How to Monitor Your Cron Jobs

Quick answer
A heartbeat monitor pings a URL every time a job runs; if the ping stops, you get an alert.
Add one line to your cron job — curl the check URL — and failures become visible.
Free tiers of healthchecks.io or cron-job.org cover a small project completely.

What is the simplest way to monitor a cron job?

Sign up for healthchecks.io (free for 20 checks), create a check, and add its unique URL to the end of your cron command. Every run pings the URL; the service alerts you if no ping arrives within the schedule you set. Your job gains a visible heartbeat with one line of change, and you can require a success signal by making the ping conditional on the job exiting cleanly.

# Alert if the backup job misses its 2:30 a.m. run by more than 15 minutes
30 2 * * * /home/me/backup.sh && curl -fsS -m 10 --retry 5 -o /dev/null https://hc-ping.com/REPLACE_WITH_UUID

# Ping only on success; failures trigger the alert because no ping is sent
30 2 * * * /home/me/backup.sh && curl -fsS -m 10 --retry 5 -o /dev/null https://hc-ping.com/REPLACE_WITH_UUID

How do I choose alert settings?

Set the check period to your schedule (for a daily job, 24 hours) and the grace to roughly the longest the job could legitimately run plus a buffer. Choose where alerts go: email for free, or Slack/Telegram/phone push if the free plan offers it. The point is that a missed run should reach you as a notification, not a surprise discovered days later.

What else should I monitor about a job?

A heartbeat proves the job started, not that it succeeded. Two additions close the gap: ping the success URL only when the job exits zero, and have the job write a status line to a log you can inspect. For high-stakes jobs like backups, also verify the output — a backup that pings successfully while writing an empty file is still a failure.

Where this bites vibecoders
The classic AI-era incident is a ‘working’ cron job that has been failing for three weeks. Because the assistant set it up and the dashboard shows nothing, nobody knows. Heartbeat monitoring is a ten-minute habit that converts silent failures into phone notifications, and it’s exactly the kind of operational detail an AI assistant won’t volunteer — you have to ask for it or know to add it.

Where AI coding assistants get this wrong

Suggesting bare cron with no monitoring because the assistant has no concept of operational alerting.
Putting the heartbeat ping after the command with &&, so failures also ping and the monitor is useless.
Choosing a grace period shorter than the job’s real runtime, causing false alerts.
Monitoring the job but not the data it produces, so corrupt output goes unnoticed.

Checklist

Add a heartbeat check to every scheduled job that matters.
Ping the success URL only on clean exit (use &&).
Set period and grace to match the real schedule, with buffer for slow runs.
Configure alerts to reach you on a channel you actually check.

FAQ

What’s the difference between a heartbeat check and an uptime monitor?

An uptime monitor pings an endpoint from outside to check availability. A heartbeat check is the reverse: your job pings the monitor. Heartbeats are for scheduled jobs that run briefly; uptime monitors are for services that should be reachable 24/7.

Can I monitor cron jobs without a third-party service?

Yes: log output to a file and set up a separate alert when the log stops updating, or run a wrapper script that emails you on failure. A service like healthchecks.io is simpler and harder to get wrong, which is why it’s the recommended path.