(r.Context(),
 			"DELETE FROM escalation_levels WHERE team_id = $1", teamID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -661,7 +687,7 @@ func handleSetEscalation(db *sql.DB) http.HandlerFunc {
 				INSERT INTO escalation_levels (team_id, position, timeout_seconds)
 				VALUES ($1, $2, $3) RETURNING id`,
 				teamID, int64(i+1), l.TimeoutSeconds).Scan(&levelID); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			for _, t := range l.Targets {
@@ -676,13 +702,13 @@ func handleSetEscalation(db *sql.DB) http.HandlerFunc {
 		}
 
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		policy, err := loadEscalationPolicy(r.Context(), db, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, escalationResponse(policy, teamID))

internal/api/escalation_test.go

@@ -502,3 +502,45 @@ func TestEscalation_StatusWithoutALadder(t *testing.T) {
 		t.Errorf("a team with no ladder should read as empty, got %+v", v)
 	}
 }
+
+// A user target may name the person instead of carrying an id; the server
+// resolves it and stores the id.
+func TestEscalation_UserTargetByUsername(t *testing.T) {
+	s := newTS(t)
+	id := teamUser(t, s, "alice", "")
+
+	put := func(username string) *http.Response {
+		return s.req(t, http.MethodPut, "/api/teams/"+defaultTeam+"/escalation", map[string]any{
+			"repeat_count": 0,
+			"levels": []map[string]any{{
+				"timeout_seconds": 300,
+				"targets":         []map[string]any{{"kind": "user", "username": username}},
+			}},
+		})
+	}
+
+	resp := put("alice")
+	resp.Body.Close()
+	if resp.StatusCode >= 300 {
+		t.Fatalf("PUT by username: %d", resp.StatusCode)
+	}
+	var got struct {
+		Levels []struct {
+			Targets []struct {
+				UserID   *int64 `json:"user_id"`
+				Username string `json:"username"`
+			} `json:"targets"`
+		} `json:"levels"`
+	}
+	decode(t, s.req(t, http.MethodGet, "/api/teams/"+defaultTeam+"/escalation", nil), &got)
+	if len(got.Levels) != 1 || len(got.Levels[0].Targets) != 1 ||
+		got.Levels[0].Targets[0].UserID == nil || *got.Levels[0].Targets[0].UserID != id {
+		t.Errorf("expected the target stored as user %d, got %+v", id, got)
+	}
+
+	bad := put("nobody")
+	bad.Body.Close()
+	if bad.StatusCode != http.StatusBadRequest {
+		t.Errorf("unknown username should be a 400, got %d", bad.StatusCode)
+	}
+}

internal/api/export_test.go

@@ -0,0 +1,31 @@
+package api
+
+import (
+	"strings"
+	"time"
+)
+
+// DeadmanConfig is a test fixture only: a set of matchers with one timeout and
+// severity, turned into switches over the API by the test helpers. Production
+// has no server-wide default any more -- switches belong to teams.
+type DeadmanConfig struct {
+	Matchers []DeadmanMatcher
+	Timeout  time.Duration
+	Severity string
+}
+
+// ParseDeadmanConfig reads a ";"-separated matcher list the way the removed
+// environment variable did, dropping malformed entries.
+func ParseDeadmanConfig(matchers string, timeout time.Duration, severity string) DeadmanConfig {
+	cfg := DeadmanConfig{Timeout: timeout, Severity: severity}
+	for _, entry := range strings.Split(matchers, ";") {
+		entry = strings.TrimSpace(entry)
+		if entry == "" {
+			continue
+		}
+		if m, err := parseDeadmanMatcher(entry); err == nil {
+			cfg.Matchers = append(cfg.Matchers, m)
+		}
+	}
+	return cfg
+}

internal/api/helpers.go

@@ -5,6 +5,7 @@ import (
 	"database/sql"
 	"encoding/json"
 	"errors"
+	"github.com/go-chi/chi/v5"
 	"log"
 	"net/http"
 	"strconv"
@@ -17,8 +18,7 @@ import (
 // sqlArgs accumulates query arguments and hands back the placeholder for each.
 //
 // Postgres numbers its placeholders, so a dynamically assembled WHERE clause has
-// to keep its $1, $2, … in step with the order of the values — which SQLite's
-// positional `?` did for free. Handing out the placeholder and storing the value
+// to keep its $1, $2, … in step with the order of the values. Handing out the placeholder and storing the value
 // in one call is what keeps them in step: a filter can be added, removed or
 // reordered without renumbering anything by hand.
 type sqlArgs struct{ vals []any }
@@ -31,8 +31,7 @@ func (a *sqlArgs) add(v any) string {
 
 // addList stores every value and returns their placeholders as "$1, $2, …",
 // ready to drop into an IN (…) clause. Returns an empty string for no values,
-// which no caller should reach: `IN ()` is a syntax error in Postgres as it was
-// in SQLite, so callers check for an empty set before building the query.
+// which no caller should reach: `IN ()` is a syntax error in Postgres, so callers check for an empty set before building the query.
 func (a *sqlArgs) addList(vs []any) string {
 	parts := make([]string, len(vs))
 	for i, v := range vs {
@@ -45,7 +44,7 @@ func (a *sqlArgs) addList(vs []any) string {
 func (a *sqlArgs) all() []any { return a.vals }
 
 // nowEpoch is the SQL expression for "now, as unix seconds", matching how every
-// timestamp in this schema is stored. SQLite spelled it unixepoch().
+// timestamp in this schema is stored.
 //
 // FLOOR, not a bare cast: EXTRACT returns fractional seconds and casting to
 // bigint rounds half up, so a row written at .6 of a second would claim a
@@ -56,11 +55,9 @@ const nowEpoch = "FLOOR(EXTRACT(EPOCH FROM now()))::bigint"
 // isUniqueViolation reports whether err is a broken unique constraint, which
 // callers turn into 409 Conflict rather than 500.
 //
-// Postgres reports it as SQLSTATE 23505 on a typed error; the SQLite driver this
-// replaced only put "UNIQUE constraint failed" in the message, which is why the
-// check used to be a substring match. Matching the code means a renamed
-// constraint or a translated message cannot quietly turn a conflict back into a
-// 500.
+// Postgres reports it as SQLSTATE 23505 on a typed error. Matching the code
+// means a renamed constraint or a translated message cannot quietly turn a
+// conflict back into a 500.
 func isUniqueViolation(err error) bool {
 	var pgErr *pgconn.PgError
 	return errors.As(err, &pgErr) && pgErr.Code == pgerrcode.UniqueViolation
@@ -72,6 +69,18 @@ func respond(w http.ResponseWriter, status int, v any) {
 	json.NewEncoder(w).Encode(v)
 }
 
+// serverError answers 500 and logs why. The response stays opaque, so the log
+// line is the only record of what failed.
+func serverError(w http.ResponseWriter, r *http.Request, err error) {
+	// The route pattern, not the path: two routes carry a credential in it.
+	route := r.URL.Path
+	if rc := chi.RouteContext(r.Context()); rc != nil && rc.RoutePattern() != "" {
+		route = rc.RoutePattern()
+	}
+	log.Printf("%s %s: %v", r.Method, route, err)
+	respond(w, http.StatusInternalServerError, errResp("internal error"))
+}
+
 // maxBodyBytes caps an ordinary JSON request body. 1 MiB is far more than any
 // endpoint below needs — it exists so an unauthenticated caller (signup,
 // login, bootstrap) can't make the server buffer an arbitrarily large body

internal/api/incidents.go

@@ -37,7 +37,7 @@ func handleListClusters(db *sql.DB) http.HandlerFunc {
 				label, strings.Join(where, " AND ")),
 			args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -46,13 +46,13 @@ func handleListClusters(db *sql.DB) http.HandlerFunc {
 		for rows.Next() {
 			var v string
 			if err := rows.Scan(&v); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			clusters = append(clusters, v)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, clusters)
@@ -138,7 +138,7 @@ func handleListIncidents(db *sql.DB) http.HandlerFunc {
 				incidentSelectFrom, strings.Join(where, " AND "), order, args.add(limit)),
 			args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -147,13 +147,13 @@ func handleListIncidents(db *sql.DB) http.HandlerFunc {
 		for rows.Next() {
 			i, err := scanIncident(rows)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			incidents = append(incidents, i)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, incidents)
@@ -172,35 +172,17 @@ func handleGetIncident(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if inc.Alerts, err = incidentAlerts(r, db, id); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, inc)
 	}
 }
 
-func handleIncidentAlerts(db *sql.DB) http.HandlerFunc {
-	return func(w http.ResponseWriter, r *http.Request) {
-		id, ok := incidentIDParam(w, r, db)
-		if !ok {
-			return
-		}
-		if !incidentExists(w, r, db, id) {
-			return
-		}
-		alerts, err := incidentAlerts(r, db, id)
-		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
-			return
-		}
-		respond(w, http.StatusOK, alerts)
-	}
-}
-
 func handleIncidentTimeline(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
 		id, ok := incidentIDParam(w, r, db)
@@ -225,7 +207,7 @@ func handleIncidentTimeline(db *sql.DB) http.HandlerFunc {
 			WHERE  e.incident_id = $1
 			ORDER  BY e.created_at ASC, e.id ASC`, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -239,14 +221,14 @@ func handleIncidentTimeline(db *sql.DB) http.HandlerFunc {
 				&e.ActorUserID, &e.ActorUsername,
 				&e.ActorServiceAccountID, &e.ActorServiceAccountName,
 				&e.AlertID, &e.Detail, &ts); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			e.CreatedAt = time.Unix(ts, 0).UTC()
 			events = append(events, e)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, events)
@@ -262,7 +244,7 @@ func handleIncidentAcknowledge(db *sql.DB) http.HandlerFunc {
 		userID, saID := callerActorIDs(r.Context())
 		acked, err := acknowledgeIncidentAs(r.Context(), db, id, userID, saID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if !acked {
@@ -272,7 +254,7 @@ func handleIncidentAcknowledge(db *sql.DB) http.HandlerFunc {
 			// (acknowledged) already holds.
 			inc, err := fetchIncident(r.Context(), db, id)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if inc.Status == "resolved" {
@@ -300,7 +282,7 @@ func handleIncidentUnacknowledge(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err := logEvent(r.Context(), db, id, evUnacknowledged, userID, saID, nil, nil); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)
@@ -335,16 +317,16 @@ func handleIncidentResolve(db *sql.DB) http.HandlerFunc {
 		}
 		// A person closing an incident is the clearest possible "I have this".
 		if err := stopEscalation(r.Context(), db, id); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if err := logEvent(r.Context(), db, id, evResolved, userID, saID, nil, nil); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if req.Resolution != "" {
 			if err := logEvent(r.Context(), db, id, evResolutionNote, userID, saID, nil, &req.Resolution); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 		}
@@ -385,7 +367,7 @@ func handleIncidentAssign(db *sql.DB) http.HandlerFunc {
 		// the actor_* columns.
 		actorUserID, actorSAID := callerActorIDs(r.Context())
 		if err := logAssignedEvent(r.Context(), db, id, req.UserID, actorUserID, actorSAID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respondIncident(w, r, db, id)
@@ -443,7 +425,7 @@ func handleIncidentSnooze(db *sql.DB) http.HandlerFunc {
 		}
 		detail := until.UTC().Format(time.RFC3339)
 		if err := logEvent(r.Context(), db, id, evSnoozed, userID, saID, nil, &detail); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respondIncident(w, r, db, id)
@@ -462,7 +444,7 @@ func handleIncidentUnsnooze(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err := logEvent(r.Context(), db, id, evUnsnoozed, userID, saID, nil, nil); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)
@@ -478,7 +460,7 @@ func handleIncidentArchive(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"UPDATE incidents SET archived_at = "+nowEpoch+" WHERE id = $1", id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -487,7 +469,7 @@ func handleIncidentArchive(db *sql.DB) http.HandlerFunc {
 		}
 		userID, saID := callerActorIDs(r.Context())
 		if err := logEvent(r.Context(), db, id, evArchived, userID, saID, nil, nil); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respondIncident(w, r, db, id)
@@ -503,7 +485,7 @@ func handleIncidentUnarchive(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"UPDATE incidents SET archived_at = NULL WHERE id = $1", id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -512,7 +494,7 @@ func handleIncidentUnarchive(db *sql.DB) http.HandlerFunc {
 		}
 		userID, saID := callerActorIDs(r.Context())
 		if err := logEvent(r.Context(), db, id, evUnarchived, userID, saID, nil, nil); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)
@@ -557,7 +539,7 @@ func handleCreateNote(db *sql.DB) http.HandlerFunc {
 			VALUES ($1, $2, $3, $4, $5, $6)
 			RETURNING id`, id, noteType, userID, saID, req.Content, now.Unix()).Scan(&eventID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -598,7 +580,7 @@ func handleDeleteNote(db *sql.DB) http.HandlerFunc {
 			   AND (user_id = $5 OR service_account_id = $6)`,
 			eventID, id, evNote, evResolutionNote, userID, saID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -652,7 +634,7 @@ func incidentExists(w http.ResponseWriter, r *http.Request, db *sql.DB, id int64
 func updateOpenIncident(w http.ResponseWriter, r *http.Request, db *sql.DB, id int64, query string, args ...any) bool {
 	res, err := db.ExecContext(r.Context(), query, args...)
 	if err != nil {
-		respond(w, http.StatusInternalServerError, errResp("internal error"))
+		serverError(w, r, err)
 		return false
 	}
 	if n, _ := res.RowsAffected(); n > 0 {
@@ -668,7 +650,7 @@ func updateOpenIncident(w http.ResponseWriter, r *http.Request, db *sql.DB, id i
 func respondIncident(w http.ResponseWriter, r *http.Request, db *sql.DB, id int64) {
 	inc, err := fetchIncident(r.Context(), db, id)
 	if err != nil {
-		respond(w, http.StatusInternalServerError, errResp("internal error"))
+		serverError(w, r, err)
 		return
 	}
 	respond(w, http.StatusOK, inc)

internal/api/incidents_test.go

@@ -794,8 +794,7 @@ func TestStats_Incidents(t *testing.T) {
 }
 
 // An empty window is a report of zero, not a failure. SUM over no rows is NULL
-// in Postgres as it was in SQLite, and that used to come back as a 500 the
-// moment every incident was archived — the state a quiet installation settles
+// in Postgres, and that used to come back as a 500 the moment every incident was archived — the state a quiet installation settles
 // into.
 func TestStats_IncidentsEmptyWindowIsZeroNotAnError(t *testing.T) {
 	s := newTS(t)

internal/api/middleware.go

@@ -5,10 +5,14 @@ import (
 	"crypto/sha256"
 	"database/sql"
 	"encoding/hex"
+	"log"
 	"net/http"
 	"strings"
 	"time"
 
