Cron every hour — `0 * * * *` explained

The minute field is what determines the cadence here. Wildcarding the minute (* * * * *) gives you every minute — 60 firings per hour. Pinning the minute to 0 gives you exactly one firing per hour, at xx:00.

The hour wildcard * says “every hour”. Combined with minute 0, the full schedule is “every hour, at minute 0”.

Linux crontab

0 * * * * /path/to/script.sh
# Or with the @hourly alias:
@hourly /path/to/script.sh

Vercel cron jobs

{
  "crons": [
    { "path": "/api/cron/hourly", "schedule": "0 * * * *" }
  ]
}

AWS EventBridge

cron(0 * * * ? *)
# or:
rate(1 hour)

Quartz (Java)

Trigger trigger = newTrigger()
    .withSchedule(cronSchedule("0 0 * * * ?"))
    .build();

Kubernetes CronJob

spec:
  schedule: "0 * * * *"
  concurrencyPolicy: Forbid

• 30 * * * * — every hour at xx:30.
• 0,30 * * * * — every half hour (twice an hour).
• 0 */2 * * * — every 2 hours at the top of the hour.
• 0 9-17 * * 1-5 — hourly during business hours on weekdays.
• 0 0,12 * * * — at midnight and noon every day (twice daily).
• */15 * * * * — every 15 minutes (4x as frequent).
• 0 0 * * * — every day at midnight (24x less frequent).

• Daylight saving — at the spring-forward hour (typically 02:00 → 03:00), the 02:00 firing is skipped. At the fall-back hour, the 02:00 firing happens twice. Pin to UTC if your job must fire exactly 24 times a day.
• Top-of-the-hour load spike — every hourly cron in your crontab fires at the same instant. Spread them out: some at 0, some at 15, some at 30, etc. Or use the Jenkins H modifier for deterministic hashing.
• Long-running hourly jobs — if the work takes more than 60 minutes, you need a lock or a concurrency-aware scheduler. Otherwise the next firing starts before the previous finishes.