Help Center/ Cloud Eye/ User Guide/ Alarm Management/ Alarm Rules/ Creating an Alarm Rule (Earlier Version)
Updated on 2026-10-15 GMT+08:00
View PDF
Share

Creating an Alarm Rule (Earlier Version)

To monitor the usage of cloud service resources or key operations on them, you can create an alarm rule. After the alarm rule is created, if a metric reaches the specified threshold or the specified event occurs, Cloud Eye immediately informs you of the exception through SMN.

This topic describes how to create an alarm rule.

In some regions, Cloud Eye has been upgraded to the latest version. The alarm rule page has been optimized. The upgrade is being rolled out in other regions. For details, see the console. This section describes how to configure alarm rules in the earlier version. For details about how to configure alarm rules in the latest version, see Creating an Alarm Rule (New Edition).

Billing

Alarm notifications sent by SMN will be billed based on their frequency and message length. For details, see Product Pricing Details.

Constraints

  • When Specific resources is selected for the monitoring scope, a maximum of 1,000 resources can be selected for a single alarm rule by default. If the current resource quota cannot meet your requirements, you can apply for a higher quota by referring to Quotas. The maximum quota is 1,500. The actual quota varies depending on the service. If the number of resources exceeds 1,000, you are advised to use resource groups for configuration.
  • When creating an alarm rule of the metric type, if you select All resources or Resource groups as the monitoring scope, you can exclude resources. By default, a maximum of 100 resources can be excluded for a single alarm rule. If the current resource quota cannot meet your requirements, you can apply for a higher quota by referring to Quotas. The maximum quota is 500. The actual quota varies depending on the service.

Prerequisites

Alarm Type

Prerequisites

Metric
  • Before creating a metric alarm rule for a cloud service, ensure that its monitoring data can be queried in the cloud service monitoring list. For details about the conditions for displaying cloud services in the cloud service monitoring list, see Cloud Services Displayed in the Cloud Service Monitoring List.
  • Before creating an alarm rule for Agent metrics, ensure that Agent has been installed on the server. For details, see Installing the Agent.
  • Before creating an alarm rule for process monitoring metrics, ensure that you have added custom processes for monitoring. For details, see Adding Process Monitoring.

Event

Before creating an alarm rule for a custom event, ensure that the event source is the same as that specified in Reporting an Event.

Creating an Alarm Rule

  1. Log in to the Cloud Eye console.
  1. In the navigation pane, choose Alarm Management > Alarm Rules.
  2. On the displayed page, click Create Alarm Rule in the upper right corner.
  3. On the Create Alarm Rule page, configure parameters.
    Assume that you want to notify all contacts in the contact group Notification_Group every hour when the raw data of CPU usage of any ECS exceeds 80% for three consecutive periods. In this case:
    1. Set Alarm Type to Metric.
    2. Set Cloud Product to Elastic Cloud Server - ECSs.
    3. Set Resource Level to Cloud product.
    4. Set Monitoring Scope to All resources.
    5. Set Method to Configure manually and set an alarm policy to trigger Hourly alarms when Raw data of ECSs / CPU Usage > 80% (marked as Major) for 3 times (consecutively).
    6. Enable Alarm Notifications and set Notified By to Contact groups.
    7. Select Notification_Group for Contact Group. Retain the default values for other parameters or leave them blank.
    8. Click Create.

Basic Parameters of an Alarm Rule

Table 1 Basic information

Parameter

Description

Name

Name of the alarm rule. The name is automatically generated, but you can change it to a custom one. The rule name cannot exceed 128 characters and can contain only letters, digits, underscores (_), and hyphens (-).

Description

(Optional) Alarm rule description. It can contain up to 256 characters.

Parameters of the Alarm Content

The alarm content parameters vary depending on the alarm type. You can check the parameter description based on the selected alarm type.

When setting Alarm Type to Metric, you can configure the alarm content by referring to the following table.

Table 2 Parameters of the metric alarm content

Parameter

Description

Example Value

Cloud Product

Name of the monitored service.

For details about supported cloud products and their metrics, see Cloud Product Metrics.

Elastic Cloud Server - ECSs

Resource Level

Resource level of the monitored object. You can select Cloud product or Specific dimension. Cloud product is recommended.

Take ECS as an example. ECS is the cloud product. Specific dimensions are disks, mount points, processes, and more. For details about dimensions, see Dimensions.

