Monitor Flapping Metric incidents

A Flapping Metric incident helps you track whether a particular server resource or monitor point associated with a server instance is generating incidents frequently within a specified time window. You can set the minimum number of incidents required to trigger an alert. In addition to individual metrics, Flapping Metric incidents also work for instance-level heartbeat alerts such as SNMP or Fabric.

For example, a server has two resources: CPU and Memory. You might have set thresholds such as CPU usage above 80% for 5 minutes and Memory usage above 90% for 5 minutes. You can configure Flapping Metric incidents to monitor whether any of these resources are causing too many incidents. If you set a 2-hour time period with a minimum of 3 incidents, then during that time, 6 CPU incidents and 2 Memory incidents occur, then only the CPU incidents would meet the threshold. A Flapping Metric incident for the CPU will then be triggered for your server. Flapping Metric incidents will remain open indefinitely until manually closed.

Note: Flapping Metric incidents being enabled do not suppress alerts for other metrics or alerting on the instance.

Flapping Metric incidents, when they occur, are displayed in the Instance Details page or the page of the incident that caused the flapping condition.

Note: If multiple metrics breach the configured Flapping Metric threshold, they will be displayed under one incident. Expanding the description will list the metrics that were above the threshold.

Configuration

To configure a Flapping Metric incident, perform the following steps:

1. Go to the Instance Details page of the server that you want to configure.

2. Under Flapping Metric Threshold, select Configure.
   

The Flapping Incident Threshold modal is displayed.

Each configurable option is described in the following table.

Field	Description
Generate a flapping incident when number of incidents exceed	Specifies the minimum number of incidents for any server resource or monitoring point within a server instance that must occur to trigger a Flapping Metric incident.
in time period	Defines the duration (from 30 minutes to 24 hours) during which incidents for any server resource or monitoring point will be considered.
creating an incident with severity	Indicates the severity level for the alert, either as a warning or critical.
and notify with the Alert Timeline	Determines the schedule for notifying users about the Flapping Metric incident.

You can also use templates to configure Flapping Metric incident thresholds.