View host performance

OceanBase Cloud Platform (OCP) allows you to view the performance monitoring data of hosts in different dimensions from the following entries:

• Entry 1:

  1. Log on to the OCP console.

  2. In the left-side navigation pane, click Clusters to go to the Clusters page.

  3. In the Clusters list, find the target cluster and click its name.

  4. In the left-side navigation pane on the page that appears, click Performance Monitoring.

  5. On the Cluster Performance tab, view the performance monitoring data of hosts.

• Entry 2:

  1. Log on to the OCP console.

  2. Click Hosts.

  3. In the Hosts list, find the host whose monitoring information that you want to view and click its IP address.

  4. On the page that appears, click the Monitoring tab.

  5. In the Host Performance tab, view the performance monitoring data of the host.

• Entry 3:

  1. Log on to the OCP console.

  2. In the left-side navigation pane, choose Performance Monitoring.

  3. The OceanBase Cluster page automatically appears.

  4. Click the Host Performance tab and view the performance monitoring data of hosts.

Prerequisites

• You have the read-only permission on the target cluster.
• The drill-down analysis feature is supported only when you go to the Host Performance tab from Entry 1 or 3.

Procedure

The following procedure takes Entry 3 as an example.

1. Log on to the OCP console.

2. In the left-side navigation pane, choose Performance Monitoring.

3. The OceanBase Cluster tab automatically appears.

4. In the Data Filtering section, configure the conditions for filtering monitoring data.

   The following table describes the filter conditions.

   Filter condition	Description
   Cluster	This parameter is not displayed in auto-refresh mode. Select a cluster to view its monitoring information.
   Zone	By default, the monitoring information of all zones is displayed.
   OBServer	By default, the monitoring information of all OBServer nodes is displayed.
   Select Time	This parameter is not displayed in auto-refresh mode. Select a time range for the data that you want to query.
   Auto Refresh	By default, auto-refresh is disabled. After auto-refresh is enabled, data is refreshed every 5s.

   

5. Click the Host Performance tab and view the performance monitoring data of hosts. If the statistical range covers multiple OBServer nodes, the aggregate metric values of the OBServer nodes are displayed.

   For more information about monitoring metrics, see Monitoring metrics. The following features enable you to better view and analyze the performance monitoring data on a trend chart. You can learn about the features with the help of the figure below.

   

   1. Hide the data of a submetric: To use this feature, click the icon of the submetric. The submetric will then turn gray and its data will be hidden from the chart.

   2. View detailed data of a specific point in time or a period of time: You can hover the pointer over a point in time on the trend chart to view the detailed data. You can also drag the pointer to select a period of time on the trend chart and have a zoomed-in view of data in this period of time. To resume the normal view, double-click the trend chart.

   3. Zoom in on the trend chart: To use this feature, click the zoom-in icon in the upper-right corner. The features indicated by ①, ②, and ④ in the preceding figure are also applicable to a zoomed-in chart.

   4. Drill down to view monitoring data in different dimensions: When an abnormal metric exists, such as a long response time of a tenant, perform a drill-down analysis to identify the faulty server. Click Drill-down Analysis in the upper-right corner of the trend chart. The following drill-down dimensions are supported: cluster > host.

      The features indicated by ① and ② in the preceding figure are also applicable to a trend chart on a drill-down page. You can also perform the following operations:

      

      1. In the Data Filtering section of the drill-down page, specify filter conditions. If you do not specify any filter conditions, the drill-down page automatically inherits filter conditions from the upper-level page, as indicated by ① in the preceding figure.

      2. If the metric has submetrics, you can view the submetrics, as indicated by ② in the preceding figure, or click the link indicated by ③ to drill down to other dimensions for further analysis.

      3. If the metric is related to host performance, you can hover the pointer over the Go to Host Performance Monitoring link indicated by ③ in the preceding figure and click the IP address of a host to go to the Monitoring page of the host.

      4. You can also click the icon indicated by ④ in the upper-right corner of a trend chart, and select the hosts whose performance monitoring data needs to be displayed on the trend chart.

6. (Optional) Configure the monitoring data collection interval.

   By default, OCP collects monitoring data per second. This collection interval has high requirements for the OCP server performance and MonitorDB storage performance. If you do not require a high collection frequency, you can change the value of the ocp.metric.collect.interval.second parameter in the config_properties file to increase the monitoring data collection interval, thus reducing required resources and increasing the manageable scale.

   Note

   The system supports only the following values for this parameter: 1, 5, 10, and 15. If the value you entered is not one of them, the system adjusts the value based on the following rules:

   • If the entered value is smaller than 1, it is adjusted to 1.
   • If the entered value is in the range of [1,5), it is adjusted to 1.
   • If the entered value is in the range of [5,10), it is adjusted to 5.
   • If the entered value is in the range of [10,15), it is adjusted to 10.
   • If the entered value is greater than 15, it is adjusted to 15.

   After you change the value, the configuration immediately takes effect. OCP collects monitoring data at this new interval. By default, the monitoring data display interval is three times the monitoring data collection interval, except for the case when the collection interval is 1 second. For example, if the monitoring data is collected every 5 seconds, the queried monitoring data is displayed by an interval of 15 seconds.