Alerts v2 can automatically clear informational alerts when a recovery condition is met or after a configured duration. This helps prevent informational events from remaining open and reduces the need to dismiss them manually.
How automatic clearing works
An Alerts v2 alert can clear automatically when:
- The monitored condition returns to normal.
- A custom clear condition is met.
- An Auto Clear period expires, when available.
If an alert does not have a clear condition or Auto Clear setting, it may need to be dismissed manually.
Migrating an alert from legacy alerting
If you are replacing a legacy informational alert:
- Review the existing alert’s trigger condition, scope, severity, and notification settings.
- Disable the legacy alert to prevent duplicate notifications during testing.
- Recreate the alert in Alerts v2 with the required scope and trigger condition.
- Configure a clear condition or Auto Clear setting.
- Test the Alerts v2 alert using a safe, non-production condition.
- Disable or remove the legacy alert after the Alerts v2 version is confirmed.
Create an automatically clearing alert
- Go to Alerts > Manage alerts.
- Create a new alert or open an existing Alerts v2 alert.
- Define the alert scope, such as devices, interfaces, services, groups, or tags.
- Configure the trigger condition.
- Set the severity to Informational, if appropriate.
- Configure the clear behavior:
- Use the default inverse of the trigger condition.
- Create a custom clear condition that defines the recovery state.
- Use Auto Clear, when available, to resolve the alert after a specified period.
- Add the appropriate notification channels.
- Add a clear message if one is available for the alert type.
- Save the alert.
Examples of clear conditions include:
- CPU usage remains below 70% for a defined period.
- A service returns to a running state.
- A monitored value returns below its trigger threshold.
- No matching event is detected during the configured clear period.
Test the alert
Use a non-production device or a safe test condition whenever possible.
- Trigger the alert condition.
- Confirm that the alert and notification are generated.
- Restore the monitored condition.
- Confirm that the clear condition is met.
- Verify that the alert changes to a resolved or cleared state.
- Confirm that notifications are not duplicated by a remaining legacy alert.
Troubleshooting
The alert does not clear
- Confirm that the clear condition uses the correct metric and scope.
- Verify that the recovery value meets the configured threshold.
- Check that the device or service is reporting current data.
- Review any alert delay or evaluation window.
- Confirm that the alert is not affected by a maintenance window, override, or suppression.
Duplicate notifications are generated
- Confirm that the legacy version of the alert is disabled.
- Check whether both alert definitions use the same scope and notification channel.
- Review parent and child site alert configurations for overlapping definitions.
The alert clears and triggers repeatedly
- Increase the trigger or clear delay.
- Use a wider difference between trigger and clear thresholds.
- Confirm that the monitored metric is not fluctuating around the threshold.
Alerts v2 is unavailable
- Confirm that Alerts v2 is enabled for the organization or site.
- Verify that your user role has permission to manage Alerts v2 definitions.
- Contact your Auvik administrator or Auvik Support if Alerts v2 is not available.