+	"github.com/go-chi/chi/v5"
+	"github.com/go-chi/chi/v5/middleware"
+
 	"git.ryuvia.com/niklas/terdut-server/internal/config"
 	"git.ryuvia.com/niklas/terdut-server/internal/models"
 )
@@ -106,7 +110,7 @@ func securityHeaders(publicURL string) func(http.Handler) http.Handler {
 //
 // 403 and not 404: the route exists and the caller is authenticated, they are
 // simply not allowed. Hiding the endpoint would buy nothing — every one of them
-// is in the README.
+// is in docs/api.md.
 func AdminOnly(next http.Handler) http.Handler {
 	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
 		caller, ok := userFromContext(r.Context())
@@ -141,19 +145,23 @@ func requireSelfOrAdmin(w http.ResponseWriter, r *http.Request, targetID int64)
 // resolve and has to be caught afterwards.
 func apiKeyUser(ctx context.Context, db *sql.DB, token string) (int64, bool) {
 	var keyID, userID int64
+	var lastUsed sql.NullInt64
 	err := db.QueryRowContext(ctx,
-		`SELECT id, user_id FROM api_keys
+		`SELECT id, user_id, last_used_at FROM api_keys
 		  WHERE key_hash = $1 AND (expires_at IS NULL OR expires_at > $2)`,
 		hashToken(token), time.Now().Unix(),
-	).Scan(&keyID, &userID)
+	).Scan(&keyID, &userID, &lastUsed)
 	if err != nil {
 		return 0, false
 	}
 
-	// best-effort; don't fail the request if this update fails
+	// best-effort; don't fail the request if this update fails. Throttled like
+	// the session expiry, so a polling client does not write a row per request.
+	if now := time.Now(); !lastUsed.Valid || now.Sub(time.Unix(lastUsed.Int64, 0)) > keyTouchEvery {
 		db.ExecContext(ctx,
 			"UPDATE api_keys SET last_used_at = $1 WHERE id = $2",
-		time.Now().Unix(), keyID)
+			now.Unix(), keyID)
+	}
 	return userID, true
 }
 
@@ -206,7 +214,7 @@ func serveAs(w http.ResponseWriter, r *http.Request, next http.Handler, db *sql.
 	// table with one row per membership.
 	teams, err := callerMemberships(r.Context(), db, userID)
 	if err != nil {
-		respond(w, http.StatusInternalServerError, errResp("internal error"))
+		serverError(w, r, err)
 		return
 	}
 
@@ -222,10 +230,8 @@ func hashToken(token string) string {
 	return hex.EncodeToString(h[:])
 }
 
-// userFromContext is a thin compatibility wrapper over Caller.AsHuman(), so
-// every call site written before the Caller abstraction (alerts.go,
-// incidents.go, schedule.go, stats.go, and more) needs no change and keeps
-// its exact existing behavior.
+// userFromContext returns the human behind the request, or false for a service
+// account: Caller.AsHuman() on the request's Caller.
 func userFromContext(ctx context.Context) (models.User, bool) {
 	c, _ := callerFromContext(ctx)
 	return c.AsHuman()
@@ -245,13 +251,13 @@ type serviceAccountPrincipal struct {
 func serviceAccountFor(ctx context.Context, db *sql.DB, token string) (serviceAccountPrincipal, bool) {
 	var sa serviceAccountPrincipal
 	var keyID int64
-	var teamID sql.NullInt64
+	var teamID, lastUsed sql.NullInt64
 	err := db.QueryRowContext(ctx, `
-		SELECT k.id, a.id, a.name, a.scope, a.team_id
+		SELECT k.id, a.id, a.name, a.scope, a.team_id, k.last_used_at
 		FROM   service_account_keys k
 		JOIN   service_accounts a ON a.id = k.service_account_id
 		WHERE  k.key_hash = $1`, hashToken(token),
-	).Scan(&keyID, &sa.id, &sa.name, &sa.scope, &teamID)
+	).Scan(&keyID, &sa.id, &sa.name, &sa.scope, &teamID, &lastUsed)
 	if err != nil {
 		return serviceAccountPrincipal{}, false
 	}
@@ -260,9 +266,11 @@ func serviceAccountFor(ctx context.Context, db *sql.DB, token string) (serviceAc
 	}
 
 	// best-effort; don't fail the request if this update fails
+	if now := time.Now(); !lastUsed.Valid || now.Sub(time.Unix(lastUsed.Int64, 0)) > keyTouchEvery {
 		db.ExecContext(ctx,
 			"UPDATE service_account_keys SET last_used_at = $1 WHERE id = $2",
-		time.Now().Unix(), keyID)
+			now.Unix(), keyID)
+	}
 	return sa, true
 }
 
@@ -286,10 +294,8 @@ func serveAsServiceAccount(w http.ResponseWriter, r *http.Request, next http.Han
 	next.ServeHTTP(w, r.WithContext(ctx))
 }
 
-// isInstanceServiceAccount is a thin compatibility wrapper over
-// Caller.IsInstanceServiceAccount(), for call sites outside this package's
-// core predicates (handleCreateTeam, handleCreateServiceAccount) that
-// needed this exact, narrow check before the Caller abstraction existed.
+// isInstanceServiceAccount is Caller.IsInstanceServiceAccount() on the
+// request's Caller.
 func isInstanceServiceAccount(ctx context.Context) bool {
 	c, _ := callerFromContext(ctx)
 	return c.IsInstanceServiceAccount()
@@ -397,6 +403,13 @@ func requireTeamOwner(w http.ResponseWriter, r *http.Request, teamID int64) bool
 	if ok && role == models.RoleOwner {
 		return true
 	}
+	// The operator's instance-scoped account manages every team's
+	// configuration, which is what lets it use one credential instead of
+	// minting one per team. This is owner reach only: it does not make the
+	// account a member, so it still reads no team's incidents.
+	if c, _ := callerFromContext(r.Context()); c.IsInstanceServiceAccount() {
+		return true
+	}
 	if caller, _ := userFromContext(r.Context()); caller.IsAdmin {
 		return true
 	}
@@ -414,3 +427,26 @@ func sessionFromContext(ctx context.Context) (int64, bool) {
 	id, ok := ctx.Value(ctxSession).(int64)
 	return id, ok
 }
+
+// requestLogger logs one line per request with the matched route pattern in
+// place of the URL path. Two routes carry a credential in the path (the
+// integration key and the ack token), and chi's stock logger would write it to
+// the log verbatim.
+func requestLogger(next http.Handler) http.Handler {
+	return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		start := time.Now()
+		ww := middleware.NewWrapResponseWriter(w, r.ProtoMajor)
+		next.ServeHTTP(ww, r)
+		route := "unmatched"
+		if rc := chi.RouteContext(r.Context()); rc != nil {
+			if p := rc.RoutePattern(); p != "" {
+				route = p
+			}
+		}
+		status := ww.Status()
+		if status == 0 {
+			status = http.StatusOK
+		}
+		log.Printf("%s %s %d %dB %s", r.Method, route, status, ww.BytesWritten(), time.Since(start).Round(time.Millisecond))
+	})
+}

internal/api/notifier.go

@@ -449,7 +449,7 @@ func renderNotification(inc models.Incident, n outboxRow, firing int, cfg Notify
 // originLabel is the label that says where an alert came from, for a team with
 // several Kubernetes clusters behind it. It comes from Prometheus's
 // externalLabels and reaches an incident through Alertmanager's group_by; the
-// web UI reads the same label, and the README ("Several clusters, one team")
+// web UI reads the same label, and docs/incidents.md ("Several clusters, one team")
 // explains how to set it up.
 const originLabel = "cluster"
 

internal/api/notify_ack.go

@@ -62,7 +62,7 @@ func handleNotifyAck(db *sql.DB) http.HandlerFunc {
 
 		acked, err := acknowledgeIncident(r.Context(), db, incidentID, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if !acked {
@@ -73,7 +73,7 @@ func handleNotifyAck(db *sql.DB) http.HandlerFunc {
 			// ntfy show a success toast rather than a failure.
 			inc, err := fetchIncident(r.Context(), db, incidentID)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			respond(w, http.StatusOK, map[string]any{

internal/api/oidc.go

@@ -130,12 +130,12 @@ func handleOIDCLogin(db *sql.DB, prov *oidc.Provider, limiter *loginLimiter, pub
 
 		state, stateHash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		nonce, _, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		verifier := oidc.NewVerifier()
@@ -149,7 +149,7 @@ func handleOIDCLogin(db *sql.DB, prov *oidc.Provider, limiter *loginLimiter, pub
 			INSERT INTO oidc_logins (state_hash, nonce, pkce_verifier, next, expires_at)
 			VALUES ($1, $2, $3, $4, $5)`,
 			stateHash, nonce, verifier, next, now.Add(oidcLoginTTL).Unix()); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 

internal/api/oidc_teams.go

@@ -31,7 +31,7 @@ func handleGetTeamOIDCGroups(db *sql.DB) http.HandlerFunc {
 			"SELECT COALESCE(oidc_member_group, ''), COALESCE(oidc_owner_group, '') FROM teams WHERE id = $1",
 			teamID).Scan(&g.MemberGroup, &g.OwnerGroup)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, g)
@@ -66,7 +66,7 @@ func handleSetTeamOIDCGroups(db *sql.DB) http.HandlerFunc {
 			       oidc_owner_group  = NULLIF($2, '')
 			 WHERE id = $3`,
 			req.MemberGroup, req.OwnerGroup, teamID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)

internal/api/operator_key.go

@@ -0,0 +1,72 @@
+package api
+
+import (
+	"context"
+	"database/sql"
+	"fmt"
+
+	"git.ryuvia.com/niklas/terdut-server/internal/models"
+)
+
+const (
+	// operatorAccountName is the instance-scoped service account the operator
+	// key belongs to.
+	operatorAccountName = "terdut-operator"
+
+	// operatorKeyName names the one key SeedOperatorKey manages on it, so a
+	// rotation replaces that key and leaves any others alone.
+	operatorKeyName = "seed"
+)
+
+// SeedOperatorKey makes key the operator account's credential: it creates the
+// instance-scoped service account if needed and replaces its "seed" key with
+// this one. Idempotent, so every replica can run it at every start, and a
+// rotated key simply wins on the next restart.
+//
+// The key is hashed like any other, so only the caller that generated it ever
+// holds the raw value. An empty key does nothing.
+func SeedOperatorKey(ctx context.Context, db *sql.DB, key string) error {
+	if key == "" {
+		return nil
+	}
+	tx, err := db.BeginTx(ctx, nil)
+	if err != nil {
+		return fmt.Errorf("seed operator key: %w", err)
+	}
+	defer tx.Rollback() //nolint:errcheck
+
+	// Serialise replicas starting together; transaction-scoped, so it needs no
+	// explicit release.
+	if _, err := tx.ExecContext(ctx, "SELECT pg_advisory_xact_lock($1)", operatorKeyLockKey); err != nil {
+		return fmt.Errorf("seed operator key: lock: %w", err)
+	}
+
+	var accountID int64
+	err = tx.QueryRowContext(ctx,
+		"SELECT id FROM service_accounts WHERE name = $1 AND scope = $2",
+		operatorAccountName, models.ServiceAccountScopeInstance).Scan(&accountID)
+	if err == sql.ErrNoRows {
+		err = tx.QueryRowContext(ctx,
+			"INSERT INTO service_accounts (name, scope) VALUES ($1, $2) RETURNING id",
+			operatorAccountName, models.ServiceAccountScopeInstance).Scan(&accountID)
+	}
+	if err != nil {
+		return fmt.Errorf("seed operator key: account: %w", err)
+	}
+
+	if _, err := tx.ExecContext(ctx,
+		"DELETE FROM service_account_keys WHERE service_account_id = $1 AND name = $2",
+		accountID, operatorKeyName); err != nil {
+		return fmt.Errorf("seed operator key: drop old key: %w", err)
+	}
+	if _, err := tx.ExecContext(ctx,
+		"INSERT INTO service_account_keys (service_account_id, key_hash, name) VALUES ($1, $2, $3)",
+		accountID, hashToken(key), operatorKeyName); err != nil {
+		return fmt.Errorf("seed operator key: store key: %w", err)
+	}
+	return tx.Commit()
+}
+
+// operatorKeyLockKey is the transaction-scoped advisory lock SeedOperatorKey
+// holds; distinct from the other lock keys in this package.
+const operatorKeyLockKey int64 = 7265_0010

