setrip/docs/archive/ADMIN_SYSTEM_HEALTH_ROADMAP.md

Setrip — Admin System Health Roadmap (ARCHIVED — DELIVERED 2026-05-18)

Admin perlu visibilitas atas job otomatis (cron) dan alert untuk state stale.

---

Status delivery

Phase	Status	Catatan
Phase 1 — Cron Run Log	✅ Delivered	Model CronRun, helper runCron(), wire ke cron existing.
Phase 2 — System Status Page	✅ Delivered	/admin/system tampilkan summary cron + 20 run terakhir + health badge.
Phase 3 — Stale State Alerts	✅ Delivered	systemHealthService.detectStale() cek 4 kategori (Payment AWAITING > 25h, AWAITING_PAY past departure, Payout HELD overdue, Refund APPROVED > 7d). Banner di /admin/system.
Phase 4 — Discord Webhook Notify	✅ Delivered	notifyAdmins() POST ke ADMIN_ALERT_WEBHOOK_URL. Trigger otomatis saat cron FAILED via runCron.

---

Phase 1 — Cron Run Log ✅

#	Item	Status	File
1.1	Model CronRun + enum CronRunStatus + migration	✅	prisma/schema.prisma + prisma/migrations/20260518170000_add_cron_run/
1.2	Helper runCron(jobName, fn) — auto create RUNNING row → SUCCESS/FAILED	✅	lib/cron-runner.ts
1.3	Wire runCron di auto-complete-trips cron	✅	app/api/cron/auto-complete-trips/route.ts

---

Phase 2 — System Status Page ✅

#	Item	Status	File
2.1	Per-job summary: last run, last success, count 7d, error count 7d	✅	app/admin/system/page.tsx
2.2	20 cron run terakhir di table bawah	✅	app/admin/system/page.tsx
2.3	Health badge (🟢 OK < 25h, 🟡 STALE, 🔴 FAILED)	✅	app/admin/system/page.tsx
2.4	Link "System" di admin navbar	✅	components/admin/admin-sidebar.tsx

---

Phase 3 — Stale State Alerts ✅

#	Item	Status	File
3.1	systemHealthService.detectStale() return 4 count	✅	server/services/system-health.service.ts
3.2	Banner alerts kuning di /admin/system kalau ada count > 0	✅	app/admin/system/page.tsx
3.3	Link tiap alert ke filtered list page yang relevan	✅ (untuk Payout HELD & Refund APPROVED)	app/admin/system/page.tsx

Threshold draft (review setelah jalan 1-2 minggu):

• Payment MIDTRANS AWAITING createdAt > 25 jam — suspect webhook tertunda
• Booking AWAITING_PAY dengan trip.date < today — peserta lupa bayar
• Payout HELD dengan heldUntil > 1 hari lewat — cron release tidak jalan
• Refund APPROVED reviewedAt > 7 hari — admin lupa process

---

Phase 4 — Discord Webhook Notify ✅

#	Item	Status	File
4.1	Helper notifyAdmins(message) — POST ke Discord webhook URL dari env	✅	lib/admin-notify.ts
4.2	Trigger notify di runCron saat FAILED (fire-and-forget)	✅	lib/cron-runner.ts
4.3	Trigger notify dari systemHealthService.detectStale rate-limited	⏳	Skip — admin sudah lihat banner di /admin/system saat buka pagi. Push notif harian baru worth it kalau admin sering miss; bisa ditambah belakangan.

Tindakan manual ops:

1. Apply migration (sudah di Phase 1).
2. (Opsional) Buat Discord channel internal + webhook URL → set env ADMIN_ALERT_WEBHOOK_URL di PM2/server. Tanpa env, notifyAdmins no-op.
3. Test alert: trigger cron secara sengaja fail (mis. matikan DB sebentar) → cek Discord channel menerima 🚨 message.