Alerts for abnormal data

更新时间:
复制 MD 格式

Alibaba Cloud Real-Time Communication is integrated with the Cloud Monitor service. You can monitor metrics such as the percentage of abnormal meeting entries, stream ingest and pulling success rates, stuttering rate, and blur rate. You can also configure alert rules to report abnormal metrics in real time.

Notes

The Cloud Monitor service provides features for monitoring abnormal metrics and triggering alerts. To view detailed, end-to-end communication data in real time, see Call Inspector.

Configure Cloud Monitor permissions for a RAM user

If a Resource Access Management (RAM) user needs to view Cloud Monitor metrics, you must grant the required permissions to the user. For more information, see Manage RAM user permissions.

Permission nameDescription
AliyunCloudMonitorFullAccessPermissions to manage Cloud Monitor.
AliyunCloudMonitorReadOnlyAccessRead-only permissions for Cloud Monitor.
AliyunCloudMonitorMetricDataReadOnlyAccessPermissions to access time series metric data in Cloud Monitor.

Perform cloud service monitoring

Operation typeMethod
View monitoring dataCloud service monitoring
Set alert rules
View alert rules
Note When you use Cloud Monitor for Real-Time Communication (RTC), set the cloud service to Real-Time Communication.

RTC monitoring metrics

MetricDescription (Monitoring epoch: 60s)
Ratio of abnormal joins (by count)The percentage of abnormal meeting entries relative to the total number of meeting entries.
Slow First-Frame Ratio (Sessions)The percentage of first frames that take longer than 5000 ms to load relative to the total number of first frames pulled.
Stream ingest success rateThe percentage of successful stream ingests relative to the total number of stream ingests.
Stream Pull Session Success RateThe percentage of successful stream pulls relative to the total number of stream pulls.
Audio stuttering rateThe percentage of online users who experience audio stuttering.
Video stuttering rateThe percentage of online users who experience video stuttering.
Blurred person instance percentageThe percentage of online users who experience video blur.
End-to-End Network Latency (Percentage of Sessions)The percentage of online users who experience an end-to-end latency of over 1200 ms.