internal/api/operator_key_test.go

@@ -0,0 +1,134 @@
+package api_test
+
+import (
+	"context"
+	"encoding/json"
+	"net/http"
+	"testing"
+
+	"git.ryuvia.com/niklas/terdut-server/internal/api"
+)
+
+const (
+	operatorKeyOne = "tdsa_operator-key-number-one-0123456789"
+	operatorKeyTwo = "tdsa_operator-key-number-two-0123456789"
+)
+
+func TestSeedOperatorKey_AuthenticatesAsInstanceAccount(t *testing.T) {
+	s := newTS(t)
+	if err := api.SeedOperatorKey(context.Background(), s.db, operatorKeyOne); err != nil {
+		t.Fatal(err)
+	}
+
+	resp := s.reqAs(t, operatorKeyOne, http.MethodPost, "/api/teams", map[string]string{"name": "seeded"})
+	resp.Body.Close()
+	if resp.StatusCode != http.StatusCreated {
+		t.Fatalf("seeded key should create a team, got %d", resp.StatusCode)
+	}
+}
+
+func TestSeedOperatorKey_RotationReplacesAndIsIdempotent(t *testing.T) {
+	s := newTS(t)
+	ctx := context.Background()
+	for _, key := range []string{operatorKeyOne, operatorKeyOne, operatorKeyTwo} {
+		if err := api.SeedOperatorKey(ctx, s.db, key); err != nil {
+			t.Fatal(err)
+		}
+	}
+
+	old := s.reqAs(t, operatorKeyOne, http.MethodGet, "/api/teams", nil)
+	old.Body.Close()
+	if old.StatusCode != http.StatusUnauthorized {
+		t.Errorf("rotated-out key should be refused, got %d", old.StatusCode)
+	}
+	cur := s.reqAs(t, operatorKeyTwo, http.MethodGet, "/api/teams", nil)
+	cur.Body.Close()
+	if cur.StatusCode != http.StatusOK {
+		t.Errorf("current key should work, got %d", cur.StatusCode)
+	}
+
+	var accounts, keys int
+	s.db.QueryRow("SELECT COUNT(*) FROM service_accounts WHERE name = 'terdut-operator'").Scan(&accounts)
+	s.db.QueryRow("SELECT COUNT(*) FROM service_account_keys").Scan(&keys)
+	if accounts != 1 || keys != 1 {
+		t.Errorf("expected one account and one key, got %d and %d", accounts, keys)
+	}
+}
+
+func TestSeedOperatorKey_EmptyKeyDoesNothing(t *testing.T) {
+	s := newTS(t)
+	if err := api.SeedOperatorKey(context.Background(), s.db, ""); err != nil {
+		t.Fatal(err)
+	}
+	var n int
+	s.db.QueryRow("SELECT COUNT(*) FROM service_accounts").Scan(&n)
+	if n != 0 {
+		t.Errorf("expected no service account, got %d", n)
+	}
+}
+
+// The instance account configures a team it did not create, which is what lets
+// the operator hold one credential instead of one per team, yet it is not a
+// member and so reads none of the team's incidents.
+func TestInstanceAccount_ActsAsOwnerOfAnyTeamButIsNoMember(t *testing.T) {
+	s := newTS(t)
+	other := newTeam(t, s, "other")
+	if err := api.SeedOperatorKey(context.Background(), s.db, operatorKeyOne); err != nil {
+		t.Fatal(err)
+	}
+
+	rename := s.reqAs(t, operatorKeyOne, http.MethodPut, "/api/teams/"+id64(other.id), map[string]string{"name": "renamed"})
+	rename.Body.Close()
+	if rename.StatusCode >= 300 {
+		t.Errorf("instance account should rename any team, got %d", rename.StatusCode)
+	}
+
+	// Not a member: the team's queue is not visible to it.
+	var queue []map[string]any
+	decode(t, s.reqAs(t, operatorKeyOne, http.MethodGet, "/api/incidents", nil), &queue)
+	if len(queue) != 0 {
+		t.Errorf("instance account should see no incidents, got %v", queue)
+	}
+}
+
+// external_id lets automation find its own team again after a crash, without
+// trusting a display name.
+func TestCreateTeam_ExternalIDIsIdempotentAndInstanceOnly(t *testing.T) {
+	s := newTS(t)
+	if err := api.SeedOperatorKey(context.Background(), s.db, operatorKeyOne); err != nil {
+		t.Fatal(err)
+	}
+	create := func(name string) (int, map[string]any) {
+		resp := s.reqAs(t, operatorKeyOne, http.MethodPost, "/api/teams",
+			map[string]string{"name": name, "external_id": "ns/platform"})
+		var out map[string]any
+		_ = json.NewDecoder(resp.Body).Decode(&out)
+		resp.Body.Close()
+		return resp.StatusCode, out
+	}
+
+	code, first := create("Platform")
+	if code != http.StatusCreated {
+		t.Fatalf("first create: %d", code)
+	}
+	// Same identity, even under a new display name: the same team comes back.
+	code, again := create("Platform renamed")
+	if code != http.StatusOK || again["id"] != first["id"] {
+		t.Errorf("repeat with the same external_id: want 200 and team %v, got %d %v", first["id"], code, again)
+	}
+
+	// A different identity cannot take the name.
+	resp := s.reqAs(t, operatorKeyOne, http.MethodPost, "/api/teams",
+		map[string]string{"name": "Platform", "external_id": "other/platform"})
+	resp.Body.Close()
+	if resp.StatusCode != http.StatusConflict {
+		t.Errorf("taken name under another external_id: want 409, got %d", resp.StatusCode)
+	}
+
+	// A person cannot set one.
+	resp = s.req(t, http.MethodPost, "/api/teams", map[string]string{"name": "Mine", "external_id": "x/y"})
+	resp.Body.Close()
+	if resp.StatusCode != http.StatusForbidden {
+		t.Errorf("a user setting external_id: want 403, got %d", resp.StatusCode)
+	}
+}

internal/api/router.go

@@ -23,12 +23,13 @@ func NewRouter(db *sql.DB, notify NotifyConfig, cfg config.Config, version strin
 	// One limiter each, both process-wide for the life of the router: login
 	// counts failed passwords, sign-up counts account creation, and mixing the
 	// two would let a burst of sign-ups lock somebody out of logging in.
+	trustedProxies.Store(int32(cfg.TrustedProxies))
 	loginLimit := newLoginLimiter(db)
 	signupLimiter := newLoginLimiter(db)
 	oidcLimit := newLoginLimiter(db)
 
 	r := chi.NewRouter()
-	r.Use(middleware.Logger)
+	r.Use(requestLogger)
 	r.Use(middleware.Recoverer)
 	r.Use(securityHeaders(notify.PublicURL))
 
@@ -48,11 +49,7 @@ func NewRouter(db *sql.DB, notify NotifyConfig, cfg config.Config, version strin
 
 	// Alert ingestion. The key in the path says both that the sender may post
 	// and which team the alerts belong to, which is why it needs no session.
-	//
-	// This is the only way in. The pre-teams /api/alertmanager/webhook, which
-	// took no credential at all, was removed in v0.13.0 once the cluster's
-	// Alertmanager had moved onto a key; a sender still posting there gets the
-	// JSON 404 every unknown /api path gets.
+	// This is the only way in.
 	r.Post("/api/integrations/{key}/alertmanager", handleIntegrationWebhook(db, notify))
 
 	// Signing up. Both are unauthenticated by necessity: the caller has no
@@ -116,9 +113,7 @@ func NewRouter(db *sql.DB, notify NotifyConfig, cfg config.Config, version strin
 		r.Post("/api/users/{id}/api-keys", handleCreateAPIKey(db))
 		r.Delete("/api/users/{id}/api-keys/{keyID}", handleDeleteAPIKey(db))
 
-		// Administration: who exists, and who is an administrator. Until #3
-		// these were open to any authenticated caller, which meant every user
-		// could delete every other one.
+		// Administration: who exists, and who is an administrator.
 		r.Group(func(r chi.Router) {
 			r.Use(AdminOnly)
 
@@ -147,7 +142,6 @@ func NewRouter(db *sql.DB, notify NotifyConfig, cfg config.Config, version strin
 		r.Get("/api/incidents", handleListIncidents(db))
 		r.Get("/api/incidents/clusters", handleListClusters(db))
 		r.Get("/api/incidents/{id}", handleGetIncident(db))
-		r.Get("/api/incidents/{id}/alerts", handleIncidentAlerts(db))
 		r.Get("/api/incidents/{id}/timeline", handleIncidentTimeline(db))
 		r.Get("/api/incidents/{id}/similar", handleIncidentSimilar(db))
 		r.Post("/api/incidents/{id}/acknowledge", handleIncidentAcknowledge(db))

internal/api/schedule.go

@@ -69,7 +69,7 @@ func handleCreateSchedule(db *sql.DB) http.HandlerFunc {
 		// be, so the delete and the insert share one transaction.
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback()
@@ -79,7 +79,7 @@ func handleCreateSchedule(db *sql.DB) http.HandlerFunc {
 				if _, err := tx.ExecContext(r.Context(),
 					"DELETE FROM schedule_entries WHERE team_id = $1 AND date = $2",
 					teamID, d); err != nil {
-					respond(w, http.StatusInternalServerError, errResp("internal error"))
+					serverError(w, r, err)
 					return
 				}
 			}
@@ -91,12 +91,12 @@ func handleCreateSchedule(db *sql.DB) http.HandlerFunc {
 						errResp("date already assigned: "+d+" (pass replace to take it)"))
 					return
 				}
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 		}
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -107,7 +107,7 @@ func handleCreateSchedule(db *sql.DB) http.HandlerFunc {
 		}
 		all, err := scheduleRange(r.Context(), db, teamID, req.Dates[0], req.Dates[len(req.Dates)-1])
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		created := []models.ScheduleEntry{}
@@ -147,7 +147,7 @@ func handleListSchedule(db *sql.DB) http.HandlerFunc {
 
 		entries, err := scheduleRange(r.Context(), db, teamID, from, to)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, entries)
@@ -171,7 +171,7 @@ func handleDeleteSchedule(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM schedule_entries WHERE id = $1 AND team_id = $2", id, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -197,7 +197,7 @@ func handleCurrentSchedule(db *sql.DB) http.HandlerFunc {
 			WHERE  s.date = $1 AND s.team_id = ANY($2)
 			ORDER  BY t.name`, today, callerTeamIDs(r.Context()))
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -207,14 +207,14 @@ func handleCurrentSchedule(db *sql.DB) http.HandlerFunc {
 			var e models.ScheduleEntry
 			var ts int64
 			if err := rows.Scan(&e.ID, &e.TeamID, &e.TeamName, &e.UserID, &e.Username, &e.Date, &ts); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			e.CreatedAt = time.Unix(ts, 0).UTC()
 			entries = append(entries, e)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, entries)

internal/api/service_accounts.go

@@ -50,6 +50,9 @@ func callerIsAdmin(ctx context.Context) bool {
 // writing a response: callers here need to combine it with other ways of
 // being allowed, not stop at the first no.
 func callerOwnsTeam(ctx context.Context, teamID int64) bool {
+	if c, _ := callerFromContext(ctx); c.IsInstanceServiceAccount() {
+		return true
+	}
 	role, ok := callerRole(ctx, teamID)
 	return ok && role == models.RoleOwner
 }
@@ -132,7 +135,7 @@ func handleCreateServiceAccount(db *sql.DB) http.HandlerFunc {
 
 		key, err := mintServiceAccountKey(r.Context(), db, sa.ID, "initial")
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusCreated, map[string]any{"service_account": sa, "key": key})
@@ -233,7 +236,7 @@ func handleCreateServiceAccountKey(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if !callerMayManageServiceAccount(r.Context(), sa) {
@@ -255,7 +258,7 @@ func handleCreateServiceAccountKey(db *sql.DB) http.HandlerFunc {
 
 		key, err := mintServiceAccountKey(r.Context(), db, sa.ID, req.Name)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusCreated, key)
@@ -274,7 +277,7 @@ func handleDeleteServiceAccountKey(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if !callerMayManageServiceAccount(r.Context(), sa) {
@@ -290,7 +293,7 @@ func handleDeleteServiceAccountKey(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM service_account_keys WHERE id = $1 AND service_account_id = $2", keyID, sa.ID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -325,7 +328,7 @@ func handleListServiceAccounts(db *sql.DB) http.HandlerFunc {
 
 		rows, err := db.QueryContext(r.Context(), query, args...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -335,14 +338,14 @@ func handleListServiceAccounts(db *sql.DB) http.HandlerFunc {
 			var sa models.ServiceAccount
 			var created int64
 			if err := rows.Scan(&sa.ID, &sa.Name, &sa.Scope, &sa.TeamID, &sa.CreatedBy, &created); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			sa.CreatedAt = time.Unix(created, 0).UTC()
 			accounts = append(accounts, sa)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, accounts)

internal/api/settings.go

@@ -203,7 +203,7 @@ func handleSetSettings(db *sql.DB) http.HandlerFunc {
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback() //nolint:errcheck
@@ -215,12 +215,12 @@ func handleSetSettings(db *sql.DB) http.HandlerFunc {
 				ON CONFLICT (key) DO UPDATE SET
 					value = excluded.value, updated_at = excluded.updated_at`,
 				key, value); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 		}
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -261,7 +261,7 @@ func handleAdminListTeams(db *sql.DB) http.HandlerFunc {
 			FROM   teams t
 			ORDER  BY t.name`)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -272,14 +272,14 @@ func handleAdminListTeams(db *sql.DB) http.HandlerFunc {
 			var created int64
 			if err := rows.Scan(&t.ID, &t.Name, &created, &t.Members, &t.OpenIncidents,
 				&t.OIDCMemberGroup, &t.OIDCOwnerGroup); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			t.CreatedAt = time.Unix(created, 0).UTC()
 			teams = append(teams, t)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, teams)
@@ -319,7 +319,7 @@ func handleAdminGetTeam(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		t.CreatedAt = time.Unix(created, 0).UTC()
@@ -333,7 +333,7 @@ func handleAdminGetTeam(db *sql.DB) http.HandlerFunc {
 			WHERE  m.team_id = $1
 			ORDER  BY u.username`, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -343,14 +343,14 @@ func handleAdminGetTeam(db *sql.DB) http.HandlerFunc {
 			var m models.TeamMember
 			var joined int64
 			if err := rows.Scan(&m.TeamID, &m.UserID, &m.Username, &m.Role, &joined, &m.Source); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			m.JoinedAt = time.Unix(joined, 0).UTC()
 			members = append(members, m)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -396,7 +396,7 @@ func handleRenameTeam(db *sql.DB) http.HandlerFunc {
 				respond(w, http.StatusConflict, errResp("a team with that name already exists"))
 				return
 			}
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -435,7 +435,7 @@ func handleSetUserDisabled(db *sql.DB) http.HandlerFunc {
 			}
 			last, err := isLastAdmin(r.Context(), db, id)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if last {
@@ -453,7 +453,7 @@ func handleSetUserDisabled(db *sql.DB) http.HandlerFunc {
 				"UPDATE users SET disabled_at = NULL WHERE id = $1", id)
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -475,7 +475,7 @@ func handleSetUserDisabled(db *sql.DB) http.HandlerFunc {
 
 		user, err := fetchUser(r.Context(), db, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, user)

internal/api/signup.go

@@ -170,13 +170,13 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 
 		hash, err := hashPassword(req.Password)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback() //nolint:errcheck
@@ -194,7 +194,7 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 				respond(w, http.StatusConflict, errResp("username or email already exists"))
 				return
 			}
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -207,7 +207,7 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 					respond(w, http.StatusConflict, errResp("a team with that name already exists"))
 					return
 				}
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			role = models.RoleOwner
@@ -216,7 +216,7 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 		if _, err := tx.ExecContext(r.Context(),
 			"INSERT INTO team_members (team_id, user_id, role) VALUES ($1, $2, $3)",
 			teamID, userID, role); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -226,7 +226,7 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 			res, err := tx.ExecContext(r.Context(),
 				"UPDATE invites SET uses = uses + 1 WHERE id = $1 AND uses < max_uses", inv.id)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if n, _ := res.RowsAffected(); n == 0 {
@@ -236,14 +236,14 @@ func handleSignup(db *sql.DB, limiter *loginLimiter, publicURL string) http.Hand
 		}
 
