Skip to content
Feature · 2026

Backup monitoring: find out tonight, not during the restore

The worst time to learn your backup stopped running is the day you need it. Backup jobs fail quietly: a full disk, an expired credential, a renamed bucket, a commented-out crontab line. Nothing errors in front of you; the job just stops. UptimePad backup monitoring is a heartbeat on the job itself: the script pings a unique URL after each successful run, and if a scheduled backup goes missing, you are alerted within your grace window.

Last updated
Ways teams verify backups actually ran
ToolCatches a silently stopped jobCatches a crashing scriptWorks with any backup toolEffort to set upCost
UptimePad heartbeat on the job
One curl line appended to the script
✓✓✓One lineFree plan
Success emails from the backup tool
You must notice an email that did NOT arrive
--✓LowFree
Checking manually now and then
Works until the week nobody checks
--✓RecurringYour time
Storage-side object alerts
Cloud-specific, and misses partial/corrupt runs that never upload
✓--Per-cloud setupVaries

Why success emails do not work

Most backup tools can send a “backup completed” email. The failure mode is human: an email that arrives every day trains you to ignore it, and the alarm signal is an email that does NOT arrive, which nobody notices. Monitoring has to invert that: silence must be the thing that fires. That is exactly what a heartbeat monitor does; the daily success ping is invisible, and the missing ping pages you.

Wiring it up: one line per backup job

Append the ping to the job so it only fires when the backup actually succeeded:

# nightly pg_dump, ping only on success
0 3 * * * pg_dump mydb | gzip > /backups/mydb-$(date +\%F).sql.gz && curl -fsS -m 10 https://app.uptimepad.com/api/ping/YOUR-TOKEN > /dev/null

Set the expected schedule to daily with a grace period that covers your slowest run (an hour is typical for large dumps). For scripts with error handling, call the URL with /fail appended from the failure branch to alert immediately instead of waiting out the schedule. The same pattern covers pg_dump and mysqldump, rsync and restic and borg, S3 sync jobs, snapshot scripts, and managed tools that can run a post-backup hook.

Monitoring the run is step one, not the whole answer

A heartbeat proves the job ran to completion on schedule. It does not prove the archive is restorable; only a periodic test restore proves that. The honest playbook is both: heartbeat monitoring catches the silent stop the night it happens (the overwhelmingly common failure), and a scheduled restore drill (which can ping its own heartbeat) catches corruption. Two monitors, and the 3 a.m. surprise is gone.

Backups next to everything else you monitor

Backup heartbeats are ordinary UptimePad monitors, so they sit on the same timeline as your other cron jobs, uptime checks, SSL certificates, and DNS records, and they alert through the same channels (Email, Slack, Discord, Telegram, webhooks, and more on higher tiers). The free plan covers 10 monitors with commercial use allowed, which is enough to put a heartbeat on every backup you have tonight.

FAQ

What is backup monitoring?+

Backup monitoring verifies that scheduled backup jobs actually run. The reliable pattern is a heartbeat: the backup script pings a unique URL after each successful run, and the monitoring service alerts you when an expected ping goes missing. It catches silent failures that success emails and manual spot checks miss.

Which backup tools does this work with?+

Any tool that runs as a script or offers a post-backup hook: pg_dump, mysqldump, mongodump, rsync, restic, borg, duplicity, S3/rclone sync jobs, ZFS snapshot scripts, and commercial tools that can execute a command after a run. If it can run curl, it can be monitored.

Does a heartbeat prove my backup is restorable?+

No, and be wary of anything that claims otherwise. A heartbeat proves the job completed on schedule, which is where the large majority of backup failures happen. Restorability is proven by test restores; schedule one periodically and give it its own heartbeat monitor.

What grace period should a nightly backup have?+

Set it to comfortably cover your slowest normal run. For a nightly job, a grace period of 30 to 60 minutes is common: long enough that a slow dump does not page you, short enough that a genuinely missing backup is flagged the same night.

Is backup monitoring free on UptimePad?+

Yes. Heartbeat monitors are ordinary monitors on every plan, including Free (10 monitors, no credit card, commercial use allowed). Paid tiers add more monitors, faster schedules, and more alert channels.

Put a heartbeat on tonight’s backup

One curl line at the end of the script, and you will know every morning that it ran. Free plan, no card, commercial use allowed.

Start monitoring free →