Incident Response
When a monitor detects an outage, an “incident” is created. This guide explains the incident flow and how to respond.
Incident Lifecycle
Section titled “Incident Lifecycle”- Detection: after 5 consecutive failed checks, an incident is created automatically and your notification channels are notified (a single failure never notifies)
- Acknowledge (optional): pressing “Acknowledge” sets the status to acknowledged, signaling to your team that someone is on it
- Resolution: when the monitor recovers, the incident is resolved automatically and a recovery notification is sent. You can also resolve manually
If the monitor remains down after a manual resolution, the consecutive-down count restarts from zero. A new incident is created and notifications are sent after 5 more consecutive failures.
See the monitoring lifecycle for the exact detection rules.
Viewing Incidents
Section titled “Viewing Incidents”The “Incidents” tab on the service detail screen (click a service card) lists that service’s incidents.
| Status | Meaning |
|---|---|
| Ongoing (red) | Outage in progress |
| Acknowledged | Someone is handling it |
| Resolved | Recovered |
Manual Incidents
Section titled “Manual Incidents”You can record incidents independently of automatic detection — useful for user-reported outages or components outside monitoring that should appear on your status page.
- On the service detail’s “Incidents” tab, press “Create incident”
- Enter a title and register (it gets a “Manual” badge)
- Press “Resolve” once handled
Manual incidents stay open until resolved manually, and only manual incidents can be deleted.
Warning Incidents (degraded)
Section titled “Warning Incidents (degraded)”When a check cannot be evaluated — as opposed to failing — a warning incident (severity warning) is created, distinct from a normal outage. The current HTTP / TCP / Ping monitors only return up / down, so this state does not occur.
When Notifications Don’t Arrive
Section titled “When Notifications Don’t Arrive”- If a monitor has channels attached directly, team-wide channels are not notified (see notification behavior)
- Rapid down/up cycling triggers flap suppression, pausing notifications (incidents are still recorded)
- During maintenance windows, incident detection itself is suspended
- Check the integrations settings for unverified (email) or disabled channels