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		// Signed in immediately: the alternative is a form that says "now go
 		// and log in", which is the same credential typed twice.
 		if err := startSession(w, r, db, userID, publicURL); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		user, _ := fetchUser(r.Context(), db, userID)
@@ -291,7 +291,7 @@ func handleListInvites(db *sql.DB) http.HandlerFunc {
 			WHERE  team_id = $1
 			ORDER  BY id DESC`, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -303,7 +303,7 @@ func handleListInvites(db *sql.DB) http.HandlerFunc {
 			var revoked *int64
 			if err := rows.Scan(&i.ID, &i.TeamID, &i.Role, &created, &expires,
 				&i.MaxUses, &i.Uses, &revoked); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			i.CreatedAt = time.Unix(created, 0).UTC()
@@ -312,7 +312,7 @@ func handleListInvites(db *sql.DB) http.HandlerFunc {
 			out = append(out, i)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, out)
@@ -356,7 +356,7 @@ func handleCreateInvite(db *sql.DB, publicURL string) http.HandlerFunc {
 
 		raw, hash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		// created_by is nullable (ON DELETE SET NULL) for exactly this
@@ -382,7 +382,7 @@ func handleCreateInvite(db *sql.DB, publicURL string) http.HandlerFunc {
 			RETURNING id, team_id, role, created_at, expires_at, max_uses, uses`,
 			hash, teamID, req.Role, createdBy, expires.Unix(), req.MaxUses).
 			Scan(&out.ID, &out.TeamID, &out.Role, &created, &expiresAt, &out.MaxUses, &out.Uses); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		out.CreatedAt = time.Unix(created, 0).UTC()
@@ -412,7 +412,7 @@ func handleRevokeInvite(db *sql.DB) http.HandlerFunc {
 			"UPDATE invites SET revoked_at = "+nowEpoch+
 				" WHERE id = $1 AND team_id = $2 AND revoked_at IS NULL", id, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -450,7 +450,7 @@ func handleTestNotification(cfg NotifyConfig, db *sql.DB) http.HandlerFunc {
 		var topic *string
 		if err := db.QueryRowContext(r.Context(),
 			"SELECT ntfy_topic FROM users WHERE id = $1", caller.ID).Scan(&topic); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if topic == nil || *topic == "" {
@@ -501,7 +501,7 @@ func handleDismissOnboarding(db *sql.DB) http.HandlerFunc {
 				"UPDATE users SET onboarding_dismissed_at = NULL WHERE id = $1", caller.ID)
 		}
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		w.WriteHeader(http.StatusNoContent)

internal/api/similar.go

@@ -37,7 +37,7 @@ func handleIncidentSimilar(db *sql.DB) http.HandlerFunc {
 
 		out, err := similarIncidents(r.Context(), db, id, limit)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, out)

internal/api/stats.go

@@ -24,7 +24,7 @@ func handleStatsAlerts(db *sql.DB) http.HandlerFunc {
 			FROM alerts WHERE %s`, where), args.all()...,
 		).Scan(&total, &firing, &resolved)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, map[string]int64{
@@ -55,7 +55,7 @@ func handleStatsTop(db *sql.DB) http.HandlerFunc {
 			ORDER  BY cnt DESC
 			LIMIT  %s`, where, args.add(limit)), args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -68,13 +68,13 @@ func handleStatsTop(db *sql.DB) http.HandlerFunc {
 		for rows.Next() {
 			var e entry
 			if err := rows.Scan(&e.Name, &e.Count); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			result = append(result, e)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, result)
@@ -93,7 +93,7 @@ func handleStatsByHour(db *sql.DB) http.HandlerFunc {
 			GROUP  BY hr
 			ORDER  BY hr ASC`, where), args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -103,13 +103,13 @@ func handleStatsByHour(db *sql.DB) http.HandlerFunc {
 			var hr int
 			var cnt int64
 			if err := rows.Scan(&hr, &cnt); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			counts[hr] = cnt
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -129,8 +129,7 @@ func handleStatsByDay(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
 		where, args := statsFilter(r.URL.Query(), "received_at", callerTeamIDs(r.Context()))
 
-		// Postgres EXTRACT(DOW …) → 0=Sunday … 6=Saturday, the same numbering
-		// SQLite's strftime('%w') returned, so the frontend needs no change.
+		// Postgres EXTRACT(DOW …) → 0=Sunday … 6=Saturday.
 		rows, err := db.QueryContext(r.Context(), fmt.Sprintf(`
 			SELECT EXTRACT(DOW FROM to_timestamp(received_at) AT TIME ZONE 'UTC')::int AS dow,
 			       COUNT(*) AS cnt
@@ -139,7 +138,7 @@ func handleStatsByDay(db *sql.DB) http.HandlerFunc {
 			GROUP  BY dow
 			ORDER  BY dow ASC`, where), args.all()...)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -149,13 +148,13 @@ func handleStatsByDay(db *sql.DB) http.HandlerFunc {
 			var dow int
 			var cnt int64
 			if err := rows.Scan(&dow, &cnt); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			counts[dow] = cnt
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -198,7 +197,7 @@ func handleStatsIncidents(db *sql.DB) http.HandlerFunc {
 			FROM incidents WHERE %s`, where), args.all()...,
 		).Scan(&total, &triggered, &acknowledged, &resolved, &mtta, &mttr)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 

internal/api/teams.go

@@ -13,19 +13,12 @@ import (
 	"github.com/go-chi/chi/v5"
 )
 
-// handleListTeams lists the caller's own teams, each with their role in it,
-// or — with ?name= — looks up one team by exact name regardless of caller
-// identity (TEAM-LOOKUP.md). An administrator listing every team goes
-// through the admin endpoint instead: the no-name case here answers "what am
-// I part of", which is what the UI's team filter and the combined queue are
-// built from.
+// handleListTeams lists the caller's own teams, each with their role in it.
+// An administrator listing every team goes through the admin endpoint instead:
+// this answers "what am I part of", which is what the UI's team filter and the
+// combined queue are built from.
 func handleListTeams(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
-		if name := strings.TrimSpace(r.URL.Query().Get("name")); name != "" {
-			handleListTeamsByName(db, w, r, name)
-			return
-		}
-
 		caller, _ := userFromContext(r.Context())
 		rows, err := db.QueryContext(r.Context(), `
 			SELECT t.id, t.name, t.created_at, m.role, m.source
@@ -34,7 +27,7 @@ func handleListTeams(db *sql.DB) http.HandlerFunc {
 			WHERE  m.user_id = $1
 			ORDER  BY t.name`, caller.ID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -44,45 +37,20 @@ func handleListTeams(db *sql.DB) http.HandlerFunc {
 			var t models.Team
 			var created int64
 			if err := rows.Scan(&t.ID, &t.Name, &created, &t.Role, &t.Source); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			t.CreatedAt = time.Unix(created, 0).UTC()
 			teams = append(teams, t)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, teams)
 	}
 }
 
-// handleListTeamsByName answers "is there a team named exactly this", open to
-// any authenticated caller including a service account (TEAM-LOOKUP.md) —
-// mirrors handleListServiceAccounts' own ?name= lookup: a one-or-zero-length
-// array, never an error on no match, and no caller-identity filtering at
-// all, since what it discloses (a name is taken, nothing about who's in it
-// or any of its data) is the same low sensitivity that lookup already
-// accepts for service-account names.
-func handleListTeamsByName(db *sql.DB, w http.ResponseWriter, r *http.Request, name string) {
-	var t models.Team
-	var created int64
-	err := db.QueryRowContext(r.Context(),
-		"SELECT id, name, created_at FROM teams WHERE name = $1", name,
-	).Scan(&t.ID, &t.Name, &created)
-	if errors.Is(err, sql.ErrNoRows) {
-		respond(w, http.StatusOK, []models.Team{})
-		return
-	}
-	if err != nil {
-		respond(w, http.StatusInternalServerError, errResp("internal error"))
-		return
-	}
-	t.CreatedAt = time.Unix(created, 0).UTC()
-	respond(w, http.StatusOK, []models.Team{t})
-}
-
 // handleUserTeams lists one user's teams, for the admin page's per-user view:
 // "what is this person in", which /api/teams cannot answer because it is always
 // about the caller.
@@ -106,7 +74,7 @@ func handleUserTeams(db *sql.DB) http.HandlerFunc {
 		var exists bool
 		if err := db.QueryRowContext(r.Context(),
 			"SELECT EXISTS (SELECT 1 FROM users WHERE id = $1)", id).Scan(&exists); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if !exists {
@@ -121,7 +89,7 @@ func handleUserTeams(db *sql.DB) http.HandlerFunc {
 			WHERE  m.user_id = $1
 			ORDER  BY t.name`, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -131,14 +99,14 @@ func handleUserTeams(db *sql.DB) http.HandlerFunc {
 			var t models.Team
 			var created int64
 			if err := rows.Scan(&t.ID, &t.Name, &created, &t.Role, &t.Source); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			t.CreatedAt = time.Unix(created, 0).UTC()
 			teams = append(teams, t)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, teams)
@@ -161,6 +129,12 @@ func handleCreateTeam(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
 		var req struct {
 			Name string `json:"name"`
+			// ExternalID makes the call idempotent for automation: a team
+			// already carrying it is returned as-is (200) instead of created, so
+			// a client that crashed between the POST and recording the id finds
+			// its own team again. Instance-scoped service accounts only; a name
+			// that belongs to a different team is still a 409.
+			ExternalID string `json:"external_id"`
 		}
 		if err := decodeJSON(r, &req); err != nil {
 			respond(w, http.StatusBadRequest, errResp("invalid request body"))
@@ -173,6 +147,10 @@ func handleCreateTeam(db *sql.DB) http.HandlerFunc {
 		}
 
 		caller, isUser := userFromContext(r.Context())
+		if req.ExternalID != "" && !isInstanceServiceAccount(r.Context()) {
+			respond(w, http.StatusForbidden, errResp("external_id is for instance-scoped service accounts"))
+			return
+		}
 		if !isUser && !isInstanceServiceAccount(r.Context()) {
 			// A team-scoped service account authenticates as owner of exactly
 			// one team already (see serveAsServiceAccount); letting it create
@@ -183,37 +161,57 @@ func handleCreateTeam(db *sql.DB) http.HandlerFunc {
 
 		tx, err := db.BeginTx(r.Context(), nil)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer tx.Rollback() //nolint:errcheck
 
 		var team models.Team
 		var created int64
+		if req.ExternalID != "" {
+			err := tx.QueryRowContext(r.Context(),
+				"SELECT id, name, created_at FROM teams WHERE external_id = $1", req.ExternalID).
+				Scan(&team.ID, &team.Name, &created)
+			if err == nil {
+				team.ExternalID = &req.ExternalID
+				team.CreatedAt = time.Unix(created, 0).UTC()
+				respond(w, http.StatusOK, team)
+				return
+			}
+			if !errors.Is(err, sql.ErrNoRows) {
+				serverError(w, r, err)
+				return
+			}
+		}
+		var externalID *string
+		if req.ExternalID != "" {
+			externalID = &req.ExternalID
+		}
 		if err := tx.QueryRowContext(r.Context(),
-			"INSERT INTO teams (name) VALUES ($1) RETURNING id, name, created_at",
-			req.Name).Scan(&team.ID, &team.Name, &created); err != nil {
+			"INSERT INTO teams (name, external_id) VALUES ($1, $2) RETURNING id, name, created_at",
+			req.Name, externalID).Scan(&team.ID, &team.Name, &created); err != nil {
 			if isUniqueViolation(err) {
 				respond(w, http.StatusConflict, errResp("a team with that name already exists"))
 				return
 			}
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if isUser {
 			if _, err := tx.ExecContext(r.Context(),
 				"INSERT INTO team_members (team_id, user_id, role) VALUES ($1, $2, $3)",
 				team.ID, caller.ID, models.RoleOwner); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 		}
 		if err := tx.Commit(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		team.CreatedAt = time.Unix(created, 0).UTC()
+		team.ExternalID = externalID
 		if isUser {
 			team.Role = models.RoleOwner
 		}
@@ -241,7 +239,7 @@ func handleDeleteTeam(db *sql.DB) http.HandlerFunc {
 		if err := db.QueryRowContext(r.Context(),
 			"SELECT COUNT(*) FROM incidents WHERE team_id = $1 AND resolved_at IS NULL", teamID).
 			Scan(&open); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if open > 0 {
@@ -251,7 +249,7 @@ func handleDeleteTeam(db *sql.DB) http.HandlerFunc {
 
