Uptime Monitoring

A monitor combines a target, a schedule, check locations, and rules for deciding whether the result is healthy. Scheduled checks run from the locations selected in the dashboard. Heartbeat monitors instead wait for signals from your job.

Read the result

A successful check meets the protocol requirements and any configured assertions. A failed connection, timeout, or failed assertion produces a failure. A degraded threshold lets you distinguish a slow response from a healthy one.

Use the monitor detail view to inspect recent results and latency. A successful TCP connection or ping proves reachability at that layer; it does not prove your application works. Use an HTTP or gRPC health endpoint when application health matters.

Choose the schedule

The dashboard shows the intervals and locations available to your workspace. Shorter intervals detect changes sooner and generate more checks. Configure timeouts and attempts for the service you actually run: the execution budget is bounded, and attempts are not an unlimited retry loop.

Connect notifications to receive status changes. Publish customer communication separately through status reports.