Cron & Scheduler Expert (2026 Edition)
English | Bahasa Indonesia
<a name="english"></a>
English
Orchestration & Integration
Connects and orchestrates with relevant domain skills like brainstorming, zero-to-prod-orchestrator, and session-memory-manager to ensure cohesive execution.
Description
Production-grade guide for implementing scheduled tasks, cron jobs, and recurring background work in modern web applications. Covers Vercel Cron Jobs, Cloudflare Workers Cron Triggers, Inngest scheduled functions, node-cron, BullMQ repeatable jobs, distributed cron locking (Redis-based), timezone-aware scheduling, and common use cases (cleanup jobs, report generation, health checks, subscription renewals).
Trigger Conditions
Activate this skill when:
- Setting up scheduled/recurring tasks (daily reports, cleanup, health checks).
- Configuring Vercel Cron Jobs or Cloudflare Workers Cron Triggers.
- Implementing subscription renewal checks or billing cycle jobs.
- Building data aggregation or ETL pipelines on a schedule.
- Needing distributed cron locking to prevent duplicate execution.
- Scheduling email digests or notification batches.
Platform Selection Guide
| Platform | Best For | Max Frequency | Invocation Model |
|---|
| Vercel Cron | Vercel-hosted Next.js apps | 1/min (Pro), 1/day (Hobby) | HTTP endpoint trigger |
| Cloudflare Workers Cron | Edge-first apps | 1/min | Worker script trigger |
| Inngest | Complex workflows, fan-out | Any (event-driven) | Serverless function |
| BullMQ Repeatable | Self-hosted, Redis-backed | Any | Worker process |
| node-cron | Simple Node.js servers | Any | In-process |
| Trigger.dev v3 | Durable scheduled tasks | Any | Serverless function |
Recommendation: Use Vercel Cron for Next.js apps on Vercel. Use Inngest for complex multi-step scheduled workflows. Use BullMQ repeatable jobs for self-hosted environments.
1. Vercel Cron Jobs
// vercel.json
{
"crons": [
{
"path": "/api/cron/daily-cleanup",
"schedule": "0 3 * * *"
},
{
"path": "/api/cron/hourly-health-check",
"schedule": "0 * * * *"
},
{
"path": "/api/cron/weekly-report",
"schedule": "0 9 * * 1"
}
]
}
// app/api/cron/daily-cleanup/route.ts
import { NextResponse } from 'next/server';
export async function GET(request: Request) {
// Verify the request is from Vercel Cron (not public)
const authHeader = request.headers.get('authorization');
if (authHeader !== `Bearer ${process.env.CRON_SECRET}`) {
return NextResponse.json({ error: 'Unauthorized' }, { status: 401 });
}
try {
// Cleanup expired sessions
const deletedSessions = await db.delete(sessions)
.where(lt(sessions.expiresAt, new Date()));
// Cleanup orphaned uploads (older than 24h, not linked to any record)
const deletedUploads = await db.delete(uploads)
.where(and(
isNull(uploads.linkedRecordId),
lt(uploads.createdAt, new Date(Date.now() - 86400000)),
));
return NextResponse.json({
ok: true,
cleaned: { sessions: deletedSessions.rowCount, uploads: deletedUploads.rowCount },
});
} catch (error) {
console.error('Daily cleanup failed:', error);
return NextResponse.json({ ok: false, error: 'Cleanup failed' }, { status: 500 });
}
}
export const runtime = 'nodejs';
export const maxDuration = 60; // Allow up to 60s for cleanup
2. Inngest Scheduled Functions
// inngest/functions/weekly-report.ts
import { inngest } from '@/lib/inngest';
export const weeklyReportJob = inngest.createFunction(
{
id: 'weekly-usage-report',
name: 'Generate Weekly Usage Report',
},
{ cron: 'TZ=Asia/Jakarta 0 9 * * 1' }, // Every Monday 9 AM WIB
async ({ step }) => {
// Step 1: Gather data
const metrics = await step.run('gather-metrics', async () => {
return await db.query.usageMetrics.findMany({
where: gte(usageMetrics.createdAt, subDays(new Date(), 7)),
});
});
// Step 2: Generate report
const report = await step.run('generate-report', async () => {
return generateWeeklyReport(metrics);
});
// Step 3: Send to admins
await step.run('send-report', async () => {
const admins = await db.query.users.findMany({
where: eq(users.role, 'admin'),
});
for (const admin of admins) {
await emailQueue.add('weekly-report', { to: admin.email, report });
}
});
return { sent: true, reportId: report.id };
},
);
3. BullMQ Repeatable Jobs
// queues/scheduled-jobs.ts
import { Queue, Worker } from 'bullmq';
const scheduledQueue = new Queue('scheduled', {
connection: { host: process.env.REDIS_HOST },
});
// Add repeatable jobs on startup
export async function registerScheduledJobs() {
// Daily cleanup at 3 AM
await scheduledQueue.add('daily-cleanup', {}, {
repeat: { pattern: '0 3 * * *', tz: 'Asia/Jakarta' },
jobId: 'daily-cleanup', // Prevent duplicates on restart
});
// Every 5 minutes: health check
await scheduledQueue.add('health-check', {}, {
repeat: { every: 5 * 60 * 1000 }, // 5 minutes in ms
jobId: 'health-check',
});
// Monthly: subscription renewal check (1st of each month at midnight)
await scheduledQueue.add('subscription-renewal', {}, {
repeat: { pattern: '0 0 1 * *', tz: 'Asia/Jakarta' },
jobId: 'subscription-renewal',
});
}
// Worker
const worker = new Worker('scheduled', async (job) => {
switch (job.name) {
case 'daily-cleanup':
await performDailyCleanup();
break;
case 'health-check':
await performHealthCheck();
break;
case 'subscription-renewal':
await checkSubscriptionRenewals();
break;
}
}, { connection: { host: process.env.REDIS_HOST } });
4. Distributed Cron Locking (Redis)
// lib/cron-lock.ts
import { Redis } from 'ioredis';
const redis = new Redis(process.env.REDIS_URL!);
/**
* Acquire a distributed lock to prevent duplicate cron execution
* across multiple server instances.