 		res, err := db.ExecContext(r.Context(), "DELETE FROM teams WHERE id = $1", teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -322,7 +320,7 @@ func handleListTeamMembers(db *sql.DB) http.HandlerFunc {
 			WHERE  m.team_id = $1
 			ORDER  BY u.username`, teamID, todayUTC())
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -334,7 +332,7 @@ func handleListTeamMembers(db *sql.DB) http.HandlerFunc {
 			var hasTopic, disabled bool
 			if err := rows.Scan(&m.TeamID, &m.UserID, &m.Username, &m.Role, &joined, &m.Source,
 				&hasTopic, &disabled, &lastActive, &m.OnCall, &m.NextShift); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			m.JoinedAt = time.Unix(joined, 0).UTC()
@@ -360,7 +358,7 @@ func handleListTeamMembers(db *sql.DB) http.HandlerFunc {
 			members = append(members, m)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, members)
@@ -396,7 +394,7 @@ func handleAddTeamMember(db *sql.DB) http.HandlerFunc {
 		}
 
 		if managed, err := isSSOManagedMember(r.Context(), db, teamID, req.UserID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		} else if managed {
 			respond(w, http.StatusConflict, errResp(ssoManagedMsg))
@@ -408,7 +406,7 @@ func handleAddTeamMember(db *sql.DB) http.HandlerFunc {
 		if req.Role == models.RoleMember {
 			last, err := isLastTeamOwner(r.Context(), db, teamID, req.UserID)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if last {
@@ -453,7 +451,7 @@ func handleRemoveTeamMember(db *sql.DB) http.HandlerFunc {
 		}
 
 		if managed, err := isSSOManagedMember(r.Context(), db, teamID, userID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		} else if managed {
 			respond(w, http.StatusConflict, errResp(ssoManagedMsg))
@@ -462,7 +460,7 @@ func handleRemoveTeamMember(db *sql.DB) http.HandlerFunc {
 
 		last, err := isLastTeamOwner(r.Context(), db, teamID, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if last {
@@ -473,7 +471,7 @@ func handleRemoveTeamMember(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM team_members WHERE team_id = $1 AND user_id = $2", teamID, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -572,7 +570,7 @@ func handleListIntegrations(db *sql.DB) http.HandlerFunc {
 			WHERE  i.team_id = $1
 			ORDER  BY i.id`, teamID, now.Add(-sourceQuietAfter).Unix())
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -584,7 +582,7 @@ func handleListIntegrations(db *sql.DB) http.HandlerFunc {
 			var lastUsed, lastAlert *int64
 			if err := rows.Scan(&i.ID, &i.TeamID, &i.Kind, &i.Name, &created, &lastUsed,
 				&lastAlert, &i.Alerts24h); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			i.CreatedAt = time.Unix(created, 0).UTC()
@@ -601,7 +599,7 @@ func handleListIntegrations(db *sql.DB) http.HandlerFunc {
 			integrations = append(integrations, i)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, integrations)
@@ -644,7 +642,7 @@ func handleCreateIntegration(db *sql.DB, publicURL string) http.HandlerFunc {
 
 		raw, hash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
@@ -656,7 +654,11 @@ func handleCreateIntegration(db *sql.DB, publicURL string) http.HandlerFunc {
 			RETURNING id, team_id, kind, name, created_at`,
 			teamID, req.Kind, req.Name, hash).
 			Scan(&i.ID, &i.TeamID, &i.Kind, &i.Name, &created); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			if isUniqueViolation(err) {
+				respond(w, http.StatusConflict, errResp("an integration with that name already exists in this team"))
+				return
+			}
+			serverError(w, r, err)
 			return
 		}
 		i.CreatedAt = time.Unix(created, 0).UTC()
@@ -703,7 +705,11 @@ func handleRenameIntegration(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"UPDATE integrations SET name = $1 WHERE id = $2 AND team_id = $3", req.Name, id, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			if isUniqueViolation(err) {
+				respond(w, http.StatusConflict, errResp("an integration with that name already exists in this team"))
+				return
+			}
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -732,7 +738,7 @@ func handleDeleteIntegration(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM integrations WHERE id = $1 AND team_id = $2", id, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -790,16 +796,6 @@ func teamParam(w http.ResponseWriter, r *http.Request) (int64, bool) {
 	return id, true
 }
 
-// defaultTeamID is the oldest team, which on an upgraded install is the
-// "Default" team every pre-teams row was moved into and on a fresh one is the
-// team migration 003 creates. Bootstrap puts the first user in it, so somebody
-// signing in to a new server lands somewhere rather than in no team at all.
-func defaultTeamID(ctx context.Context, db *sql.DB) (int64, error) {
-	var id int64
-	err := db.QueryRowContext(ctx, "SELECT id FROM teams ORDER BY id LIMIT 1").Scan(&id)
-	return id, err
-}
-
 // ---------------------------------------------------------------------------
 // A team's dead man's switches
 // ---------------------------------------------------------------------------
@@ -833,12 +829,12 @@ func handleListTeamDeadman(db *sql.DB) http.HandlerFunc {
 
 		set, err := deadmanSetForTeam(r.Context(), db, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		out, err := deadmanStatuses(r.Context(), db, teamID, set, time.Now())
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, out)
@@ -901,7 +897,11 @@ func handleCreateTeamDeadman(db *sql.DB) http.HandlerFunc {
 			INSERT INTO deadman_switches (team_id, name, matcher, timeout_seconds, severity)
 			VALUES ($1, $2, $3, $4, $5) RETURNING id`,
 			teamID, req.Name, m.config(), req.TimeoutSeconds, req.Severity).Scan(&id); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			if isUniqueViolation(err) {
+				respond(w, http.StatusConflict, errResp("a switch with that name already exists in this team"))
+				return
+			}
+			serverError(w, r, err)
 			return
 		}
 
@@ -975,7 +975,11 @@ func handleUpdateTeamDeadman(db *sql.DB) http.HandlerFunc {
 			 WHERE id = $5 AND team_id = $6`,
 			req.Name, m.config(), req.TimeoutSeconds, req.Severity, switchID, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			if isUniqueViolation(err) {
+				respond(w, http.StatusConflict, errResp("a switch with that name already exists in this team"))
+				return
+			}
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -990,12 +994,12 @@ func handleUpdateTeamDeadman(db *sql.DB) http.HandlerFunc {
 		// handleListTeamDeadman would give it.
 		set, err := deadmanSetForTeam(r.Context(), db, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		statuses, err := deadmanStatuses(r.Context(), db, teamID, set, time.Now())
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		for _, s := range statuses {
@@ -1004,7 +1008,7 @@ func handleUpdateTeamDeadman(db *sql.DB) http.HandlerFunc {
 				return
 			}
 		}
-		respond(w, http.StatusInternalServerError, errResp("internal error"))
+		serverError(w, r, err)
 	}
 }
 
@@ -1029,7 +1033,7 @@ func handleDeleteTeamDeadman(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM deadman_switches WHERE id = $1 AND team_id = $2", switchID, teamID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {

internal/api/teams_test.go

@@ -6,8 +6,6 @@ import (
 	"io"
 	"net/http"
 	"testing"
-
-	"git.ryuvia.com/niklas/terdut-server/internal/models"
 )
 
 // The whole point of #4: two teams sharing one server must not see each other's
@@ -146,7 +144,6 @@ func TestTeams_IncidentsAreScopedToTheReceivingTeam(t *testing.T) {
 	otherID := int64(blueIncidents[0]["id"].(float64))
 	for _, path := range []string{
 		"/api/incidents/" + id64(otherID),
-		"/api/incidents/" + id64(otherID) + "/alerts",
 		"/api/incidents/" + id64(otherID) + "/timeline",
 	} {
 		resp := red.call(http.MethodGet, path, nil)
@@ -461,70 +458,58 @@ func TestTeams_OutsiderSeesNothing(t *testing.T) {
 // GET /api/teams?name= (TEAM-LOOKUP.md)
 // ---------------------------------------------------------------------------
 
-func TestListTeamsByName_FindsExactMatch(t *testing.T) {
-	s := newTS(t)
-	instanceKey := createServiceAccount(t, s, s.key, "terdut-operator", models.ServiceAccountScopeInstance, 0)
-	teamID := createTeamAs(t, s, instanceKey, "platform")
-
-	teams := list(t, s.reqAs(t, instanceKey, http.MethodGet, "/api/teams?name=platform", nil))
-	if len(teams) != 1 {
-		t.Fatalf("expected exactly one match for ?name=platform, got %d: %v", len(teams), teams)
-	}
-	if int64(teams[0]["id"].(float64)) != teamID {
-		t.Errorf("id = %v, want %d", teams[0]["id"], teamID)
-	}
-	// No membership, so no role to report (models.Team's own doc comment:
-	// "empty when nobody in particular is asking").
-	if _, has := teams[0]["role"]; has {
-		t.Errorf("expected no role on a name-lookup match, got %v", teams[0]["role"])
-	}
-}
-
-func TestListTeamsByName_NoMatchIsAnEmptyArrayNotAnError(t *testing.T) {
-	s := newTS(t)
-	instanceKey := createServiceAccount(t, s, s.key, "terdut-operator", models.ServiceAccountScopeInstance, 0)
-
-	resp := s.reqAs(t, instanceKey, http.MethodGet, "/api/teams?name=does-not-exist", nil)
-	if resp.StatusCode != http.StatusOK {
-		t.Fatalf("expected 200 on no match, got %d", resp.StatusCode)
-	}
-	teams := list(t, resp)
-	if len(teams) != 0 {
-		t.Errorf("expected an empty array, got %v", teams)
-	}
-}
-
-// The actual motivating scenario (TEAM-LOOKUP.md): a service account that
-// already created a team, interrupted before it could remember the id,
-// recovers it via ?name= on the same name its own POST 409s on.
-func TestListTeamsByName_RecoversAfterCreateConflict(t *testing.T) {
-	s := newTS(t)
-	instanceKey := createServiceAccount(t, s, s.key, "terdut-operator", models.ServiceAccountScopeInstance, 0)
-	original := createTeamAs(t, s, instanceKey, "recovered")
-
-	conflict := s.reqAs(t, instanceKey, http.MethodPost, "/api/teams", map[string]string{"name": "recovered"})
-	if conflict.StatusCode != http.StatusConflict {
-		t.Fatalf("expected 409 recreating the same name, got %d", conflict.StatusCode)
-	}
-	conflict.Body.Close()
-
-	teams := list(t, s.reqAs(t, instanceKey, http.MethodGet, "/api/teams?name=recovered", nil))
-	if len(teams) != 1 || int64(teams[0]["id"].(float64)) != original {
-		t.Fatalf("expected to recover the original team %d via ?name=, got %v", original, teams)
-	}
-}
-
-// Not gated by isInstanceServiceAccount or AdminOnly (TEAM-LOOKUP.md): any
-// authenticated caller may ask whether a name is taken, the same low
-// sensitivity GET /api/service-accounts?name= already accepts.
-func TestListTeamsByName_OpenToAnyAuthenticatedCaller(t *testing.T) {
+// Everybody signed in can list users to name them, but only an admin (or the
+// row's owner) sees an email or an ntfy topic, which is a publish secret.
+func TestListUsers_RedactsEmailAndTopicForOthers(t *testing.T) {
 	s := newTS(t)
 	red := newTeam(t, s, "red")
-	_ = createTeamAs(t, s, s.key, "blue-target")
+	_ = newTeam(t, s, "blue")
+	s.exec(t, "UPDATE users SET ntfy_topic = 'secret-topic'")
 
-	// red's own member, not a member of "blue-target", still gets a match.
-	teams := list(t, red.call(http.MethodGet, "/api/teams?name=blue-target", nil))
-	if len(teams) != 1 || teams[0]["name"] != "blue-target" {
-		t.Errorf("expected a non-member caller to still find the team by name, got %v", teams)
+	var asAdmin []map[string]any
+	decode(t, s.req(t, http.MethodGet, "/api/users", nil), &asAdmin)
+	for _, u := range asAdmin {
+		if u["email"] == "" || u["ntfy_topic"] != "secret-topic" {
+			t.Errorf("admin should see everything, got %v", u)
+		}
+	}
+
+	asMember := list(t, red.call(http.MethodGet, "/api/users", nil))
+	if len(asMember) < 3 {
+		t.Fatalf("expected the whole user list, got %v", asMember)
+	}
+	for _, u := range asMember {
+		own := u["username"] == "red-user"
+		if own != (u["email"] != "") || own != (u["ntfy_topic"] != nil) {
+			t.Errorf("only red-user's own row should keep email and topic, got %v", u)
+		}
+	}
+}
+
+// Names identify integrations and switches within a team.
+func TestTeamNames_AreUniquePerTeam(t *testing.T) {
+	s := newTS(t) // creates one integration named "test" in the default team
+
+	dup := s.req(t, http.MethodPost, "/api/teams/"+defaultTeam+"/integrations", map[string]string{"name": "test"})
+	dup.Body.Close()
+	if dup.StatusCode != http.StatusConflict {
+		t.Errorf("duplicate integration name: expected 409, got %d", dup.StatusCode)
+	}
+
+	body := map[string]any{"matcher": "alertname=Watchdog", "timeout_seconds": 60, "severity": "critical"}
+	first := s.req(t, http.MethodPost, "/api/teams/"+defaultTeam+"/deadman/switches", body)
+	first.Body.Close()
+	second := s.req(t, http.MethodPost, "/api/teams/"+defaultTeam+"/deadman/switches", body)
+	second.Body.Close()
+	if first.StatusCode != http.StatusCreated || second.StatusCode != http.StatusConflict {
+		t.Errorf("duplicate switch name: expected 201 then 409, got %d then %d", first.StatusCode, second.StatusCode)
+	}
+
+	// The same name in another team is fine.
+	other := newTeam(t, s, "elsewhere")
+	ok := s.req(t, http.MethodPost, "/api/teams/"+id64(other.id)+"/integrations", map[string]string{"name": "test"})
+	ok.Body.Close()
+	if ok.StatusCode != http.StatusCreated {
+		t.Errorf("same name in another team: expected 201, got %d", ok.StatusCode)
 	}
 }

internal/api/testdb_test.go

@@ -13,9 +13,8 @@ import (
 	"git.ryuvia.com/niklas/terdut-server/internal/db"
 )
 
