Create threshold-triggered alert rules

You can create threshold-triggered alert rules in CloudMonitor to monitor IoT resources, and send alert notifications to specific contacts by using the specified method. You can monitor the number of online devices that are under a specific product, the number of TSL model communication failures, and the number of times that data is forwarded by the rule engine.

1. Log on to the CloudMonitor console.
2. In the left-side navigation pane, choose Alarms > Alarm Rules.
3. On the Threshold Value Alarm tab of the Alarm Rules page, click Create Alarm Rule.
4. Select IoT Platform as the related resource and specify the product to be monitored.
5. Set alert rules. Add the required alert rules.

   In this example, three threshold-triggered alert rules are set, and the alert mute period is set to 10 minutes. (If an alert is not cleared 10 minutes after the alert is triggered, the alert notification will be sent again.)

   

   The following table describes the rules shown in the preceding figure.

   Rule name	Rule description	Description
   DevicePropertyReportError	The statistical period is 15 minutes. An alert is triggered if a minimum of 100 property reporting failures are detected for three consecutive statistical periods.	This rule is used to monitor the property reporting status of all devices under the temperature-based sensory product. If a large number of consecutive property reporting failures occur, you must check the devices and network status.
   MessageCountForwardedThroughRuleEngine_FC	The statistical period is 1 minute. An alert is triggered if a minimum of 10,000 messages are forwarded to Function Compute for three consecutive statistical periods.	According to the design, when the temperature reported by a device is higher than 25°C, the rules engine forwards the temperature data to Function Compute. If 10,000 messages are forwarded, a large number of devices have reported more than two temperatures that are above 25°C within 1 minute. We recommend that you monitor the temperature change and perform early troubleshooting of potential safety risks.
   OnlineDevicesCount_MQTT	The statistical period is 5 minutes. An alert is triggered if less than 4,800 devices are online for three consecutive statistical periods.	A large number of devices are offline at the same time. The devices and network status must be checked.

6. Set the alert notification method. Select an alert group and a notification method. You can specify or configure the other fields.
7. Click Confirm to complete rule creation.