Cron Job Monitoring: Catch Silent Failures Before They Compound

Your cron job failed last night. You don't know about it. It'll fail again tonight. And tomorrow night. By the time you notice, you'll have lost data, missed backups, or broken customer workflows.

Cron jobs are the invisible backbone of most applications — scheduled tasks, cleanup jobs, report generation, billing cycles. When they fail, they fail silently. No error page. No user complaint. Just a gap in your data that you discover weeks later.

Why Cron Jobs Fail Silently

Unlike web requests, cron jobs don't have a user waiting for a response. If a job crashes at 3AM, there's no ticket, no alert, no visibility. The job just... stops.

Common failure modes:

• Database connection timeout: Job tries to connect, times out, exits
• Disk full: Backup job can't write, fails silently
• Memory limit: PHP/Python job hits memory limit and crashes
• Missing dependency: Job expects a file that doesn't exist
• Network issues: Job can't reach external API, gives up

How to Monitor Cron Jobs

Effective cron monitoring has three components:

1. Heartbeat Checks

At the end of each job, ping a monitoring endpoint:

• curl https://your-monitor.com/ping/job-name
• If the monitor doesn't receive a ping within X minutes, alert
• Simple, works for any job type

2. Exit Code Tracking

Cron jobs should exit with code 0 on success, non-zero on failure:

• 0 = Success
• 1 = Warning (job completed but with issues)
• 2+ = Failure (job didn't complete)
• Monitor these exit codes and alert on non-zero

3. Duration Tracking

If a backup job normally takes 5 minutes but suddenly takes 30, something's wrong:

• Track job duration over time
• Alert on significant deviations (2x normal duration)
• Helps catch degraded performance before total failure

Common Cron Monitoring Mistakes

• Only checking if job ran: Knowing it ran doesn't mean it succeeded
• Alerting on every failure: One-off failures happen. Alert on consecutive failures
• No timeout alerts: If a job hangs, you need to know it didn't complete
• Ignoring duration: A job that takes 10x longer is a problem even if it succeeds

The OpsPulse Approach

OpsPulse monitors cron jobs the same way we monitor endpoints — with no-noise alerting:

• Consecutive failure requirement: Alert after 2-3 missed heartbeats, not 1
• Deduplication: One alert per incident, not per missed check
• Severity routing: Critical jobs = immediate alert, maintenance jobs = morning digest

The goal isn't to know about every cron job that runs. It's to know about the ones that matter — before they compound into bigger problems.

Start monitoring your cron jobs →