-// Tests run against a real Postgres, because the server does. SQLite's
-// ":memory:" gave every test a private database for free; Postgres has no
-// equivalent, so isolation is bought with a schema per test.
+// Tests run against a real Postgres, because the server does. Isolation is
+// bought with a schema per test.
 //
 // A schema rather than a database: CREATE DATABASE copies a template on disk and
 // costs a hundred milliseconds or so each time, while CREATE SCHEMA plus the one

internal/api/users.go

@@ -15,6 +15,10 @@ import (
 	"github.com/go-chi/chi/v5"
 )
 
+// bootstrapLockKey is the transaction-scoped advisory lock handleBootstrap
+// holds; distinct from the migration and notifier keys.
+const bootstrapLockKey = 0x7465726475744254
+
 func handleBootstrap(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
 		var req struct {
@@ -40,15 +44,30 @@ func handleBootstrap(db *sql.DB) http.HandlerFunc {
 			}
 			h, err := hashPassword(req.Password)
 			if err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			passwordHash = &h
 		}
 
+		// Check-then-insert has to be one atomic step: two concurrent calls on
+		// an empty install would otherwise both see zero users and both create
+		// an admin. The transaction-scoped lock serialises them, and the loser
+		// sees the winner's row.
+		tx, err := db.BeginTx(r.Context(), nil)
+		if err != nil {
+			serverError(w, r, err)
+			return
+		}
+		defer tx.Rollback() //nolint:errcheck
+		if _, err := tx.ExecContext(r.Context(), "SELECT pg_advisory_xact_lock($1)", bootstrapLockKey); err != nil {
+			serverError(w, r, err)
+			return
+		}
+
 		var count int
-		if err := db.QueryRowContext(r.Context(), "SELECT COUNT(*) FROM users").Scan(&count); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+		if err := tx.QueryRowContext(r.Context(), "SELECT COUNT(*) FROM users").Scan(&count); err != nil {
+			serverError(w, r, err)
 			return
 		}
 		if count > 0 {
@@ -57,34 +76,29 @@ func handleBootstrap(db *sql.DB) http.HandlerFunc {
 		}
 
 		var userID int64
-		if err := db.QueryRowContext(r.Context(),
+		if err := tx.QueryRowContext(r.Context(),
 			"INSERT INTO users (username, email, password_hash, is_admin) VALUES ($1, $2, $3, true) RETURNING id",
 			req.Username, req.Email, passwordHash).Scan(&userID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
 		raw, hash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		var keyID int64
-		if err := db.QueryRowContext(r.Context(),
+		if err := tx.QueryRowContext(r.Context(),
 			"INSERT INTO api_keys (user_id, key_hash, name) VALUES ($1, $2, $3) RETURNING id",
 			userID, hash, "bootstrap").Scan(&keyID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 
-		// The default team exists from migration 003, on a fresh install too.
-		// Without a membership the first user signs in to a working server with
-		// no queue, no schedule and nowhere for an integration to hang off.
-		if teamID, err := defaultTeamID(r.Context(), db); err == nil {
-			db.ExecContext(r.Context(), //nolint:errcheck
-				"INSERT INTO team_members (team_id, user_id, role) VALUES ($1, $2, $3) "+
-					"ON CONFLICT (team_id, user_id) DO NOTHING",
-				teamID, userID, models.RoleOwner)
+		if err := tx.Commit(); err != nil {
+			serverError(w, r, err)
+			return
 		}
 
 		user, _ := fetchUser(r.Context(), db, userID)
@@ -93,12 +107,19 @@ func handleBootstrap(db *sql.DB) http.HandlerFunc {
 	}
 }
 
+// handleListUsers is readable by anyone signed in, because the assignment
+// control and the schedule need to name people. What it returns about other
+// people is therefore only what naming them takes: email and ntfy_topic are
+// blanked unless the caller is an admin or the row is their own. The topic in
+// particular is a publish secret.
 func handleListUsers(db *sql.DB) http.HandlerFunc {
 	return func(w http.ResponseWriter, r *http.Request) {
+		caller, _ := userFromContext(r.Context())
+		seeAll := caller.IsAdmin
 		rows, err := db.QueryContext(r.Context(),
 			"SELECT id, username, email, created_at, ntfy_topic, is_admin, admin_source, disabled_at FROM users ORDER BY id")
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -109,15 +130,19 @@ func handleListUsers(db *sql.DB) http.HandlerFunc {
 			var ts int64
 			var disabled *int64
 			if err := rows.Scan(&u.ID, &u.Username, &u.Email, &ts, &u.NtfyTopic, &u.IsAdmin, &u.AdminSource, &disabled); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			u.CreatedAt = time.Unix(ts, 0).UTC()
 			u.DisabledAt = unixPtr(disabled)
+			if !seeAll && u.ID != caller.ID {
+				u.Email = ""
+				u.NtfyTopic = nil
+			}
 			users = append(users, u)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, users)
@@ -147,7 +172,7 @@ func handleCreateUser(db *sql.DB) http.HandlerFunc {
 				respond(w, http.StatusConflict, errResp("username or email already exists"))
 				return
 			}
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		user, _ := fetchUser(r.Context(), db, id)
@@ -185,7 +210,7 @@ func handleSetNotifyTarget(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"UPDATE users SET ntfy_topic = $1 WHERE id = $2", topic, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -195,7 +220,7 @@ func handleSetNotifyTarget(db *sql.DB) http.HandlerFunc {
 
 		user, err := fetchUser(r.Context(), db, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, user)
@@ -217,7 +242,7 @@ func handleDeleteUser(db *sql.DB) http.HandlerFunc {
 			return
 		}
 		if last, err := isLastAdmin(r.Context(), db, id); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		} else if last {
 			respond(w, http.StatusConflict, errResp("cannot delete the last administrator"))
@@ -226,7 +251,7 @@ func handleDeleteUser(db *sql.DB) http.HandlerFunc {
 
 		res, err := db.ExecContext(r.Context(), "DELETE FROM users WHERE id = $1", id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		n, _ := res.RowsAffected()
@@ -283,7 +308,7 @@ func handleCreateAPIKey(db *sql.DB) http.HandlerFunc {
 
 		raw, hash, err := randomToken()
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		var expiresAt *int64
@@ -298,7 +323,7 @@ func handleCreateAPIKey(db *sql.DB) http.HandlerFunc {
 		if err := db.QueryRowContext(r.Context(),
 			"INSERT INTO api_keys (user_id, key_hash, name, expires_at) VALUES ($1, $2, $3, $4) RETURNING id",
 			userID, hash, req.Name, expiresAt).Scan(&keyID); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		key := models.APIKey{
@@ -329,7 +354,7 @@ func handleListAPIKeys(db *sql.DB) http.HandlerFunc {
 			`SELECT id, name, created_at, last_used_at, expires_at
 			   FROM api_keys WHERE user_id = $1 ORDER BY created_at DESC`, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		defer rows.Close()
@@ -340,7 +365,7 @@ func handleListAPIKeys(db *sql.DB) http.HandlerFunc {
 			var created int64
 			var lastUsed, expires *int64
 			if err := rows.Scan(&k.ID, &k.Name, &created, &lastUsed, &expires); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			k.UserID = userID
@@ -350,7 +375,7 @@ func handleListAPIKeys(db *sql.DB) http.HandlerFunc {
 			keys = append(keys, k)
 		}
 		if err := rows.Err(); err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, keys)
@@ -376,7 +401,7 @@ func handleDeleteAPIKey(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"DELETE FROM api_keys WHERE id = $1 AND user_id = $2", keyID, userID)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		n, _ := res.RowsAffected()
@@ -442,7 +467,7 @@ func handleSetAdmin(db *sql.DB) http.HandlerFunc {
 			if err := db.QueryRowContext(r.Context(),
 				"SELECT EXISTS (SELECT 1 FROM users WHERE id = $1 AND is_admin AND admin_source = 'oidc')",
 				id).Scan(&managed); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			}
 			if managed {
@@ -456,7 +481,7 @@ func handleSetAdmin(db *sql.DB) http.HandlerFunc {
 				return
 			}
 			if last, err := isLastAdmin(r.Context(), db, id); err != nil {
-				respond(w, http.StatusInternalServerError, errResp("internal error"))
+				serverError(w, r, err)
 				return
 			} else if last {
 				respond(w, http.StatusConflict, errResp("cannot revoke the last administrator"))
@@ -467,7 +492,7 @@ func handleSetAdmin(db *sql.DB) http.HandlerFunc {
 		res, err := db.ExecContext(r.Context(),
 			"UPDATE users SET is_admin = $1 WHERE id = $2", *req.IsAdmin, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		if n, _ := res.RowsAffected(); n == 0 {
@@ -477,7 +502,7 @@ func handleSetAdmin(db *sql.DB) http.HandlerFunc {
 
 		user, err := fetchUser(r.Context(), db, id)
 		if err != nil {
-			respond(w, http.StatusInternalServerError, errResp("internal error"))
+			serverError(w, r, err)
 			return
 		}
 		respond(w, http.StatusOK, user)

internal/config/config.go

@@ -5,17 +5,21 @@ import (
 	"fmt"
 	"net/url"
 	"os"
+	"strconv"
 	"strings"
 	"time"
 )
 
+// MinOperatorKeyLength is the shortest TERDUT_OPERATOR_KEY accepted: it is a
+// bearer credential with instance reach, so a short one is refused outright.
+const MinOperatorKeyLength = 32
+
 type Config struct {
 	Addr string
 
 	// DSN is the Postgres connection string, e.g.
 	// postgres://terdut:secret@host:5432/terdut?sslmode=require. Required:
-	// unlike the SQLite path it replaced there is no sensible default, and a
-	// server that silently came up against the wrong database would be worse
+	// there is no sensible default, and a server that silently came up against the wrong database would be worse
 	// than one that refuses to start.
 	DSN string
 
@@ -26,25 +30,6 @@ type Config struct {
 	// repeat_interval (default 4h), which is what refreshes the alert.
 	StaleAfter time.Duration
 
-	// DeadmanMatchers selects the alerts that are heartbeats rather than
-	// problems: receiving one opens no incident, and the absence of one does.
-	//
-	// ";" separates matchers, "," the label conditions within one, "=" is exact
-	// equality — `alertname=Watchdog,cluster=prod; alertname=Heartbeat`. Every
-	// matcher must name an alertname. See api.ParseDeadmanConfig.
-	DeadmanMatchers string
-
-	// DeadmanTimeout is how long a heartbeat may go unheard before its switch is
-	// declared dead. It must be *shorter* than the Alertmanager repeat_interval
-	// of the route carrying the heartbeat — the opposite of StaleAfter, and the
-	// reason a dead man's switch usually wants a route of its own. Zero disables
-	// dead man's switch handling entirely.
-	DeadmanTimeout time.Duration
-
-	// DeadmanSeverity is the severity a dead man's switch incident opens at.
-	// These incidents have no member alerts to derive one from.
-	DeadmanSeverity string
-
 	// NtfyURL is the ntfy server push notifications are published to. Empty
 	// disables notifications entirely.
 	NtfyURL string
@@ -70,6 +55,20 @@ type Config struct {
 	// Config, which is what a test or a new caller builds, keeps passwords working.
 	DisablePasswordLogin bool
 
+	// OperatorKey, when set, is the credential of the instance-scoped service
+	// account "terdut-operator", created or re-keyed at every start. It is how
+	// terdut-operator gets in without a bootstrap handshake: the operator
+	// generates the key, hands it to the server here, and uses it as its bearer
+	// token. Empty means no such account is managed.
+	OperatorKey string
+
+	// TrustedProxies is how many reverse proxies sit in front of the server and
+	// append to X-Forwarded-For. The per-address rate limits take the client
+	// address that many entries from the right, because everything further left
+	// is whatever the client chose to send. 0 ignores the header and uses the
+	// connection's own address.
+	TrustedProxies int
+
 	// OIDC configures single sign-on. The zero value, with no Issuer, is off.
 	OIDC OIDC
 
@@ -133,24 +132,12 @@ func Load() Config {
 	if addr == "" {
 		addr = ":8080"
 	}
-	deadmanMatchers := os.Getenv("TERDUT_DEADMAN_MATCHERS")
-	if deadmanMatchers == "" {
-		deadmanMatchers = "alertname=Watchdog"
-	}
-	deadmanSeverity := os.Getenv("TERDUT_DEADMAN_SEVERITY")
-	if deadmanSeverity == "" {
-		deadmanSeverity = "critical"
-	}
 	return Config{
 		Addr:         addr,
 		DSN:          os.Getenv("TERDUT_DB_DSN"),
 		ArchiveAfter: duration("TERDUT_ARCHIVE_AFTER", 7*24*time.Hour),
 		StaleAfter:   duration("TERDUT_STALE_AFTER", 6*time.Hour),
 
-		DeadmanMatchers: deadmanMatchers,
-		DeadmanTimeout:  duration("TERDUT_DEADMAN_TIMEOUT", 15*time.Minute),
-		DeadmanSeverity: deadmanSeverity,
-
 		NtfyURL:           os.Getenv("TERDUT_NTFY_URL"),
 		NtfyToken:         os.Getenv("TERDUT_NTFY_TOKEN"),
 		NtfyFallbackTopic: os.Getenv("TERDUT_NTFY_FALLBACK_TOPIC"),
@@ -158,6 +145,9 @@ func Load() Config {
 		NotifyRepeat:      duration("TERDUT_NOTIFY_REPEAT", 15*time.Minute),
 
 		DisablePasswordLogin: !boolean("TERDUT_PASSWORD_LOGIN", true),
+
+		OperatorKey:    strings.TrimSpace(os.Getenv("TERDUT_OPERATOR_KEY")),
+		TrustedProxies: integer("TERDUT_TRUSTED_PROXIES", 1),
 		OIDC:           loadOIDC(),
 
