Cloud monitoring by monitoring CDN's QPS, BPS, byte hit ratio, and so on, helps users get domain name usage. After a user adds an accelerated domain name, cloud monitoring automatically begins to monitor it, you are logged in to CDN for cloud monitoring You can view monitoring details on the page. You can also set alarm rules on monitoring items so that you receive alarm information when the data is abnormal.

Monitoring Services

  • Monitoring item description
  • Monitoring items Meaning Dimensions Unit Minimum monitor Granularity
    Number of visits per second Total number of visits/time granularity within the time grain Domain Name Times 1 minute
    Network bandwidth BPS Maximum network traffic per unit of time Domain Name BPS 1 minute
    Hit rate The probability of hit cache for the number of bytes requested in the time grain, note "bytes = number of requests x traffic ", the byte hit ratio more directly feedback back-to-back traffic Domain Name Percentage 1 minute
    Public network out of traffic That is, CDN's public network downstream traffic. Domain Name Bytes 5 minutes
    Return code 4xx Percentage of HTTP return code 4xx as all return codes within the time grain Domain Name Percentage 1 minute
    Return code 5xx share Percentage of HTTP return code 5xx as all return codes within the time grain Domain Name Percentage 1 minute
  • Viewing Monitoring Data
    1. Log on to the CloudMonitor console.
    2. Enter the list of CDN instances that the cloud service monitors.
    3. Click the Instance name or the monitor chart in the operation to go to the monitor details page.
    4. Click the size chart toggle button to toggle the larger image display (optional ).

Alarm service

  • Parameter descriptions
    • Monitor: Monitoring metrics provided by CDN.
    • Statistical Cycle: the alarm system checks if your monitoring data exceeds the alarm threshold for this cycle. For example, to set a memory usage alarm rule, the statistics cycle is 1 minute, an interval of 1 minute checks if memory usage exceeds the threshold.
    • Statistical Methods: Statistical Methods refer to settings that exceed the threshold range. You can set the average, maximum, minimum, count, and value in a statistical method.
      • Average: the average of the monitored data during the statistical cycle. The statistic result is the average of all metric data collected within 15 minutes. An average value of over 80% is deemed to exceed the threshold.
      • Maximum: the maximum value of the monitor data during the statistics cycle. The maximum value of monitoring data collected during the statistical cycle exceeds the threshold of 80.
      • Minimum: the minimum value of the monitored data during the statistics cycle. The minimum value for monitoring data collected during the statistical cycle exceeds the threshold of 80.
      • Value: Sum of monitoring data during the statistics cycle. When the sum of the metric data collected within the period is over 80%, it exceeds the threshold. Such statistical methods are required for traffic-class metrics.
    • Alarm after several consecutive exceeds threshold: refers to a number of consecutive statistical cycle monitoring items whose values continue to exceed the threshold to trigger an alarm.

      Example: Set CPU usage to exceed 80% Alarm, with a statistical cycle of 5 minutes, alarm three consecutive times after exceeding the threshold, when the first time the detection CPU usage exceeds 80%, no alarm notification is issued. Second Probe in 5 minutes CPU usage exceeds 80%, and no alarm will be issued. The third probe still exceeds 80% Alarm notification will be issued only when. That is, from the first time the actual data exceeds the threshold to the final alarm rule, the minimum time required is the statistical cycle (number of consecutive probes-1) = 5 (3-1) = 10 minutes.

  • Set alarm rules
    1. Log in to the clLog on to the CloudMonitor console.oud monitoring console.
    2. Enter the list of CDN instances that the cloud service monitors.
    3. Click the alarm rule in the instance List action to enter the alert Rule Page for the instance.
    4. Click new alarm rule in the upper-right corner of the alarm Rule Page to create an alarm rule based on the parameter.