Monitor lifecycle and alert events
Use this reference to distinguish a failed check from an incident and an alert notification. For the setup workflow, see Add an alert to a monitor.
Last checked: 2026-10-09 · Maintainer: UptimeObserver Support
Monitor lifecycle
For HTTP, DNS, Port, and Ping monitors:
Check → Retry (if configured) → Down / active incident → recovery / Up
| Term or state | Meaning |
|---|---|
| Check result | One observed response or network attempt from a monitoring region; evaluated using that monitor's configuration |
| Up | The monitor meets its success criteria; this does not prove every application function works |
| Retry | Failed results have been recorded, but the configured failure threshold has not yet generated Down |
| Down | The failure threshold was met; normally opens an automatic incident |
| Incident | An outage record with a start time, region details, root cause, and later resolution time |
| Alert rule | A monitor-specific subscription to one event through one notification connection/target |
| Recovery / Monitor Up | The automatic incident resolves after the required region recovery; configured Up rules are notified |
| Paused | Checks are suspended; pause is separate from the last recorded health result and is not proof of recovery |
An incident and its monitor state are saved before notification delivery. A provider failure does not undo the incident, so Down does not prove an alert arrived. Inspect the incident and test your connection separately.
Retries
For actively checked monitors, Retries is a consecutive-failure threshold—not a burst of immediate requests. Checks continue at the configured interval.
| Retries setting | Consecutive failed results needed for Down |
|---|---|
0 |
1 |
1 or 2 |
2 |
3 |
3 |
4 |
4 |
5 |
5 |
With a nonzero setting, the first failed result puts the monitor in Retry. A successful result resets the counter. Results from selected regions share this monitor-wide counter; the table counts results, not complete rounds across all regions. Every region does not need to fail before an incident opens.
Retries are different from alert retrigger, which repeats notifications after an incident is already active.
Recovery
With one region, a successful check after Down normally resolves the incident and sends Monitor Up. With multiple regions, the incident remains active until every currently configured region recorded as Down in that incident has a recovery result. Regions without a Down record in the incident do not block resolution. One region recovering is therefore not necessarily a full recovery.
A successful check while in Retry returns the monitor to Up without a recovery notification, because no Down incident was opened. Further failures during an existing incident update its region state rather than sending a new initial Down notification for every check.
Heartbeat differences
Heartbeats receive pings; they do not check your job directly. A new heartbeat remains Pending until its first ping is processed. After that, missing pings are evaluated against the interval/cron schedule, grace period, and heartbeat-specific overdue retry rule. Recovery follows a processed ping and a non-overdue scan. See overdue checks and retries.
success: false and nonzero exitCode are metadata and still refresh the ping deadline. To detect a failed run, do not send its heartbeat.
Pause, maintenance, and snooze
These controls are not interchangeable:
| Control | Checks | Incident/notification effect |
|---|---|---|
| Pause / Resume | Suspends / restarts checks | Emits separate Monitoring Paused/Resumed events; does not establish endpoint health |
| Maintenance window | Checks continue | Covered Down results do not open a new automatic incident; a recovery can resolve an existing incident without a recovery notification |
| Alert Snooze | Checks continue | Suppresses retrigger notifications for one alert on one incident until expiry; does not suppress all event notifications |
| Deactivate alert | Checks continue | Stops notifications for that rule until reactivated |
Maintenance windows require a paid plan. See monitoring controls and the dashboard's maintenance settings. Do not pause a monitor as an outage simulation: use a dedicated endpoint test.
Alert Event Types
When adding an alert to a monitor, you can choose from the following event types:
| Event | Description | When It Triggers |
|---|---|---|
| Monitor Down | A monitored service is unreachable or failing | When the configured failure threshold is reached, including overdue heartbeat checks |
| Monitor Up | An automatic incident has recovered | When the required regions recover; heartbeats recover after a processed ping and a non-overdue scan |
| Monitoring Paused | Monitoring has been manually paused | When a user pauses a monitor |
| Monitoring Resumed | A paused monitor has been restarted | When a user resumes monitoring |
| SSL Expiry 14 Days | SSL certificate expires in 14 days | 14 days before certificate expiration |
| SSL Expiry 7 Days | SSL certificate expires in 7 days | 7 days before certificate expiration |
| SSL Expiry 1 Day | SSL certificate expires in 1 day | 1 day before certificate expiration |
Recommended Setup
We recommend configuring alerts for both Monitor Down and Monitor Up events. This ensures your team is notified when an issue starts and, just as importantly, when it has been resolved.
Monitor Down
The most common alert type. For HTTP, DNS, Port, and Ping monitors, it triggers when failed results meet the Retries threshold, unless covered by maintenance. For Heartbeat monitors, it follows the overdue-check retry setting after the interval and grace period pass. Use this to be notified of downtime, server errors, connectivity issues, or jobs that did not report success.
Monitor Up
Triggers when a Down incident resolves, subject to region recovery and maintenance. A Retry-to-Up transition does not generate this notification.
Monitoring Paused & Resumed
These events track when monitoring is manually paused or resumed. Use them for: - Tracking maintenance windows - Notifying teams when monitoring coverage changes - Recording when services are intentionally taken offline
SSL Expiry Events
For HTTPS monitors on paid plans, UptimeObserver tracks SSL/TLS certificate expiration. Three staggered alerts give you time to renew:
| Days Remaining | Alert | Action |
|---|---|---|
| 14 days | Warning | Start renewal process |
| 7 days | Urgent | Renewal should be in progress |
| 1 day | Critical | Immediate action required |
Alert Retrigger
Paid-plan feature
Alert retrigger is available on paid plans for Monitor Down rules only. Free users will see the message: "Alert retrigger is a Premium feature."
The alert retrigger feature sends repeated notifications while an incident is still active, keeping your team aware until the issue is resolved.
Configuration
- On a monitor's alert settings, enable the toggle: "Re-send alert while incident is active"
- Set the retrigger interval — how often to resend the alert
| Setting | Range | Default |
|---|---|---|
| Retrigger Interval | 10 minutes to 24 hours | — |
Pushover Emergency alerts
Pushover alerts using Emergency priority cannot use alert retrigger. Emergency notifications have their own retry schedule and repeat until acknowledged or expired. See the Pushover integration guide for configuration requirements.
Error: This feature requires a premium plan. Please upgrade your plan.
Free plan users who attempt to enable alert retrigger will receive this 403 error. See plans and limits.
Use Cases
- Extended outages — Receive hourly reminders that an incident is still active
- On-call escalation — Combine with incident management integrations for periodic re-notification
- After-hours incidents — Ensure teams don't miss alerts during off-hours
Alert Snooze
In an incident's Alert Snooze section, snooze one retrigger-enabled alert to temporarily silence its repeated reminders for that incident. This does not pause checks, resolve the incident, or silence a separate Monitor Up rule.
Snooze States
| State | Description |
|---|---|
| Alert Snoozed | That alert's retrigger notifications for this incident are temporarily suppressed |
| Snooze Removed | Snooze was manually canceled before expiry |
| Snooze Expired | Snooze duration ended; eligible retrigger notifications can resume |
When to Snooze
- Ongoing incident — Stop repeated reminders while responders investigate
- Known issues — Suppress notifications for issues you're already aware of and working on
- A specific reminder channel — Snooze one alert without muting the others
Note
Snooze applies to incident retrigger notifications, not all alerts on a monitor. Use an alert's active switch to disable a rule, or a maintenance window for planned work.
Need Help?
If you need assistance configuring alerts, reach out using the "Need Help?" button on the bottom right corner or email us at support@uptimeobserver.com.