Browse documentation
Monitoring
How checks and incidents work
Understand site state, failure confirmation, recovery, and recorded evidence.
Check results
A check succeeds when the target responds before the configured timeout with the expected HTTP status. Downbot records the status code, response time, timestamp, and failure details when available.
| State | Meaning |
|---|---|
| Healthy | The latest recorded check met the configured expectation |
| Down | A failure was confirmed and an incident is active |
| Paused | Scheduled checks are disabled for the site |
| Waiting | No monitoring evidence has arrived yet |
Failure confirmation and recovery
Downbot requires consecutive failed checks before opening a downtime incident, which reduces alerts from brief network noise. It performs rapid confirmation checks during a suspected failure. A later successful check resolves the active incident and can send a recovery notification.
Intervals and confirmation
The normal interval controls routine checks. Confirmation checks during a suspected outage may happen sooner than the normal schedule.
Uptime and history
The site detail page calculates uptime for the last 24 hours, 7 days, and 30 days from recorded checks. It also shows a 24-hour response-time trend, the 20 most recent checks, and up to 10 recent incidents.