Grouped Alerts Feature

Overview

Group related incidents together to reduce noise and manage alerts more effectively. When multiple monitors in the same project or escalation policy experience issues within a configured time window, alerts will be grouped to prevent notification fatigue.

Configuration

You can enable Grouped Alerts at two levels:

Project Level

1. Go to Project Settings
2. Find "Alert grouping" section
3. Select a time window from the dropdown:
   • Never (disabled)
   • Within 10 seconds
   • Within 30 seconds
   • Within 1 minute
   • Within 3 minutes
   • Within 5 minutes
   • Within 10 minutes
   • Within 15 minutes
   • Within 30 minutes
   • Within 1 hour

The project-level setting can be overridden by individual escalation policies.

Escalation Policy Level

1. Go to Escalation Policy settings
2. Find "Alert grouping" section
3. Select a time window from the dropdown (same options as project level)

Note: Escalation Policy settings override Project settings when both are configured.

Rules & Examples

Down Alerts

1. When the first monitor goes down, an alert is sent
2. Any other monitors that go down within the time window won't trigger alerts
3. The first monitor's UUID is stored to track which monitor initiated the alert group

Example:

Monitor A goes down → Alert sent
Monitor B goes down (within window) → No alert
Monitor C goes down (within window) → No alert

Up Alerts

Within the time window:

• Only the first alerted monitor can send an up alert
• Other monitors going up won't trigger alerts

Outside the time window:

• First monitor to go up gets the alert
• Other monitors going up within the new window won't trigger alerts

Example:

Monitor A goes down → Alert sent
Monitor B goes down (within window) → No alert
Monitor A goes up (within window) → Up alert sent
Monitor B goes up (within window) → No up alert
Monitor C goes up (after window) → Up alert sent

Time Window Behavior

• Each alert type (up/down) has its own independent window
• Windows are tracked separately for each project/escalation policy

Example with 60-second window:

10:00:00 - Monitor A goes down → Alert sent
10:00:30 - Monitor B goes down → No alert
10:01:00 - Monitor A goes up → Up alert sent
10:01:30 - Monitor B goes up → No up alert
10:02:00 - Monitor C goes down → New down alert sent (window expired)

Benefits

• Reduces alert noise during widespread issues
• Maintains visibility of the first affected monitor
• Prevents alert fatigue for teams
• Configurable at both project and escalation policy levels
• Independent windows for up and down alerts

Best Practices

• Set window duration based on your team's response time
• Consider enabling at escalation policy level for more granular control
• Monitor alert patterns to optimize window duration
• Use in conjunction with other alerting features like alert delays and repeat intervals