Back to features
Heartbeat monitoring
Let your systems say they are still alive.
Use heartbeat check-ins for servers, devices, workers, and background processes where the only reliable signal comes from inside your system.
- Feature
- Monitor processes that cannot be checked from outside
- Feature
- Catch stopped workers and disconnected devices
- Feature
- Use any language that can send an HTTP request
- Feature
- Route missed heartbeats through normal alert channels
- Useful for internal systems.
- Some things are not public URLs. Heartbeats let internal workers, scripts, private services, and devices report that they are still running.
- Workers
- IoT devices
- Private servers
- One workflow for every failure.
- A missed heartbeat can alert Slack, SMS, PagerDuty, email, or any webhook just like a failed website check.
- Shared alerting
- Shared incident workflow
- Shared dashboard
- Good for systems you cannot poll.
- If a service lives behind a firewall, runs on a customer device, or only wakes up periodically, let it call OnlineOrNot instead of exposing it to the internet.
- Private networks
- Customer devices
- Periodic workers
- Simple enough to add everywhere.
- A heartbeat is just an HTTP request. Add it to scripts, workers, services, and devices without installing an agent or adopting a new runtime.
- curl
- fetch
- Any HTTP client
Frequently asked questions
The short version. The docs have the details when you need them.
- How does a heartbeat differ from uptime monitoring?
- Uptime checks call your service. Heartbeats wait for your service to call OnlineOrNot.
- Can I use heartbeats for cron jobs?
- Yes. Cron job monitoring is one of the most common heartbeat use cases.
- What counts as a successful heartbeat?
- Your system sends the heartbeat request when it reaches the point you consider healthy, usually after a job completes or a process confirms it is still running.
- Do heartbeats need an SDK?
- No. Any language or runtime that can make an HTTP request can send a heartbeat.
Start with heartbeat monitoring.
Start with the process you cannot monitor from the outside, then add heartbeats wherever silence would be dangerous.