 		OperatorMode: boolean("TERDUT_OPERATOR_MODE", false),
@@ -187,6 +177,9 @@ func loadOIDC() OIDC {
 // provider would come up and then fail every login, which is harder to notice
 // than not starting.
 func (c Config) Validate() error {
+	if c.OperatorKey != "" && len(c.OperatorKey) < MinOperatorKeyLength {
+		return fmt.Errorf("TERDUT_OPERATOR_KEY must be at least %d characters", MinOperatorKeyLength)
+	}
 	o := c.OIDC
 	if !o.Enabled() {
 		if c.DisablePasswordLogin {
@@ -226,6 +219,16 @@ func str(env, def string) string {
 	return def
 }
 
+// integer reads a non-negative int env var; anything else takes the default.
+func integer(env string, def int) int {
+	if s := os.Getenv(env); s != "" {
+		if n, err := strconv.Atoi(strings.TrimSpace(s)); err == nil && n >= 0 {
+			return n
+		}
+	}
+	return def
+}
+
 // list reads a comma- or space-separated env var.
 func list(env, def string) []string {
 	s := os.Getenv(env)

internal/db/db.go

@@ -35,8 +35,7 @@ const (
 //
 // The pool is modest on purpose: this server's concurrency comes from a handful
 // of HTTP handlers plus two background loops, and a cloud-native-pg instance
-// sized for it has a low max_connections. It is still a pool, unlike the single
-// connection SQLite forced, so the notifier no longer blocks a webhook.
+// sized for it has a low max_connections.
 func Open(dsn string) (*sql.DB, error) {
 	if dsn == "" {
 		return nil, fmt.Errorf("empty DSN: set TERDUT_DB_DSN")
@@ -76,9 +75,8 @@ const migrationLockKey int64 = 7265_0003
 // Migrate applies every embedded migration that has not been applied yet, in
 // filename order, recording each in schema_migrations.
 //
-// Each file runs inside a transaction, which SQLite's version did not do: a
-// migration that failed half way used to leave the schema in whatever state it
-// had reached. Postgres has transactional DDL, so the rollback is real.
+// Each file runs inside a transaction, so a migration that fails half way
+// leaves the schema as it was: Postgres has transactional DDL.
 func Migrate(db *sql.DB) error {
 	ctx := context.Background()
 	conn, err := db.Conn(ctx)

internal/db/migrations/001_schema.sql

@@ -0,0 +1,587 @@
+-- Terdut Server schema. One baseline: the project has not shipped, so the
+-- migration history that led here (SQLite import, a Default team, per-team
+-- deadman configs later replaced by switches) is not carried. Later changes are
+-- new numbered files after this one.
+--
+-- Timestamps are Unix epoch seconds in BIGINT columns throughout.
+
+CREATE TABLE alerts (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    fingerprint text NOT NULL,
+    name text NOT NULL,
+    status text NOT NULL,
+    labels jsonb DEFAULT '{}'::jsonb NOT NULL,
+    annotations jsonb DEFAULT '{}'::jsonb NOT NULL,
+    starts_at bigint NOT NULL,
+    ends_at bigint,
+    generator_url text DEFAULT ''::text NOT NULL,
+    received_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    archived_at bigint,
+    resolution_source text,
+    team_id bigint NOT NULL,
+    integration_id bigint,
+    CONSTRAINT alerts_status_check CHECK ((status = ANY (ARRAY['firing'::text, 'resolved'::text])))
+);
+
+CREATE TABLE api_keys (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    user_id bigint NOT NULL,
+    key_hash text NOT NULL,
+    name text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    last_used_at bigint,
+    expires_at bigint
+);
+
+CREATE TABLE deadman_switches (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    team_id bigint NOT NULL,
+    name text NOT NULL,
+    matcher text NOT NULL,
+    timeout_seconds bigint NOT NULL,
+    severity text DEFAULT 'critical'::text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    CONSTRAINT deadman_switches_timeout_seconds_check CHECK ((timeout_seconds > 0))
+);
+
+CREATE TABLE device_logins (
+    device_hash text NOT NULL,
+    user_code text NOT NULL,
+    status text DEFAULT 'pending'::text NOT NULL,
+    user_id bigint,
+    expires_at bigint NOT NULL,
+    last_polled_at bigint DEFAULT 0 NOT NULL,
+    CONSTRAINT device_logins_status_check CHECK ((status = ANY (ARRAY['pending'::text, 'approved'::text, 'denied'::text])))
+);
+
+CREATE TABLE escalation_levels (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    team_id bigint NOT NULL,
+    "position" bigint NOT NULL,
+    timeout_seconds bigint NOT NULL,
+    CONSTRAINT escalation_levels_timeout_seconds_check CHECK ((timeout_seconds > 0))
+);
+
+CREATE TABLE escalation_policies (
+    team_id bigint NOT NULL,
+    repeat_count bigint DEFAULT 0 NOT NULL,
+    fallback_topic text DEFAULT ''::text NOT NULL,
+    updated_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    CONSTRAINT escalation_policies_repeat_count_check CHECK (((repeat_count >= 0) AND (repeat_count <= 10)))
+);
+
+CREATE TABLE escalation_targets (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    level_id bigint NOT NULL,
+    kind text NOT NULL,
+    user_id bigint,
+    CONSTRAINT escalation_targets_check CHECK ((((kind = 'user'::text) AND (user_id IS NOT NULL)) OR ((kind = 'oncall'::text) AND (user_id IS NULL)))),
+    CONSTRAINT escalation_targets_kind_check CHECK ((kind = ANY (ARRAY['user'::text, 'oncall'::text])))
+);
+
+CREATE TABLE incident_ack_tokens (
+    token_hash text NOT NULL,
+    incident_id bigint NOT NULL,
+    user_id bigint NOT NULL,
+    created_at bigint NOT NULL,
+    expires_at bigint NOT NULL
+);
+
+CREATE TABLE incident_alerts (
+    incident_id bigint NOT NULL,
+    alert_id bigint NOT NULL,
+    added_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL
+);
+
+CREATE TABLE incident_events (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    incident_id bigint NOT NULL,
+    type text NOT NULL,
+    user_id bigint,
+    alert_id bigint,
+    detail text,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    service_account_id bigint,
+    actor_user_id bigint,
+    actor_service_account_id bigint,
+    CONSTRAINT incident_events_actor_xor_chk CHECK (((user_id IS NULL) OR (service_account_id IS NULL))),
+    CONSTRAINT incident_events_assign_actor_xor_chk CHECK (((actor_user_id IS NULL) OR (actor_service_account_id IS NULL)))
+);
+
+CREATE TABLE incidents (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    group_key text NOT NULL,
+    title text NOT NULL,
+    group_labels jsonb DEFAULT '{}'::jsonb NOT NULL,
+    status text NOT NULL,
+    severity text,
+    triggered_at bigint NOT NULL,
+    acknowledged_by bigint,
+    acknowledged_at bigint,
+    assigned_to bigint,
+    snoozed_until bigint,
+    resolved_at bigint,
+    resolution_source text,
+    archived_at bigint,
+    team_id bigint NOT NULL,
+    escalation_level bigint DEFAULT 0 NOT NULL,
+    escalation_level_at bigint,
+    escalation_round bigint DEFAULT 0 NOT NULL,
+    signature text DEFAULT ''::text NOT NULL,
+    acknowledged_by_service_account_id bigint,
+    CONSTRAINT incidents_ack_actor_xor_chk CHECK (((acknowledged_by IS NULL) OR (acknowledged_by_service_account_id IS NULL))),
+    CONSTRAINT incidents_status_check CHECK ((status = ANY (ARRAY['triggered'::text, 'acknowledged'::text, 'resolved'::text])))
+);
+
+CREATE TABLE integrations (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    team_id bigint NOT NULL,
+    kind text NOT NULL,
+    name text NOT NULL,
+    key_hash text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    last_used_at bigint,
+    CONSTRAINT integrations_kind_check CHECK ((kind = 'alertmanager'::text))
+);
+
+CREATE TABLE invites (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    token_hash text NOT NULL,
+    team_id bigint NOT NULL,
+    role text NOT NULL,
+    created_by bigint,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    expires_at bigint NOT NULL,
+    max_uses bigint DEFAULT 1 NOT NULL,
+    uses bigint DEFAULT 0 NOT NULL,
+    revoked_at bigint,
+    CONSTRAINT invites_max_uses_check CHECK (((max_uses > 0) AND (max_uses <= 100))),
+    CONSTRAINT invites_role_check CHECK ((role = ANY (ARRAY['owner'::text, 'member'::text])))
+);
+
+CREATE TABLE notifications (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    incident_id bigint NOT NULL,
+    user_id bigint,
+    topic text NOT NULL,
+    kind text NOT NULL,
+    created_at bigint NOT NULL,
+    send_after bigint NOT NULL,
+    attempts bigint DEFAULT 0 NOT NULL,
+    sent_at bigint,
+    last_error text,
+    CONSTRAINT notifications_kind_check CHECK ((kind = ANY (ARRAY['triggered'::text, 'reminder'::text, 'resolved'::text, 'escalated'::text])))
+);
+
+CREATE TABLE oidc_logins (
+    state_hash text NOT NULL,
+    nonce text NOT NULL,
+    pkce_verifier text NOT NULL,
+    expires_at bigint NOT NULL,
+    next text DEFAULT '/'::text NOT NULL
+);
+
+CREATE TABLE rate_limit_counters (
+    key text NOT NULL,
+    window_start bigint NOT NULL,
+    count integer NOT NULL
+);
+
+CREATE TABLE schedule_entries (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    user_id bigint NOT NULL,
+    date text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    team_id bigint NOT NULL
+);
+
+CREATE TABLE service_account_keys (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    service_account_id bigint NOT NULL,
+    key_hash text NOT NULL,
+    name text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    last_used_at bigint
+);
+
+CREATE TABLE service_accounts (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    name text NOT NULL,
+    scope text NOT NULL,
+    team_id bigint,
+    created_by bigint,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    CONSTRAINT service_accounts_scope_check CHECK ((scope = ANY (ARRAY['instance'::text, 'team'::text]))),
+    CONSTRAINT service_accounts_scope_team_id_chk CHECK ((((scope = 'team'::text) AND (team_id IS NOT NULL)) OR ((scope = 'instance'::text) AND (team_id IS NULL))))
+);
+
+CREATE TABLE sessions (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    token_hash text NOT NULL,
+    user_id bigint NOT NULL,
+    created_at bigint NOT NULL,
+    last_seen_at bigint NOT NULL,
+    expires_at bigint NOT NULL,
+    user_agent text,
+    max_expires_at bigint
+);
+
+CREATE TABLE settings (
+    key text NOT NULL,
+    value text NOT NULL,
+    updated_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL
+);
+
+CREATE TABLE team_members (
+    team_id bigint NOT NULL,
+    user_id bigint NOT NULL,
+    role text NOT NULL,
+    joined_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    source text DEFAULT 'manual'::text NOT NULL,
+    CONSTRAINT team_members_role_check CHECK ((role = ANY (ARRAY['owner'::text, 'member'::text]))),
+    CONSTRAINT team_members_source_check CHECK ((source = ANY (ARRAY['manual'::text, 'oidc'::text])))
+);
+
+CREATE TABLE teams (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    name text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    oidc_member_group text,
+    oidc_owner_group text,
+    -- A stable identity for a team managed by automation (terdut-operator:
+    -- "<namespace>/<name>" of its TerdutTeam), so it can find or recreate its own
+    -- team without trusting a display name. NULL for a team a person made.
+    external_id text
+);
+
+CREATE TABLE user_identities (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    user_id bigint NOT NULL,
+    issuer text NOT NULL,
+    subject text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    last_login_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL
+);
+
+CREATE TABLE users (
+    id bigint GENERATED BY DEFAULT AS IDENTITY NOT NULL,
+    username text NOT NULL,
+    email text NOT NULL,
+    created_at bigint DEFAULT (floor(EXTRACT(epoch FROM now())))::bigint NOT NULL,
+    ntfy_topic text,
+    password_hash text,
+    is_admin boolean DEFAULT false NOT NULL,
+    disabled_at bigint,
+    invited_via bigint,
+    onboarding_dismissed_at bigint,
+    admin_source text DEFAULT 'manual'::text NOT NULL,
+    CONSTRAINT users_admin_source_check CHECK ((admin_source = ANY (ARRAY['manual'::text, 'oidc'::text])))
+);
+
+ALTER TABLE alerts
+    ADD CONSTRAINT alerts_pkey PRIMARY KEY (id);
+
+ALTER TABLE api_keys
+    ADD CONSTRAINT api_keys_key_hash_key UNIQUE (key_hash);
+
+ALTER TABLE api_keys
+    ADD CONSTRAINT api_keys_pkey PRIMARY KEY (id);
+
+ALTER TABLE deadman_switches
+    ADD CONSTRAINT deadman_switches_pkey PRIMARY KEY (id);
+
+ALTER TABLE device_logins
+    ADD CONSTRAINT device_logins_pkey PRIMARY KEY (device_hash);
+
+ALTER TABLE device_logins
+    ADD CONSTRAINT device_logins_user_code_key UNIQUE (user_code);
+
+ALTER TABLE escalation_levels
+    ADD CONSTRAINT escalation_levels_pkey PRIMARY KEY (id);
+
+ALTER TABLE escalation_levels
+    ADD CONSTRAINT escalation_levels_team_id_position_key UNIQUE (team_id, "position");
+
+ALTER TABLE escalation_policies
+    ADD CONSTRAINT escalation_policies_pkey PRIMARY KEY (team_id);
+
+ALTER TABLE escalation_targets
+    ADD CONSTRAINT escalation_targets_pkey PRIMARY KEY (id);
+
+ALTER TABLE incident_ack_tokens
+    ADD CONSTRAINT incident_ack_tokens_pkey PRIMARY KEY (token_hash);
+
+ALTER TABLE incident_alerts
+    ADD CONSTRAINT incident_alerts_pkey PRIMARY KEY (incident_id, alert_id);
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_pkey PRIMARY KEY (id);
+
+ALTER TABLE incidents
+    ADD CONSTRAINT incidents_pkey PRIMARY KEY (id);
+
+ALTER TABLE integrations
+    ADD CONSTRAINT integrations_key_hash_key UNIQUE (key_hash);
+
+ALTER TABLE integrations
+    ADD CONSTRAINT integrations_pkey PRIMARY KEY (id);
+
+ALTER TABLE invites
+    ADD CONSTRAINT invites_pkey PRIMARY KEY (id);
+
+ALTER TABLE invites
+    ADD CONSTRAINT invites_token_hash_key UNIQUE (token_hash);
+
+ALTER TABLE notifications
+    ADD CONSTRAINT notifications_pkey PRIMARY KEY (id);
+
+ALTER TABLE oidc_logins
+    ADD CONSTRAINT oidc_logins_pkey PRIMARY KEY (state_hash);
+
+ALTER TABLE rate_limit_counters
+    ADD CONSTRAINT rate_limit_counters_pkey PRIMARY KEY (key);
+
+ALTER TABLE schedule_entries
+    ADD CONSTRAINT schedule_entries_pkey PRIMARY KEY (id);
+
+ALTER TABLE service_account_keys
+    ADD CONSTRAINT service_account_keys_key_hash_key UNIQUE (key_hash);
+
+ALTER TABLE service_account_keys
+    ADD CONSTRAINT service_account_keys_pkey PRIMARY KEY (id);
+
+ALTER TABLE service_accounts
+    ADD CONSTRAINT service_accounts_name_key UNIQUE (name);
+
+ALTER TABLE service_accounts
+    ADD CONSTRAINT service_accounts_pkey PRIMARY KEY (id);
+
+ALTER TABLE sessions
+    ADD CONSTRAINT sessions_pkey PRIMARY KEY (id);
+
+ALTER TABLE sessions
+    ADD CONSTRAINT sessions_token_hash_key UNIQUE (token_hash);
+
+ALTER TABLE settings
+    ADD CONSTRAINT settings_pkey PRIMARY KEY (key);
+
+ALTER TABLE team_members
+    ADD CONSTRAINT team_members_pkey PRIMARY KEY (team_id, user_id);
+
+ALTER TABLE teams
+    ADD CONSTRAINT teams_name_key UNIQUE (name);
+
+ALTER TABLE teams
+    ADD CONSTRAINT teams_external_id_key UNIQUE (external_id);
+
+ALTER TABLE teams
+    ADD CONSTRAINT teams_pkey PRIMARY KEY (id);
+
+ALTER TABLE user_identities
+    ADD CONSTRAINT user_identities_issuer_subject_key UNIQUE (issuer, subject);
+
+ALTER TABLE user_identities
+    ADD CONSTRAINT user_identities_pkey PRIMARY KEY (id);
+
+ALTER TABLE users
+    ADD CONSTRAINT users_email_key UNIQUE (email);
+
+ALTER TABLE users
+    ADD CONSTRAINT users_pkey PRIMARY KEY (id);
+
+ALTER TABLE users
+    ADD CONSTRAINT users_username_key UNIQUE (username);
+
+CREATE INDEX alerts_archived_at_idx ON alerts USING btree (archived_at);
+
+CREATE INDEX alerts_integration_idx ON alerts USING btree (integration_id, received_at) WHERE (integration_id IS NOT NULL);
+
+CREATE INDEX alerts_name_idx ON alerts USING btree (name);
+
+CREATE INDEX alerts_received_at_idx ON alerts USING btree (received_at DESC);
+
+CREATE INDEX alerts_status_idx ON alerts USING btree (status);
+
+CREATE UNIQUE INDEX alerts_team_fingerprint_idx ON alerts USING btree (team_id, fingerprint);
+
+CREATE INDEX alerts_team_received_idx ON alerts USING btree (team_id, received_at DESC);
+
+CREATE INDEX deadman_switches_team_idx ON deadman_switches USING btree (team_id);
+
+CREATE INDEX device_logins_expires_idx ON device_logins USING btree (expires_at);
+
+CREATE INDEX escalation_targets_level_idx ON escalation_targets USING btree (level_id);
+
+CREATE INDEX idx_sessions_user ON sessions USING btree (user_id);
+
+CREATE INDEX incident_ack_tokens_expires_idx ON incident_ack_tokens USING btree (expires_at);
+
+CREATE INDEX incident_alerts_alert_id_idx ON incident_alerts USING btree (alert_id);
+
+CREATE INDEX incident_events_actor_service_account_id_idx ON incident_events USING btree (actor_service_account_id);
+
+CREATE INDEX incident_events_actor_user_id_idx ON incident_events USING btree (actor_user_id);
+
+CREATE INDEX incident_events_incident_idx ON incident_events USING btree (incident_id, created_at);
+
+CREATE INDEX incident_events_service_account_id_idx ON incident_events USING btree (service_account_id);
+
+CREATE INDEX incidents_acknowledged_by_service_account_id_idx ON incidents USING btree (acknowledged_by_service_account_id);
+
+CREATE INDEX incidents_archived_at_idx ON incidents USING btree (archived_at);
+
+CREATE INDEX incidents_escalation_idx ON incidents USING btree (escalation_level_at) WHERE ((resolved_at IS NULL) AND (status = 'triggered'::text));
+
+CREATE UNIQUE INDEX incidents_open_group_key_idx ON incidents USING btree (team_id, group_key) WHERE (resolved_at IS NULL);
+
+CREATE INDEX incidents_signature_idx ON incidents USING btree (team_id, signature, triggered_at DESC);
+
+CREATE INDEX incidents_status_idx ON incidents USING btree (status);
+
+CREATE INDEX incidents_team_triggered_idx ON incidents USING btree (team_id, triggered_at DESC);
+
+CREATE INDEX incidents_triggered_at_idx ON incidents USING btree (triggered_at DESC);
+
+CREATE INDEX integrations_team_idx ON integrations USING btree (team_id);
+
+-- A name identifies an integration (and a switch) within its team, so a client
+-- that manages them declaratively can look one up by name instead of listing
+-- and matching.
+CREATE UNIQUE INDEX integrations_team_name_key ON integrations (team_id, name);
+CREATE UNIQUE INDEX deadman_switches_team_name_key ON deadman_switches (team_id, name);
+
+CREATE INDEX invites_team_idx ON invites USING btree (team_id);
+
+CREATE INDEX notifications_incident_idx ON notifications USING btree (incident_id, id DESC);
+
+CREATE INDEX notifications_pending_idx ON notifications USING btree (send_after) WHERE (sent_at IS NULL);
+
+CREATE INDEX oidc_logins_expires_idx ON oidc_logins USING btree (expires_at);
+
+CREATE INDEX schedule_entries_date_idx ON schedule_entries USING btree (date);
+
+CREATE UNIQUE INDEX schedule_entries_team_date_idx ON schedule_entries USING btree (team_id, date);
+
+CREATE INDEX service_account_keys_service_account_id_idx ON service_account_keys USING btree (service_account_id);
+
+CREATE INDEX service_accounts_team_id_idx ON service_accounts USING btree (team_id);
+
+CREATE INDEX team_members_user_idx ON team_members USING btree (user_id);
+
+CREATE INDEX user_identities_user_idx ON user_identities USING btree (user_id);
+
+CREATE INDEX users_is_admin_idx ON users USING btree (is_admin) WHERE is_admin;
+
+ALTER TABLE alerts
+    ADD CONSTRAINT alerts_integration_id_fkey FOREIGN KEY (integration_id) REFERENCES integrations(id) ON DELETE SET NULL;
+
+ALTER TABLE alerts
+    ADD CONSTRAINT alerts_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE api_keys
+    ADD CONSTRAINT api_keys_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE deadman_switches
+    ADD CONSTRAINT deadman_switches_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE device_logins
+    ADD CONSTRAINT device_logins_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE escalation_levels
+    ADD CONSTRAINT escalation_levels_team_id_fkey FOREIGN KEY (team_id) REFERENCES escalation_policies(team_id) ON DELETE CASCADE;
+
+ALTER TABLE escalation_policies
+    ADD CONSTRAINT escalation_policies_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE escalation_targets
+    ADD CONSTRAINT escalation_targets_level_id_fkey FOREIGN KEY (level_id) REFERENCES escalation_levels(id) ON DELETE CASCADE;
+
+ALTER TABLE escalation_targets
+    ADD CONSTRAINT escalation_targets_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_ack_tokens
+    ADD CONSTRAINT incident_ack_tokens_incident_id_fkey FOREIGN KEY (incident_id) REFERENCES incidents(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_ack_tokens
+    ADD CONSTRAINT incident_ack_tokens_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_alerts
+    ADD CONSTRAINT incident_alerts_alert_id_fkey FOREIGN KEY (alert_id) REFERENCES alerts(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_alerts
+    ADD CONSTRAINT incident_alerts_incident_id_fkey FOREIGN KEY (incident_id) REFERENCES incidents(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_actor_service_account_id_fkey FOREIGN KEY (actor_service_account_id) REFERENCES service_accounts(id) ON DELETE SET NULL;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_actor_user_id_fkey FOREIGN KEY (actor_user_id) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_alert_id_fkey FOREIGN KEY (alert_id) REFERENCES alerts(id) ON DELETE SET NULL;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_incident_id_fkey FOREIGN KEY (incident_id) REFERENCES incidents(id) ON DELETE CASCADE;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_service_account_id_fkey FOREIGN KEY (service_account_id) REFERENCES service_accounts(id) ON DELETE SET NULL;
+
+ALTER TABLE incident_events
+    ADD CONSTRAINT incident_events_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE incidents
+    ADD CONSTRAINT incidents_acknowledged_by_fkey FOREIGN KEY (acknowledged_by) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE incidents
+    ADD CONSTRAINT incidents_acknowledged_by_service_account_id_fkey FOREIGN KEY (acknowledged_by_service_account_id) REFERENCES service_accounts(id) ON DELETE SET NULL;
+
+ALTER TABLE incidents
+    ADD CONSTRAINT incidents_assigned_to_fkey FOREIGN KEY (assigned_to) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE incidents
+    ADD CONSTRAINT incidents_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE integrations
+    ADD CONSTRAINT integrations_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE invites
+    ADD CONSTRAINT invites_created_by_fkey FOREIGN KEY (created_by) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE invites
+    ADD CONSTRAINT invites_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE notifications
+    ADD CONSTRAINT notifications_incident_id_fkey FOREIGN KEY (incident_id) REFERENCES incidents(id) ON DELETE CASCADE;
+
+ALTER TABLE notifications
+    ADD CONSTRAINT notifications_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE schedule_entries
+    ADD CONSTRAINT schedule_entries_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE schedule_entries
+    ADD CONSTRAINT schedule_entries_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE service_account_keys
+    ADD CONSTRAINT service_account_keys_service_account_id_fkey FOREIGN KEY (service_account_id) REFERENCES service_accounts(id) ON DELETE CASCADE;
+
+ALTER TABLE service_accounts
+    ADD CONSTRAINT service_accounts_created_by_fkey FOREIGN KEY (created_by) REFERENCES users(id) ON DELETE SET NULL;
+
+ALTER TABLE service_accounts
+    ADD CONSTRAINT service_accounts_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE sessions
+    ADD CONSTRAINT sessions_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE team_members
+    ADD CONSTRAINT team_members_team_id_fkey FOREIGN KEY (team_id) REFERENCES teams(id) ON DELETE CASCADE;
+
+ALTER TABLE team_members
+    ADD CONSTRAINT team_members_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE user_identities
+    ADD CONSTRAINT user_identities_user_id_fkey FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE;
+
+ALTER TABLE users
+    ADD CONSTRAINT users_invited_via_fkey FOREIGN KEY (invited_via) REFERENCES invites(id) ON DELETE SET NULL;

