Cron job monitoring: we tell you when a job does not run

One line, right where your job ends

The address below is an example. Yours appears in your account once you create the heartbeat, with the same lines ready to copy. Every line has a short time limit and never holds up your job, even if we cannot be reached.

0 3 * * *  /path/to/backup.sh; curl -fsS -m 10 --retry 3 https://runcheck.de/a8k3m2x9q1/$? > /dev/null

// … your code …
@file_get_contents('https://runcheck.de/a8k3m2x9q1', false, stream_context_create(['http' => ['timeout' => 5]]));

# … your code …
import urllib.request
try:
    urllib.request.urlopen("https://runcheck.de/a8k3m2x9q1", timeout=10)
except OSError:
    pass

// … your code …
await fetch('https://runcheck.de/a8k3m2x9q1', { signal: AbortSignal.timeout(10000) }).catch(() => {});

Method:   GET
URL:      https://runcheck.de/a8k3m2x9q1
Options:  Timeout 10000

Method:  GET
URL:     https://runcheck.de/a8k3m2x9q1

select cron.schedule('nightly-cleanup', '0 3 * * *', $$
  delete from sessions where expires_at < now();
  select net.http_get('https://runcheck.de/a8k3m2x9q1');
$$);

// app/api/cron/route.ts
export async function GET() {
  // … your code …
  await fetch('https://runcheck.de/a8k3m2x9q1', { signal: AbortSignal.timeout(10000) }).catch(() => {});
  return Response.json({ ok: true });
}

In detail, step by step: BackupWordPressn8nSupabaseVercelNetlify

Frequently asked questions

What does monitoring cost?

One heartbeat is free; you see its status in your account. With Plus for €17.88 a year you monitor up to 10 heartbeats and get an email when a check-in is missing. Unlimited for €47.88 a year brings 50 heartbeats and also alerts in Slack, Microsoft Teams, Discord or via webhook. Every new sign-up starts with 14 days of Unlimited, no payment details needed.

When exactly do you alert me?

You set how often your job runs and how much leeway it gets – for a daily backup, one hour for example. If no check-in has arrived by then, we alert you once; the next check-in brings the all-clear. If your job reports a failure, for example through an exit code other than 0, we alert you right away.

What happens if runcheck.de cannot be reached?

Your job keeps running. Every line on this page has a short time limit and ignores errors while checking in. And if no heartbeat at all checks in during a period, the problem is on our side – then we do not alert you for that period.

What happens to my data?

Check-ins arrive on our servers in Germany. We store the time of each check-in, the IP address it came from and – only if your job sends it via POST – up to 10 KB of output. It is kept like the call log of your cronjobs. Do not put personal data into the output; it is meant for short status messages.

Is this a dead man’s switch?

Yes. A heartbeat works like a dead man’s switch: as long as your job checks in, nothing happens. If the check-in does not arrive, we trigger the alert. Other services call this cron job monitoring or heartbeat monitoring.

Do I need to install anything?

No. Checking in is an ordinary call to an address – with curl, with what PHP, Python or Node.js bring along, or as an HTTP step in n8n and Make. Laravel can do it out of the box.

Can I report run times and failures too?

Yes. https://runcheck.de/a8k3m2x9q1/start at the start measures the run time and reports a run that gets stuck. https://runcheck.de/a8k3m2x9q1/fail or an exit code from 1 to 255 reports a failure right away. Via POST we accept up to 10 KB of output, which shows up in the log.