*/
export async function acquireCronLock(
jobName: string,
ttlSeconds: number = 300,
): Promise<boolean> {
const lockKey = `cron-lock:${jobName}`;
// SET NX (only if not exists) with TTL
const result = await redis.set(lockKey, Date.now().toString(), 'EX', ttlSeconds, 'NX');
return result === 'OK';
}
export async function releaseCronLock(jobName: string): Promise<void> {
await redis.del(`cron-lock:${jobName}`);
}
// Usage in cron handler
export async function GET(request: Request) {
const locked = await acquireCronLock('daily-cleanup', 300);
if (!locked) {
return Response.json({ skipped: true, reason: 'Already running on another instance' });
}
try {
await performDailyCleanup();
return Response.json({ ok: true });
} finally {
await releaseCronLock('daily-cleanup');
}
}
Common Cron Patterns
| Schedule | Cron Expression | Description |
|---|
| Every minute | * * * * * | Health checks, queue monitoring |
| Every 5 minutes | */5 * * * * | Cache warming, metric aggregation |
| Every hour | 0 * * * * | Data sync, report generation |
| Daily at 3 AM | 0 3 * * * | Cleanup, backups |
| Weekly Monday 9 AM | 0 9 * * 1 | Weekly reports |
| Monthly 1st at midnight | 0 0 1 * * | Billing cycles, subscription renewals |
| Weekdays at 8 AM | 0 8 * * 1-5 | Business notifications |
Common Pitfalls to Avoid
| Anti-Pattern | Problem | Correct Approach |
|---|
| No distributed locking | Duplicate execution across instances | Use Redis-based NX lock |
| Long-running cron without timeout | Resource exhaustion, overlapping runs | Set maxDuration and use lock TTL |
| No monitoring of cron failures | Silent failures go unnoticed | Log results, alert on failures |
| Hardcoded timezone | Jobs run at wrong time after DST | Always use explicit tz parameter |
| Cron job does too much work | Timeout, partial completion | Break into smaller steps (Inngest/Temporal) |
Integration with Other Skills
async-queue-temporal-expert — BullMQ repeatable jobs, Inngest scheduled functions
cloud-hosting-expert — Vercel Cron, Cloudflare Workers Cron configuration
saas-billing — Subscription renewal checks, billing cycle jobs
email-notification-expert — Scheduled email digests, daily summaries
data-telemetry-expert — Scheduled metric aggregation and reporting
logging-error-tracking-expert — Cron job failure monitoring
<a name="bahasa-indonesia"></a>
Bahasa Indonesia
Integrasi Orkestrasi
Terhubung dan mengorkestrasi skill domain yang relevan seperti brainstorming, zero-to-prod-orchestrator, dan session-memory-manager untuk memastikan eksekusi yang kohesif.
async-queue-temporal-expert — BullMQ repeatable jobs dan alur kerja Inngest.
cloud-hosting-expert — Konfigurasi Vercel Cron dan Cloudflare Workers Cron.
saas-billing — Pemeriksaan siklus perpanjangan invoice langganan berkala.
email-notification-expert — Pengiriman digest email harian/mingguan terjadwal.
Deskripsi
Panduan tingkat produksi untuk mengimplementasikan tugas terjadwal, cron job, dan pekerjaan latar belakang berulang di aplikasi web modern. Mencakup Vercel Cron Jobs, Cloudflare Workers Cron Triggers, Inngest scheduled functions, BullMQ repeatable jobs, distributed cron locking (berbasis Redis), penjadwalan sadar zona waktu, dan kasus penggunaan umum (pembersihan data, pembuatan laporan, health check, perpanjangan langganan).
