feat(monitoring): add snapshot collector to queue worker

Periodic metric collection running in the queue-worker PM2 process.
Collects system metrics every 60s (configurable), stores them in
MonitoringSnapshots, and evaluates alert rules against each snapshot.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

ecosystem.config.cjs

@@ -39,6 +39,8 @@ module.exports = {
         // Worker aktivieren/deaktivieren
         QUEUE_ENABLE_EMAIL: 'true',
         QUEUE_ENABLE_PDF: 'true',
+        QUEUE_ENABLE_MONITORING: 'true',
+        MONITORING_SNAPSHOT_INTERVAL: '60000',
         // PDF-spezifische Optionen
         PDF_OUTPUT_DIR: '/tmp/payload-pdfs',
         // PDF_GENERATION_DISABLED: 'true', // Für Tests

scripts/run-queue-worker.ts

@@ -43,6 +43,9 @@ async function main() {
   const ENABLE_RETENTION_SCHEDULER = process.env.QUEUE_ENABLE_RETENTION_SCHEDULER !== 'false'
   const ENABLE_YOUTUBE_UPLOAD_WORKER = process.env.QUEUE_ENABLE_YOUTUBE_UPLOAD !== 'false'
 
+  // Monitoring (optional)
+  const ENABLE_MONITORING = process.env.QUEUE_ENABLE_MONITORING !== 'false'
+
   console.log('='.repeat(50))
   console.log('[QueueRunner] Starting queue workers...')
   console.log(`[QueueRunner] PID: ${process.pid}`)
@@ -52,6 +55,7 @@ async function main() {
   console.log(`[QueueRunner] Retention Worker: ${ENABLE_RETENTION_WORKER ? 'enabled' : 'disabled'}`)
   console.log(`[QueueRunner] Retention Scheduler: ${ENABLE_RETENTION_SCHEDULER ? 'enabled' : 'disabled'}`)
   console.log(`[QueueRunner] YouTube Upload Worker: ${ENABLE_YOUTUBE_UPLOAD_WORKER ? 'enabled' : 'disabled'}`)
+  console.log(`[QueueRunner] Monitoring Collector: ${ENABLE_MONITORING ? 'enabled' : 'disabled'}`)
   console.log('='.repeat(50))
 
   // Workers starten
@@ -78,6 +82,15 @@ async function main() {
     startYouTubeUploadWorker()
   }
 
+  // Monitoring Snapshot Collector
+  if (ENABLE_MONITORING) {
+    const { startSnapshotCollector, stopSnapshotCollector } = await import('../src/lib/monitoring/snapshot-collector')
+    await startSnapshotCollector()
+
+    // Store stop function for shutdown
+    ;(global as any).__stopSnapshotCollector = stopSnapshotCollector
+  }
+
   // Graceful Shutdown
   async function shutdown(signal: string) {
     console.log(`\n[QueueRunner] Received ${signal}, shutting down gracefully...`)
@@ -97,6 +110,9 @@ async function main() {
       if (ENABLE_YOUTUBE_UPLOAD_WORKER) {
         stopPromises.push(stopYouTubeUploadWorker())
       }
+      if (ENABLE_MONITORING && (global as any).__stopSnapshotCollector) {
+        stopPromises.push((global as any).__stopSnapshotCollector())
+      }
 
       await Promise.all(stopPromises)
       console.log('[QueueRunner] All workers stopped')

src/lib/monitoring/snapshot-collector.ts

@@ -0,0 +1,56 @@
+/**
+ * Snapshot Collector
+ *
+ * Periodische Metrik-Erfassung, laeuft im Queue-Worker PM2-Prozess.
+ * Sammelt alle 60s (konfigurierbar) System-Metriken und speichert
+ * sie in MonitoringSnapshots. Evaluiert dabei Alert-Regeln.
+ */
+
+import { collectMetrics } from './monitoring-service.js'
+import { AlertEvaluator } from './alert-evaluator.js'
+
+let interval: ReturnType<typeof setInterval> | null = null
+const alertEvaluator = new AlertEvaluator()
+
+export async function startSnapshotCollector(): Promise<void> {
+  const INTERVAL = parseInt(process.env.MONITORING_SNAPSHOT_INTERVAL || '60000', 10)
+  console.log(`[SnapshotCollector] Starting (interval: ${INTERVAL}ms)`)
+
+  // Run immediately once, then on interval
+  await collectAndSave()
+
+  interval = setInterval(async () => {
+    await collectAndSave()
+  }, INTERVAL)
+}
+
+async function collectAndSave(): Promise<void> {
+  try {
+    const { getPayload } = await import('payload')
+    const config = (await import('@payload-config')).default
+    const payload = await getPayload({ config })
+
+    const metrics = await collectMetrics()
+
+    await payload.create({
+      collection: 'monitoring-snapshots',
+      data: {
+        timestamp: new Date().toISOString(),
+        ...metrics,
+      },
+    })
+
+    // Evaluate alert rules against collected metrics
+    await alertEvaluator.evaluateRules(payload, metrics)
+  } catch (error) {
+    console.error('[SnapshotCollector] Error:', error)
+  }
+}
+
+export async function stopSnapshotCollector(): Promise<void> {
+  if (interval) {
+    clearInterval(interval)
+    interval = null
+  }
+  console.log('[SnapshotCollector] Stopped')
+}