Heartbeat (cron) monitors

Some things can't be checked from the outside: a nightly backup, a queue worker, a script on a server behind a firewall. A heartbeat monitor turns it around — your job calls a unique URL every time it runs, and we alert you when a call doesn't arrive in time.

How it works#

1. Create a monitor of type Heartbeat and set the interval to how often the job runs (for example 1 hour for an hourly job, 24 hours for a nightly one).
2. Set the grace period (graceSeconds, 0–86,400 seconds, default 60): how late a call may be before it counts as missed. Give jobs with a variable run time more grace.
3. Save. The monitor shows its heartbeat URL:

https://api.isituprightnow.com/api/v1/heartbeat/<token>

4. Call this URL at the end of every successful run, with GET, POST or HEAD. It answers 200 ok.

If no call arrives within interval + grace after the previous one, the monitor is confirmed down, an incident opens and your alert rules apply. The next call marks it up again and resolves the incident.

Responses#

Calls are cheap and the endpoint accepts any origin, so you can also call it from serverless functions or CI pipelines.

Examples#

Replace <token> with your monitor's token, or copy the full URL from the monitor page.

crontab + curl#

Ping only when the job succeeded (&&), with retries and a timeout so a network blip doesn't fail the job:

# m h dom mon dow  command
30 2 * * * /usr/local/bin/backup.sh && curl -fsS -m 10 --retry 3 -o /dev/null https://api.isituprightnow.com/api/v1/heartbeat/<token>

wget#

/usr/local/bin/backup.sh && wget -q -T 10 -t 3 -O /dev/null https://api.isituprightnow.com/api/v1/heartbeat/<token>

Inside a shell script#

#!/usr/bin/env bash
set -euo pipefail

pg_dump --format=custom mydb > /backups/mydb.dump

# Reached only if every command above succeeded.
curl -fsS -m 10 --retry 3 -o /dev/null "https://api.isituprightnow.com/api/v1/heartbeat/<token>"

PowerShell (Windows Task Scheduler)#

& "C:\jobs\export.ps1"
if ($LASTEXITCODE -eq 0) {
    Invoke-RestMethod -Uri "https://api.isituprightnow.com/api/v1/heartbeat/<token>" -Method Post -TimeoutSec 10
}

systemd timer#

Ping from a separate unit that runs only when the job unit succeeded:

# /etc/systemd/system/backup.service
[Unit]
Description=Nightly backup
OnSuccess=backup-heartbeat.service

[Service]
Type=oneshot
ExecStart=/usr/local/bin/backup.sh

# /etc/systemd/system/backup-heartbeat.service
[Unit]
Description=Report the backup to isituprightnow

[Service]
Type=oneshot
ExecStart=/usr/bin/curl -fsS -m 10 --retry 3 -o /dev/null https://api.isituprightnow.com/api/v1/heartbeat/<token>

# /etc/systemd/system/backup.timer
[Unit]
Description=Run the backup every night

[Timer]
OnCalendar=*-*-* 02:30:00
Persistent=true

[Install]
WantedBy=timers.target

Enable it with systemctl enable --now backup.timer. (OnSuccess= needs systemd 249 or later; on older systems add ExecStartPost= with the curl command to backup.service.)

Kubernetes CronJob#

apiVersion: batch/v1
kind: CronJob
metadata:
  name: nightly-report
spec:
  schedule: "0 3 * * *"
  jobTemplate:
    spec:
      template:
        spec:
          restartPolicy: Never
          containers:
            - name: report
              image: registry.example.com/report:latest
              command: ["/bin/sh", "-c"]
              args:
                - ./run-report && wget -q -T 10 -O /dev/null https://api.isituprightnow.com/api/v1/heartbeat/<token>

Tips#

• Ping at the end, not the start. A ping at the start only proves the job began.
• Long jobs: set the interval to the schedule and the grace period to the longest normal run time.
• Irregular jobs (for example a queue worker): ping from the worker's main loop at most once a minute and use an interval of a few minutes.
• Pausing: pause the monitor before you disable the job, so the missing pings don't open an incident.