Alarm Feature

Last updated: 2020-02-25 17:05:20


Operation Scenario

You can create alarms to trigger alarms and send alarm messages when the cloud product status changes. The created alarm can determine whether an alarm-related notification needs to be triggered according to the difference between the monitored index and the given threshold at intervals.
You can take appropriate precautionary or remedial measures in a timely manner when the alarm is triggered. Therefore, properly created alarms can help you improve the robustness and reliability of your applications. For more information on alarms, see Alarm Configuration In Cloud Monitor.

When a user needs to send an alarm for a specific status of a product, it is required to create an alarm policy at first. An alarm policy is composed of three compulsory components, that is, the name, type and alarm triggering conditions. Each alarm policy is a set of alarm triggering conditions with the logical relationship "or", that is, as long as one of the conditions is met, an alarm will be triggered. The alarm will be sent to all users associated with the alarm policy. Upon receiving the alarm, the user can view the alarm and take appropriate actions in time.

Make sure that you have set the default alarm recipient; otherwise, the default alarm policy of TencentDB won't be able to send notifications.


Creating an Alarm Policy

  1. Log in to the Cloud Monitor Console And select Alarm Configuration > Alarm Policy On the left sidebar.
  2. In the alarm policy list, click CREATE .
  3. Set the policy name, policy type, target product, alarm object, and trigger.
  • Policy type: it is divided into host monitoring and slave monitoring, which are suitable for unused instance types respectively.
    • Deploy monitoring on the master: When the monitored instance isa master instance which is not a slave of any instance, replicating relevant monitoring data from the master won't work, and the IO and SQL threads are disabled. Replicating relevant monitoring data can work and the IO and SQL threads can be enabled only when the monitored instance isa disaster recovery or a read-only instance.
    • Deploy monitoring on the slave: The master instance and disaster recovery instance in HA edition come in a one-master-one-slave architecture by default. As a result, replicating relevant monitoring data from the slave can work only when the monitored instance isa master or disaster recovery instance. Such data can be used to reflect the delay distance and time between the master or disaster recovery instance and its hidden slave nodes. You are recommended to keep an eye out for the relevant monitoring data of the slave. If the master or disaster recovery instance fails, its monitored hidden slave nodes can be promoted into the master instance quickly.
  • An alarm trigger is a semantic condition composed of metric, comparison, threshold, statistical period, and duration. For example, if the metric is disk utilization, the comparison is >, the threshold is 80%, the statistical period is 5 minutes, and the duration is two statistical periods, then the data on disk utilization of a database will be collected once every five minutes, and an alarm will be triggered if the disk utilization exceeds 80% for two consecutive times.
  • The object instance to be associated with can be found by selecting the region where the object is located or searching for the instance ID of the object.
  1. After confirming everything is correct, click Complete .

Associating Objects

After the alarm policy is created, you can associate some alarm objects with it. When an alarm object satisfies an alarm trigger, an alarm notification will be sent.

  1. In the alarm policy list, click the name of an alarm policy to enter the alarm policy management page.

  2. Click CREATE .

  3. Select a desired Tencent Cloud product and click Apply To associate it with the alarm policy.

Setting an Alarm Recipient

Alarm recipients are those who will receive alarm messages.

  1. In the alarm policy list, click the name of an alarm policy.
  2. On the alarm policy management page, select Alarm Recipient Group And click Edit .
  3. Select the user group to be notified, set relevant options, and click Save .