Cron & Heartbeat Monitoring

Cron and heartbeat monitoring is inside-out: instead of UptimeStatus reaching out to your service, your job reaches out to UptimeStatus. Each successful run pings a unique URL. If the expected ping doesn't arrive on schedule, we alert you — so a backup that silently stopped running gets caught, not just one that errored loudly.

How it works

Every heartbeat monitor has a unique ping URL and an expected cadence plus a grace period. Your job requests that URL when it finishes successfully. UptimeStatus records the ping and starts a countdown for the next one:

• Received on time → the monitor stays healthy.
• Overdue past the grace period → the monitor goes down and alerts fire.

Cadence can be as tight as every 30 seconds or as loose as monthly. The grace period absorbs normal jitter (a nightly job that usually finishes at 02:03 but sometimes 02:09).

Have the job ping on success — a plain GET or POST is enough:

# Run at the end of your cron job, only on success
0 2 * * *  /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 \
  https://uptime-status.org/ping/9f3c1a2e-heartbeat-token

You can also signal start and failure to measure run duration and catch non-zero exits:

curl -fsS https://uptime-status.org/ping/<token>/start   # job began
curl -fsS https://uptime-status.org/ping/<token>/fail    # job errored

What triggers an alert

• No ping arrives within the cadence + grace period (the job is overdue, hung, or the box is down).
• An explicit /fail ping is received.
• A /start with no matching success within the grace window (run took too long).

Setting it up

1. Add monitor and choose Cron / Heartbeat.
2. Set the expected cadence and a grace period. Cadence can be a plain interval (every 30 seconds up to monthly) or a 5-field cron expression such as 0 2 * * * (daily at 02:00 UTC) or a nickname like @hourly. When a cron expression is set, the next expected check-in is the next scheduled slot; the interval is used otherwise.
3. Copy the generated ping URL.
4. Add the curl ping to the end of your job, as shown above.
5. Attach notifications.

Related

• Health Checks · Uptime · Server Metrics
• Notifications
• Marketing: Cron & heartbeat monitoring feature