Help Center/ GaussDB/ Best Practices/ Suggestions on GaussDB Metric Alarm Configuration
Updated on 2025-09-12 GMT+08:00

Suggestions on GaussDB Metric Alarm Configuration

You can set alarm rules on the Cloud Eye console to specify the monitored objects and notification policies for your instances and keep track of the instance status. This section describes how to configure GaussDB metric alarm rules.

Creating a Metric Alarm Rule

  1. Log in to the management console.
  2. Under Management & Governance of the service list, click Cloud Eye.
  3. In the navigation pane on the left, choose Cloud Service Monitoring.
  4. Click GaussDB in the list.
  5. Locate the instance for which you want to create an alarm rule, click More in the Operation column, and choose Create Alarm Rule.
  6. On the displayed page, set parameters as required.

    Table 1 Alarm rule information

    Parameter

    Description

    Name

    Alarm rule name. The system generates a random name, and you can change it if needed. The value can contain only letters, digits, underscores (_), and hyphens (-), and cannot exceed 128 characters.

    Description

    Description of the alarm rule. The value can contain a maximum of 256 characters. This parameter is optional.

    Method

    Mode for configuring an alarm policy. You can select Associate template or Configure manually.

    • Configure manually: You can create a custom alarm policy as needed.
    • Associate template: If the same alarm rule needs to be configured for multiple GaussDB instances, you can use an alarm template to simplify operation.

    Template

    This parameter is only available if you select Associate template for Method.

    You can select a default alarm template or create a custom one.

    After an associated template is modified, the policies contained in this alarm rule to be created will be updated accordingly.

    Alarm Policy

    This parameter is only available if you select Configure manually for Method.

    An alarm is triggered when the metric configured for this alarm reaches the preset threshold in consecutive periods. For example, an alarm is triggered if the average CPU usage is 80% or higher for three consecutive 5-minute periods.

    A maximum of 50 alarm policies can be added to an alarm rule. If any of these alarm policies is met, an alarm will be triggered.

    Table 2 Alarm notification

    Parameter

    Description

    Alarm Notifications

    Specifies whether to notify users when alarms are triggered. Notifications can be sent by email or text messages, or through HTTP/HTTPS requests to servers. This function is enabled by default.

    Notified By

    The following three options are available:

    • Notification policies: Flexible alarm notifications by severity and more notification channels are provided.
    • Notification groups: Configure notification templates on the Cloud Eye console.
    • Topic subscriptions: Configure notification templates on the Simple Message Notification (SMN) console.

    Notification Policies

    This parameter is only available if you select Notification policies for Notified By. Select one or more notification policies. You can specify the notification group, window, template, and other parameters in a notification policy.

    For how to create a notification policy, see Creating, Modifying, or Deleting a Notification Policy.

    Notification Group

    This parameter is only available if you select Notification groups for Notified By. Select the notification groups to which alarm notifications will be sent.

    For details about how to create a notification group, see Creating a Recipient and Notification Group.

    Recipient

    This parameter is only available if you select Topic subscriptions for Notified By. You can select the account contact or a topic as the object to which alarm notifications will be sent.

    • The account contact is the mobile phone number and email address of the registered account.
    • A topic is a specific event type for publishing messages or subscribing to notifications. If the required topic is unavailable, create one first and add subscriptions to it. For details, see Creating a Topic and Adding Subscriptions.

    Notification Template

    This parameter is only available if you select Notification groups or Topic subscriptions for Notified By. You can select an existing template or create a new one to send alarm notifications.

    Notification Window

    This parameter is only available if you select Notification groups or Topic subscriptions for Notified By.

    Cloud Eye sends notifications only within the notification window you specified.

    If Notification Window is set to 08:00-20:00, alarm notifications are sent only from 08:00 to 20:00.

    Trigger Condition

    This parameter is only available if you select Notification groups or Topic subscriptions for Notified By.

    You can select either Generated alarm or Cleared alarm, or both.

    Enterprise Project

    Enterprise project that the alarm rule belongs to. Only users who have the permissions of the enterprise project can view and manage this alarm rule.

    Tags

    Key-value pairs that you can use to easily categorize and search for cloud resources. You are advised to create predefined tags in Tag Management Service (TMS).

    If your organization has configured tag policies for Cloud Eye, you need to add tags to alarm rules based on tag policies. If a tag does not comply with the policies, an alarm rule may fail to be created. Contact your organization administrator to learn more about tag policies.

    • A key can contain up to 128 characters, and a value can contain up to 225 characters.
    • You can add up to 20 tags.

  7. Click Create. The alarm rule is created.

    For details about how to create alarm rules, see Creating an Alarm Rule.

Metric Alarm Configuration Suggestions

Metric ID

Metric Name

Definition

Threshold in Best Practices

Alarm Severity in Best Practices

io_bandwidth_usage

Disk I/O Bandwidth Usage

Percentage of the maximum disk I/O bandwidth currently used

Raw data > 80% for three consecutive periods

Major

iops_usage

IOPS Usage

Percentage of the maximum disk IOPS currently used

Raw data > 80% for three consecutive periods

Major

rds001_cpu_util

CPU Usage

CPU usage of a measured object

Raw data > 80% for three consecutive periods

Major

rds002_mem_util

Memory Usage

Memory usage of a monitored object

Raw data > 90% for three consecutive periods

Major

rds007_instance_disk_usage

Instance Disk Usage

Real-time data disk usage of the monitored instance

Raw data > 75% for three consecutive periods (The threshold should not be set above 80%.)

Major

rds020_avg_disk_ms_per_write

Time Required for per Disk Write

Average time required for a data disk write on the monitored node in a measurement period

Raw data > 8 ms for three consecutive periods

Major

rds021_avg_disk_ms_per_read

Time Required for per Disk Read

Average time required for a data disk read on the monitored node in a measurement period

Raw data > 8 ms for three consecutive periods

Major

rds036_deadlocks

Deadlocks

Incremental number of database transaction deadlocks in a measurement period

Raw data > 5 counts for three consecutive periods

Major

rds048_P80

Response Time of 80% SQL Statements

Real-time response time of 80% of database SQL statements

Raw data > 10000000 μs for three consecutive periods

Major

rds049_P95

Response Time of 95% SQL Statements

Real-time response time of 95% of database SQL statements

Raw data > 15000000 μs for three consecutive periods

Major

rds060_long_running_transaction_exectime

Maximum Execution Duration of Database Transactions

Real-time maximum execution duration of database transactions execution on a monitored object

Raw data > 7200s for three consecutive periods (You are advisable to manually terminate a transaction if its duration is longer than 2 hours. Adjust this threshold based on workload requirements.)

Major

rds063_slowquery_user

Slow SQL Statements in the User Database

Real-time number of slow SQL statements in the user databases on the primary DN or CN in a measurement period

Raw data > 15 counts for three consecutive periods

Major

rds065_dynamic_used_memory_usage

Dynamic Memory Usage

Real-time dynamic memory usage of a monitored object

Raw data > 80% for three consecutive periods

Major

rds066_replication_slot_wal_log_size

WAL Log Size in the Replication Slot

Real-time size of WAL logs reserved in the replication slot of a primary DN

Raw data > [10% of the storage] bytes for three consecutive periods (10% is the recommended value. Adjust this threshold based on the purchased storage.)

Major

rds070_thread_pool

Thread Pool Usage

Real-time thread pool usage on a CN or DN

Raw data > 85% for three consecutive periods

Major