Heartbeat monitoring lets your servers, scripts, and background jobs check in on a schedule — so you know right away when something stops running.
No credit card required.
# last line of the backup / cron jobcurl -fsS https://api.pylonmon.com/hb/YOUR-MONITOR-ID
# Task Scheduler, last stepInvoke-WebRequest -UseBasicParsing \ https://api.pylonmon.com/hb/YOUR-MONITOR-ID
import urllib.request# after the work is doneurllib.request.urlopen( "https://api.pylonmon.com/hb/YOUR-MONITOR-ID")
Send a simple HTTP request from your servers, scripts, or devices when the job finishes. No agent, no inbound firewall holes — miss the window and you are paged.
View integration docs →Free for 20 monitors. Escalation ladders on every plan, so a missed backup reaches a human.
Send one HTTP request from your servers, scripts, or devices when the job finishes. No SDK, no agent — the last line of the job is the integration:
# add one line at the end of your backup / cron job: curl -fsS https://api.pylonmon.com/hb/your-monitor-id # miss the window, and PylonMon pages you: nightly-backup missed check-in — silent for 14m · SLA 10m
Append a one-line curl to your cron job, backup, or script. It hits a unique URL every time it finishes successfully.
You set the expected window. PylonMon watches from outside your network, so it still notices even if the whole box dies.
Miss the window and you're alerted — email, Slack, Discord, SMS, voice, webhook and more, with escalation ladders. The silence is the signal.
An HTTP check tells you a URL is up. It can't tell you the 3am backup silently stopped writing, or a cron job has been failing for a week. A dead-man's-switch flips that around: instead of asking "is it up?", it expects a regular "I'm alive" — and the moment that signal doesn't arrive, you hear about it. It's the only way to catch jobs that fail by simply not running at all.
restic/borg/rsync jobs, offsite syncs.And it's not only heartbeats — the same account also runs website & server uptime, HTTP/S, TCP, ping, DNS, TLS-certificate and domain-expiry monitors, so your uptime checks and your cron checks live in one place. See every monitor type →
No. A heartbeat is a single curl line. HTTP, TCP and ping checks run from our side. There's no agent on your servers.
Free forever for 20 monitors with 1-minute intervals and heartbeat, HTTP/S, TCP & ping checks. Paid tiers start at $9/mo for 30-second intervals, more monitors, SMS, and extra monitor types. See pricing →
As soon as a job misses its window. You set the SLA per monitor, and alerts fire on the first state change — no repeat 2am nagging.
Set up in about two minutes. No credit card.