You can view the resource usage of a dedicated host and configure alert rules in the Cloud Monitor console. This allows you to monitor the running status of a dedicated host.

View the resource usage of a dedicated host

  1. Log on to the Cloud Monitor console.
  2. In the left-side navigation pane, click Cloud products.
  3. In the Elastic calculation section, click Dedicated Host.
  4. On the Dedicated Host page, select the region and ID of the instance that you want to view.
  5. View the resource usage of a dedicated host.
    You can view the monitoring data of the last 30 days. Cloud Monitor monitors the computing, networking, and storage resources of a dedicated host. The following table describes the metrics.
    Resource type Metric Unit Description
    Computing resource UserCpuUtilization % The average CPU usage of all the ECS instances on a dedicated host
    Networking resource UserNetworkRxPPS PPS The sum of inbound packets of all the ECS instances on a dedicated host per second
    UserNetworkTxPPS PPS The sum of outbound packets of all the ECS instances on a dedicated host per second
    UserNetworkRxBandwidth bit/s The total inbound bandwidth of all the ECS instances on a dedicated host per second
    UserNetworkTxBandwidth bit/s The total outbound bandwidth of all the ECS instances on a dedicated host per second
    Storage resource The read capacity of virtual machines running on a dedicated host IOPS The total number of requests that read data from the disks of all the ECS instances on a dedicated host
    UserDiskWriteIOPS IOPS The total number of requests that write data to the disks of all the ECS instances on a dedicated host
    UserDiskReadBPS Byte/s The total number of bytes that are read from the disks of all the ECS instances on a dedicated host
    UserDiskWriteBPS Byte/s The total number of bytes that are written to the disks of all the ECS instances on a dedicated host

Create an alert rule for a dedicated host

You must create alert contacts and alert contact groups before you create alert rules. For more information, see Create an alert contact or alert group.

  1. Move the pointer over the metric for which you want to create an alert rule. Click the Create an alert rule icon in the upper-right corner.
  2. Associate an alert rule with a dedicated host.
    By default, the current dedicated host is associated with the alert rule. However, you can change the associated dedicated host.
    • To associate the alert rule with other dedicated hosts, select a region from the Region list. Then, select one or more dedicated hosts from the Dedicated Host list.
    • To associate the alert rule with all dedicated hosts, select All Resources from the Resource Range list.
  3. Configure the alert rule.
    The following table describes the parameters of a sample alert rule.
    Region Example Description
    Rule Name ddhxxx-cpu The alert rule monitors the average CPU usage of all the ECS instances on the dedicated host.
    Rule Description
    • UserCpuUtilization
    • 1-minute cycle
    • Continue for three cycles
    • Value
    • >=
    • 80%
    If the average CPU usage of all the ECS instances on the dedicated host is greater than or equal to 80% for three consecutive 1-minute cycles, an alert is triggered.
    Mute for 30 minutes If the alert conditions are still met 30 minutes after the first alert, another alert notification is sent.
    Effective Period 19:00 to 20:59 Specifies whether to trigger alerts only between 19:00 and 21:00.
  4. Configure a notification method.
    1. Select a contact group from the Contact Group list. Click Move To. The selected contact group is displayed in the Selected Groups list.
    2. Select an alert level.
      The alert level determines the notification method. The following three alert levels are supported:
      • Critical. If a critical alert occurs, a notification is sent to specified recipients by using a voice call, text message, DingTalk message, and an email.
      • Warning. If a warning alert occurs, a notification is sent to specified recipients by using a text message, DingTalk message, and an email.
      • Info. If an info alert occurs, a notification is sent to specified recipients by using an email and a DingTalk message.
    3. Set other optional parameters.
      The following parameters are optional:
      • Auto Scaling: Specify a scaling rule that is automatically executed when an alert is triggered.
      • Log Service: Specify a region, project, and Logstore. If an alert is triggered, the alert information is written to the Logstore.
      • Email Subject: The default email subject is Product Name + Metric Name + Instance ID.
      • Email Remark: Enter the email description.
      • HTTP CallBack: Enter a URL that is accessible from the Internet. Cloud Monitor sends HTTP POST requests to push alert messages to this URL.

      For more information, see Create a threshold-triggered alert rule.