DocsICP FilingConsole
Documentation

Anomaly diagnosis

Updated at:
Copy as MD
Product
My Favorites

The Anomaly Diagnosis service is enabled by default for Enterprise Edition instances purchased on or before November 30, 2022. When your devices experience status or communication issues, you can use this service to view the anomaly detection results. This service helps you quickly and accurately pinpoint issues, reduce potential losses, and optimize device management.

Procedure

  1. Log on to the IoT Platform console.
  2. In the upper-left corner of the console, select the region where your Enterprise instance is located. On the Instance Overview page, find the Enterprise instance and click the instance name.

  3. In the left-side navigation pane, choose Maintenance > O&M Statistics.

  4. Click the Anomaly Diagnosis tab, and select an instance and a time range within the last month.
  5. In the Device Anomalies list, click View for the corresponding anomaly.
  6. On the Device Anomaly Details page, check metrics such as the number of online devices, uplink and downlink message TPS, online duration, and message latency for anomalies.
    Use the Basic Information, Metric Monitoring, and Influencing Factors sections to review anomaly details, such as its occurrence time, specific data, the affected devices, and contributing factors.

    For example, the page might show an anomaly for the number of online devices for instance iot-oxs****, with a sharp decrease from the previous day.

    The anomaly details page shows that the number of online devices dropped from 1,000 to 0 (a 100% day-over-day decrease), and the anomaly was detected at 2021/02/05 09:52:00. In the negative impact ranking under Influencing Factors, the device counts for two ProductKeys dropped from 999 to 0 and 1 to 0, respectively. The number of devices with the network connection method set to Wi-Fi dropped from 1,000 to 0.