internal/models/alert.go

@@ -33,7 +33,7 @@ type Alert struct {
 	// refreshed. The sweeper stale-dates against it (see expireStale), API
 	// clients render it, and GET /api/alerts is ordered by it. Anything that
 	// stops the webhook handler from advancing it on a re-send is a breaking
-	// change — see "received_at is a liveness heartbeat" in the README and
+	// change — see "received_at is a liveness heartbeat" in docs/api.md and
 	// TestWebhook_ResendBumpsReceivedAt.
 	ReceivedAt time.Time `json:"received_at"`
 
@@ -52,7 +52,7 @@ type Alert struct {
 	// inferred. Under "expiry" nothing ever reported an end, so EndsAt is only
 	// an upper bound (see expireStale) and ReceivedAt is the more truthful
 	// signal. Treat the value set as open — see "resolution_source says how much
-	// to trust ends_at" in the README, and TestWebhook_ResolvedSetsSource /
+	// to trust ends_at" in docs/api.md, and TestWebhook_ResolvedSetsSource /
 	// TestExpiry_StaleFiringAlert.
 	ResolutionSource *string `json:"resolution_source,omitempty"`
 

internal/models/team.go

@@ -9,6 +9,10 @@ type Team struct {
 	Name      string    `json:"name"`
 	CreatedAt time.Time `json:"created_at"`
 
+	// ExternalID identifies a team managed by automation; see handleCreateTeam.
+	// Shown to instance service accounts and admins only.
+	ExternalID *string `json:"external_id,omitempty"`
+
 	// Role is the caller's own role in this team, populated when a team is
 	// listed for a particular person. Empty when nobody in particular is
 	// asking, as in the admin listing.

internal/oidc/grants.go

@@ -102,6 +102,3 @@ func rank(role string) int {
 	}
 	return 0
 }
-
-// HigherRole reports whether role a outranks role b.
-func HigherRole(a, b string) bool { return rank(a) > rank(b) }

internal/web/static/js/format.js

@@ -123,7 +123,7 @@ export function initial(name) {
 // convention. It comes from Prometheus's externalLabels, so it is on every
 // alert; an incident carries it only when it is in Alertmanager's group_by,
 // which is also what keeps two clusters' identical alerts from merging into one
-// incident (see the README, "Several clusters, one team").
+// incident (see docs/incidents.md, "Several clusters, one team").
 export const ORIGIN_LABEL = 'cluster';
 
 export function originOf(labels) {