feat(broker): production hardening — caps, limits, metrics, logging

Adds the minimum ops surface area for a production broker without
over-engineering. All new config knobs are env-var driven with sane
defaults.

New modules:
- logger.ts: structured JSON logs (one line, stderr, ready for
  Loki/Datadog ingestion without preprocessing)
- metrics.ts: in-process Prometheus counters + gauges, exposed at
  GET /metrics. Tracks connections, messages, queue depth, TTL
  sweeps, hook requests, DB health.
- rate-limit.ts: token-bucket rate limiter keyed by (pid, cwd).
  Applied to POST /hook/set-status at 30/min default.
- db-health.ts: Postgres ping loop with exponential-backoff retry.
  GET /health returns 503 while DB is down.
- build-info.ts: version + gitSha (from GIT_SHA env or `git rev-parse`
  fallback) + uptime, surfaced on /health.

Behavior changes:
- Connection caps: MAX_CONNECTIONS_PER_MESH (default 100). Exceed →
  close(1008, "capacity") + metric increment.
- Message size: MAX_MESSAGE_BYTES (default 65536). WS applies it via
  `ws.maxPayload`. Hook POST bodies cap out with 413.
- Structured logs everywhere replacing the old `log()` helper.
- Env validation stricter: DATABASE_URL required + regex-checked for
  postgres:// prefix.

New endpoints:
- GET /health → {status, db, version, gitSha, uptime}. 503 if DB down.
- GET /metrics → Prometheus text format.

Verified: 21/21 tests still pass. Hit /health + /metrics live —
gitSha resolves correctly via `git rev-parse --short HEAD` in dev.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

apps/broker/src/metrics.ts

@@ -0,0 +1,121 @@
+/**
+ * Minimal in-process metrics, exposed as Prometheus plaintext.
+ *
+ * Intentionally no external deps — we track a handful of counters
+ * and gauges that matter for broker ops. Scraped by /metrics.
+ */
+
+type Labels = Record<string, string | number>;
+
+class Counter {
+  private values = new Map<string, number>();
+  constructor(
+    public name: string,
+    public help: string,
+  ) {}
+  inc(labels: Labels = {}, by = 1): void {
+    const key = labelKey(labels);
+    this.values.set(key, (this.values.get(key) ?? 0) + by);
+  }
+  toText(): string {
+    const lines = [`# HELP ${this.name} ${this.help}`, `# TYPE ${this.name} counter`];
+    if (this.values.size === 0) {
+      lines.push(`${this.name} 0`);
+    } else {
+      for (const [key, v] of this.values) {
+        lines.push(`${this.name}${key} ${v}`);
+      }
+    }
+    return lines.join("\n");
+  }
+}
+
+class Gauge {
+  private values = new Map<string, number>();
+  constructor(
+    public name: string,
+    public help: string,
+  ) {}
+  set(value: number, labels: Labels = {}): void {
+    this.values.set(labelKey(labels), value);
+  }
+  inc(labels: Labels = {}, by = 1): void {
+    const key = labelKey(labels);
+    this.values.set(key, (this.values.get(key) ?? 0) + by);
+  }
+  dec(labels: Labels = {}, by = 1): void {
+    this.inc(labels, -by);
+  }
+  toText(): string {
+    const lines = [`# HELP ${this.name} ${this.help}`, `# TYPE ${this.name} gauge`];
+    if (this.values.size === 0) {
+      lines.push(`${this.name} 0`);
+    } else {
+      for (const [key, v] of this.values) {
+        lines.push(`${this.name}${key} ${v}`);
+      }
+    }
+    return lines.join("\n");
+  }
+}
+
+function labelKey(labels: Labels): string {
+  const entries = Object.entries(labels);
+  if (entries.length === 0) return "";
+  const parts = entries
+    .sort(([a], [b]) => a.localeCompare(b))
+    .map(([k, v]) => `${k}="${String(v).replace(/"/g, '\\"')}"`)
+    .join(",");
+  return `{${parts}}`;
+}
+
+export const metrics = {
+  connectionsTotal: new Counter(
+    "broker_connections_total",
+    "Total WS connection attempts",
+  ),
+  connectionsRejected: new Counter(
+    "broker_connections_rejected_total",
+    "WS connections refused (auth failure, capacity, etc.)",
+  ),
+  connectionsActive: new Gauge(
+    "broker_connections_active",
+    "Currently connected peers",
+  ),
+  messagesRoutedTotal: new Counter(
+    "broker_messages_routed_total",
+    "Messages successfully queued + routed",
+  ),
+  messagesRejectedTotal: new Counter(
+    "broker_messages_rejected_total",
+    "Messages rejected (size, auth, malformed)",
+  ),
+  queueDepth: new Gauge(
+    "broker_queue_depth",
+    "Undelivered messages currently in the queue",
+  ),
+  ttlSweepsTotal: new Counter(
+    "broker_ttl_sweeps_total",
+    "TTL sweeper runs completed",
+  ),
+  hookRequestsTotal: new Counter(
+    "broker_hook_requests_total",
+    "POST /hook/set-status requests received",
+  ),
+  hookRequestsRateLimited: new Counter(
+    "broker_hook_requests_rate_limited_total",
+    "POST /hook/set-status rejected by rate limit",
+  ),
+  dbHealthy: new Gauge(
+    "broker_db_healthy",
+    "1 if Postgres connection is up, 0 if not",
+  ),
+};
+
+export function metricsToText(): string {
+  return (
+    Object.values(metrics)
+      .map((m) => m.toText())
+      .join("\n") + "\n"
+  );
+}