Help Center/ TaurusDB/ User Guide/ DBA Assistant/ Performance Monitoring/ Viewing the Operating Status of a DB Instance
Updated on 2024-11-06 GMT+08:00
View PDF
Share

Viewing the Operating Status of a DB Instance

The Dashboard page allows you to view the operating status of the current DB instance, including alarms, health check results, compute resource usage, storage resource usage, and key performance metrics.

Alarms

  1. Log in to the management console.
  2. Click in the upper left corner and select a region and project.
  3. Click in the upper left corner of the page and choose Databases > GaussDB(for MySQL).
  4. On the Instances page, click the instance name.
  5. In the navigation pane, choose DBA Assistant > Real-Time Diagnosis.
  6. On the Dashboard page, view instance alarms provided by Cloud Eye.

    You can customize alarm rules by adjusting alarm policies and severities for key metrics, such as CPU usage and disk usage. To view alarm details, click the number next to an alarm severity.

    Figure 1 Alarms

    Figure 2 Alarm list

Health

In the Health area, you can view real-time health check results. By default, the data for high vCPU utilization, memory bottlenecks, high-frequency slow SQL statements, and lock waits are displayed.

For abnormal metrics, click Diagnose to view diagnosis details and suggestions. For details, see Table 1.

Figure 3 Health

Table 1 Health diagnosis and suggestions

Item

Exception Trigger Condition

High vCPU utilization

Either of the following conditions is met:

  • After you configure alarm rules on Cloud Eye, an alarm is reported, indicating the CPU usage is high.
  • The CPU usage exceeds 95% for more than 2.5 minutes of a 5-minute measurement period.

Memory bottleneck

Either of the following conditions is met:

  • After you configure alarm rules on Cloud Eye, an alarm is reported, indicating the memory usage is high.
  • The memory usage exceeds 95% within a 5-minute measurement period.

High-frequency slow SQL

Either of the following conditions is met:

  • After you configure alarm rules on Cloud Eye, an alarm is reported, indicating there are too many slow logs.
  • There are more than 100 slow logs within five minutes.

Lock wait

After you configure alarm rules on Cloud Eye, any of the following alarms is reported:

  • Row Lock Time
  • InnoDB Row Locks
  • Row Lock Waits

Compute Resource Usage

In the Compute Resource Usage area, the vCPU usage and memory usage are displayed by default. The displayed values are the average values for 5-minute measurement periods.

Figure 4 Compute Resource Usage

Storage Resource Usage

In the Storage Resource Usage area, the storage usage, disk read IOPS, and disk write IOPS are displayed by default. The displayed values are the average values for 5-minute measurement periods.

Figure 5 Storage Resource Usage

Key Performance Metrics

In the Key Performance Metrics area, the CPU usage & slow query logs, connections, memory utilization, and disk reads/writes from the last hour are displayed by default. The displayed values are real-time values.

Figure 6 Key Performance Metrics

Parent Topic: Performance Monitoring