Alarm service

To detect the change in the status of a service, you can create an alarm policy to trigger an alarm, which will promptly notify you of the change.

You can create an alarm policy with the following procedure:

Create an alarm policy

Enter the policy name, select the product type, such as KCI - cluster, and then add alarm rules.

• A complete alarm rule contains the following items: Metric Name, Period, Statistic, Operator, and Threshold. For example, you can select CPU utilization for Metric Name, 1 minute for Period, Average for Statistic, > for Operator, and 70% for Threshold. This means that, in a statistical period of 1 minute, if the average CPU usage is greater than 70%, an alarm is triggered.
• You can add multiple rules to a policy. Any rule triggered will cause an alarm.

KCE enables you to create alarm policies for the following objects:

• Clusters
• Pods in a cluster
• Containers in a cluster

Select alarm objects

You can select the objects that you want to associate to the alarm rules.

Select alarm recipients

Select the recipients of the alarms triggered by the alarm rules.