Escalation
Escalation rules automatically route an alert to a higher-priority notification channel when it goes unacknowledged for too long. This prevents critical alerts from being missed if the initial notification is overlooked.
How escalation works
When an alert fires and is not acknowledged within a configured window, Beacon triggers an escalation:
- Alert fires → sends notification to the primary channel
- If not acknowledged within escalation delay → sends notification to the escalation channel
- Repeat (optional) at configurable intervals until the alert is acknowledged or resolved
Escalation only fires when the alert remains in Active state. Acknowledging the alert stops escalation immediately.
Configuring escalation
Escalation is configured per-policy at Dashboard → Policies → [policy] → Escalation.
| Setting | Description |
|---|---|
| Enabled | Toggle escalation on/off for this policy |
| Delay (minutes) | How long to wait after the initial notification before escalating |
| Escalation channel | Which notification channel to use for escalations |
| Repeat interval (minutes) | Re-notify on the escalation channel this often until acknowledged (optional) |
| Priority filter | Only escalate alerts at or above this priority (e.g. High and Critical only) |
Example configuration
A common setup for a managed services team:
- Primary channel: Slack
#alerts— all alerts land here - Escalation delay: 15 minutes — enough time for someone to see the Slack message during business hours
- Escalation channel: Email to on-call pager address — ensures someone is paged if Slack goes unnoticed
- Repeat interval: 30 minutes — re-page every 30 minutes until acknowledged
- Priority filter: Critical and High only — Moderate and Low alerts do not escalate
Escalation for off-hours coverage
For 24/7 coverage, consider a two-tier escalation with different channels for business hours vs. after hours. Beacon does not currently have a built-in business hours schedule for escalation routing — a workaround is to use a webhook channel and let your alerting platform (PagerDuty, OpsGenie) handle on-call scheduling.
See Webhooks & API for the webhook payload format.
See also: Alert Rules, Notification Channels