RDS PostgreSQL

Monitoring Indicators and Alarms

2025-06-17 08:23:48

Supported Monitoring Indicators

Application Scenarios

The RDS-PostgreSQL product provides detailed monitoring indicators viewing functions. The supported monitoring indicators are listed below.

 

Monitoring Indicators

Monitoring Type

Monitoring Indicator

Resource   Monitoring (Monitoring related to node machine)

CPU   Usage

Memory   Usage

Memory   Usage

IO   Operand

IO   Throughput

io-util

Hard   Disk Free Space

Hard   Disk Idle Rate

Hard   Disk Used Space

Hard   Disk Usage

Network   Throughput

WAL   Log Space

Engine   monitoring (Monitoring related to node   database engine)

TPS

QPS

Number   of Sessions

Number   of Database Reads and Writes in a Period

Number   of Operation Rows

bgwriter   stat

Checkpoint

Checkpoint   Performance for Writing To Disk

Maximum   time (session time)

Number   of workers

Slow   query statistics

Long   Transaction

System   Cache Hit Ratio

Average   Number of Scans per Second Throughout a Large Table Exceeding 10 MB

Number   of Bytes Per Second Generated By a Temporary File in a Period

Number   of Deadlocks Per Second in a Period

Number   of Stored Procedure Calls Per Second in a Period

 

Viewing Monitoring Indicators

Application Scenarios

View the detailed monitoring data. For the monitoring data supported by RDS-PostgreSQL, see Supported Monitoring Indicators.

Viewing Steps for Monitoring Indicators of Instance

1.           Log in to the eSurfing Cloud Portal.

2.           Go to the Console.

3.           Select the resource pool region in the upper left corner. For details of resource pools supported by RDS-PostgreSQL, see Function Loading Description.

4.           Select RDS-PostgreSQL and click it to go to the Console Administration page.

5.           Select the dashboard on the left sidebar, as shown in the picture below.

               

6.           Click the View button in the Chart column of the corresponding node to jump to the detailed monitoring information page of the corresponding node.

                

Setting Alarm Rules

Application Scenarios

The RDS-PostgreSQL product allows you to set the alarm rules. You can customize monitoring targets and notification policies to learn the running status of database services to realize warning purposes.

Setting Steps for Alarm Rules

1.           Log in to the eSurfing Cloud Portal.

2.           Go to the Console.

3.           Select the resource pool area in the upper left corner. For details of the resource pool loaded by RDS-PostgreSQL, see Function Loading Description.

4.           Find RDS-PostgreSQL and click it to go to the Console page.

5.           Click the

                

                icon in the upper right corner to jump to the Database Management platform.

6.           In the platform, click Use Now > Alarm Center > Alarm Settings > Create Policy to set the alarm rules.

7.           After completing the alarm rule parameters, click Save and enable the alarm rule.

Parameter Interpretation of Alarm Rules

Parameter Name

Parameter   Interpretation

Policy   name

The   name of the alarm policy. One alarm policy can contain multiple alarm rules.

Resource   pool

The   resource pool corresponding to the alarm policy.

Remarks

Remarks   information for the alarm policy.

Alarm   object

You   can select the eSurfing Cloud database component and select the corresponding   alarm instance. For instance, you can select Instance PostgreSQL and   then click Select Instance to select the corresponding instance into   the alarm rule.

Template   type

You   can manually create an alarm rule. You can go to Alarm Center > Alarm   Templates and create alarm rules by applying existing templates.

Threshold   alarm

You   can select the instance CPU usage, memory usage, disk usage, TPS, QPS, WAL   log space, and other monitoring indicators.

Event   alarm

You   can select instance status, agent process status, backup program status,   cluster status, and other indicators.

Silence   period

Specify   the interval for resending the alarm if the monitoring item does not return   to normal after the alarm occurs.

Alarm   level

You   can select three alarm levels: emergency, alarm, and ordinary.

Alarm   mode

You   can select mailbox, SMS, and Enterprise WeChat. You can also select multiple   alarm methods for alarms at the same time. After filling in, an additional   alarm message will be pushed to the specified alarm method.

Alarm   contact

After   selection, the alarm information will be pushed to the mailbox and a text   message bound by the contact.

Alarm   contact group

After   selection, the alert information will be pushed to contacts in the group.


E3XOqCZJvIB_