Kondisi Pemicu
Aktifkan skill ini ketika:
- Menyiapkan tugas terjadwal/berulang (laporan harian, pembersihan, health check).
- Mengonfigurasi Vercel Cron Jobs atau Cloudflare Workers Cron Triggers.
- Mengimplementasikan pemeriksaan perpanjangan langganan atau siklus billing.
- Membangun pipeline agregasi data atau ETL pada jadwal tertentu.
- Membutuhkan penguncian cron terdistribusi untuk mencegah eksekusi duplikat.
Panduan Pemilihan Solusi Cron
| Platform | Pilihan Terbaik Untuk | Frekuensi Maksimum | Model Pemicu |
|---|
| Vercel Cron | Aplikasi Next.js di Vercel | 1x/menit (Pro), 1x/hari (Hobby) | Pemicu HTTP endpoint internal |
| Cloudflare Workers Cron | Arsitektur Edge murni | 1x/menit | Scheduled event script worker |
| Inngest | Alur bertahap (step-functions) multi-langkah | Fleksibel & event-driven | Serverless durable execution |
| BullMQ Repeatable | Aplikasi mandiri berbasis Redis | Fleksibel | Antrean worker Node.js terpisah |
Implementasi Pola Kunci (TypeScript)
1. Vercel Cron dengan Proteksi Autentikasi Rahasia (CRON_SECRET)
// vercel.json
{
"crons": [
{
"path": "/api/cron/pembersihan-harian",
"schedule": "0 3 * * *"
}
]
}
// app/api/cron/pembersihan-harian/route.ts
import { NextResponse } from 'next/server';
export async function GET(request: Request) {
// Verifikasi permintaan resmi berasal dari scheduler Vercel
const authHeader = request.headers.get('authorization');
if (authHeader !== `Bearer ${process.env.CRON_SECRET}`) {
return NextResponse.json({ error: 'Tidak memiliki otorisasi' }, { status: 401 });
}
try {
// Jalankan operasi pembersihan sesi kadaluarsa
const hasil = await db.delete(sesiPengguna)
.where(lt(sesiPengguna.kadaluarsaPada, new Date()));
return NextResponse.json({ ok: true, terhapus: hasil.rowCount });
} catch (error) {
console.error('Pembersihan cron gagal:', error);
return NextResponse.json({ ok: false, error: 'Gagal membersihkan data' }, { status: 500 });
}
}
export const runtime = 'nodejs';
export const maxDuration = 60; // Izinkan eksekusi hingga 60 detik
2. Inngest Scheduled Functions Sadar Zona Waktu (TZ=Asia/Jakarta)
import { inngest } from '@/lib/inngest';
export const jobLaporanMingguan = inngest.createFunction(
{ id: 'laporan-mingguan-admin', name: 'Kirim Laporan Penggunaan Mingguan' },
{ cron: 'TZ=Asia/Jakarta 0 9 * * 1' }, // Setiap Senin pukul 09:00 WIB
async ({ step }) => {
const dataMetrik = await step.run('ambil-data-metrik', async () => {
return await kumpulkanMetrik7HariTerakhir();
});
const fileLaporan = await step.run('buat-dokumen-laporan', async () => {
return await generateReportPdf(dataMetrik);
});
await step.run('kirimkan-email-admin', async () => {
await emailQueue.add('kirim-laporan', { file: fileLaporan });
});
return { status: 'sukses', terkirim: true };
}
);
3. Distributed Cron Locking Berbasis Redis (SET NX)
import { Redis } from 'ioredis';
const redis = new Redis(process.env.REDIS_URL!);
export async function dapatkanKunciCron(namaTugas: string, durasiDetik: number = 300): Promise<boolean> {
const kunci = `kunci-cron:${namaTugas}`;
const status = await redis.set(kunci, Date.now().toString(), 'EX', durasiDetik, 'NX');
return status === 'OK';
}
export async function lepaskanKunciCron(namaTugas: string): Promise<void> {
await redis.del(`kunci-cron:${namaTugas}`);
}
Praktik Terbaik Keamanan Cron
- Gunakan Distributed Lock: Mencegah eksekusi ganda ketika aplikasi berjalan pada banyak instans server (cluster/autoscaling).
- Tentukan
maxDuration: Hindari cron menggantung tanpa henti yang menghabiskan alokasi sumber daya komputasi.
- Zona Waktu Eksplisit: Selalu sertakan parameter timezone (
TZ=Asia/Jakarta) agar jadwal tidak bergeser saat Daylight Saving Time (DST) di cloud server.