Re-notify for persistently active episodes in the experimental alerting system

The On status change frequency option notifies once for each status transition, for example when an episode activates or resolves. This is efficient for reducing noise from rules in the experimental alerting system, but a persistently active episode that only changes in severity doesn't re-trigger a notification.

To re-notify for episodes that stay active without a status change, use a time-based throttle.

• At most once every… Re-notifies after the configured interval regardless of whether severity or status changed. Setting this to 1h sends a follow-up notification every hour while the episode remains active and matched.
• On status change + repeat at interval Notifies on status change and then repeats at the configured interval while the episode stays in the same status.

In this example, you want to be re-paged if a critical episode stays open for more than an hour. Set the action policy frequency to At most once every 1h. The action policy fires when the episode first matches and then again each hour until the episode resolves or no longer matches.

• Manage severity escalation notifications for scenarios where a severity change determines whether a notification fires.
• Action policy reference for descriptions of all frequency options.
• Create and configure an action policy to set up the frequency settings described on this page.