Incidents
Incident lifecycle & statuses
Use the incident timeline to distinguish a confirmed monitor failure from an individual failed check, then follow the event through acknowledgement and recovery.
Understand the incident states
- Open
- The configured number of probe regions has confirmed that the monitor is failing. The incident needs a responder.
- Acknowledged
- A responder has taken ownership. The acknowledgement records the user and time, but it does not mean that the monitor has recovered.
- Resolved
- Probe results confirm recovery. The incident keeps its timeline and records the time that the service was restored.
How an incident opens
- Set the monitor's confirmation threshold when you configure its probe regions. The default is
2confirming regions. - Wait for probes to report results. GetPulseCheck correlates regional results in a 90-second window so nearby failures can be evaluated together.
- When the threshold is reached, GetPulseCheck creates one open incident for the monitor. A failure below the threshold is not promoted to a confirmed incident.
The incident record is protected against duplicate open incidents for the same monitor. A later failure after recovery creates a new incident with its own start time.
Check an incident
- Open Incidents from the dashboard.
- Use the status filters to show Open, Acknowledged, or Resolved incidents. Search by monitor name or confirming region when needed.
- Select an incident to open its detail page. The timeline shows when it started, whether someone acknowledged it, and when the probe confirmed recovery.
How an incident resolves
Resolution is automatic. When the correlated probe results return to an up state, GetPulseCheck sets the incident to resolved and recordsresolvedAt. The detail page calculates the total duration from the incident start and recovery timestamps. An acknowledgement remains visible in the timeline after recovery.