NOTE:

If you select Cloud product, metrics across dimensions (such as Disk Usage and CPU Usage) can be configured in the same alarm rule. If you select Specific dimension, only metrics of the specified dimension can be configured for the same alarm rule.

Cloud product

Monitoring Scope

Monitoring scope the alarm rule applies to. You can select All resources, Resource groups, or Specific resources.

  • All resources: An alarm will be triggered if any resource in the selected cloud product meets the alarm policy.
  • Resource groups: An alarm will be triggered if any resource in the selected resource group meets the alarm policy.
  • Specific resources: Click Select Specific Resources to select resources. An alarm will be triggered if any selected resource meets the alarm policy.

Specific resources

Group

When Monitoring Scope is set to Resource groups, you need to select a group. If no resource group meets your needs, click Create Resource Group to create one.

After selecting a resource group from the drop-down list, you can click View Resource Details in a Group to view the details of resources in the group. After an alarm rule is configured, the group cannot be modified.

NOTE:

If the resource group contains an EVS resource with the type in the format of ECS instance ID-volume-Volume ID, the instance cannot report monitoring data after the alarm rule is created. As a result, no alarm can be triggered.

-

Instance

When Monitoring Scope is set to Specific resources, you need to select the monitored objects for the alarm rule.

Click Select Specific Resources to select desired resources.

-

Method

Method for configuring an alarm policy. The options are as follows:

  • Configure manually: You can create a custom alarm policy as needed.
  • Associate template: If you need to configure the same alarm rule for multiple groups of resources under the same cloud product, you can use an alarm template to simplify operations.

Configure manually

Template

When you set Method to Associate template, you need to select a template.

You can select a default or custom template.

An alarm template may contain alarm policies of multiple cloud products or different dimensions of the same cloud product. When you create an alarm rule, the alarm policies vary according to the resource level.

  • When you set Resource Level to Cloud product, all alarm policies of the cloud product in the alarm template will be synchronized to the alarm rule. Alarm policies of other cloud products will not be synchronized.
  • When you set Resource Level to Specific dimension, only alarm policies of the same dimension as the current resource in the alarm template will be added to the alarm rule.

-

Alarm Policy

When you set Method to Configure manually, you need to configure alarm policies. You can add up to 50 alarm policies for a single alarm rule. You can choose to send alarm notifications when any of the policies is met or when all policies are met. For details about alarm policy parameters, see Configuring Alarm Policies for Metrics.

You can customize alarm policy parameters or select Use Template to select an alarm template and modify the alarm policies in that template. Changes made will not affect the original policies in that template.

An alarm will be triggered when the monitored data reaches the threshold for the specified consecutive times. For example, Cloud Eye triggers a major alarm if the average CPU usage of the monitored object is 80% or more for three consecutive 5-minute periods.

-

Operation

If there are multiple alarm policies, you can click Delete to delete one.

NOTE:

You can also click Create above the alarm policy list to add more alarm policies.

Delete

When setting Alarm Type to Event, you can configure the alarm content by referring to the following table.

Table 3 Parameters of the event alarm content

Parameter

Description

Example Value

Event Type

This parameter is only available if Alarm Type is set to Event. You can select either System event or Custom event. For details about the events supported by each cloud service, see Events Supported by Event Monitoring.

System event

Event Source

This parameter is only available if Alarm Type is set to Event.

  • If Event Type is set to System event, select the cloud service from which the event comes.
  • If Event Type is set to Custom event, the event source must be the same as that of the reported source. For details about how to report an event, see Reporting Events.

Elastic Cloud Server

Monitoring Scope

Monitoring scope the alarm rule applies to.

  • All resources: An alarm will be triggered if any resource in the selected cloud product meets the alarm policy.
  • Resource groups: An alarm will be triggered if any resource in the selected resource group meets the alarm policy.
  • Specific resources: Click Select Specific Resources to select resources. An alarm will be triggered if any selected resource meets the alarm policy.
NOTE:
  • This parameter is available when Event Type is set to System event.
  • Currently, Resource groups can be selected as the monitoring scope only when the event source is ECS (supported only in certain regions; for details, check the console), DDS, RDS, DCS, or GeminiDB.

Specific resources

Group

When Monitoring Scope is set to Resource groups, you need to select a group. If no resource group meets your needs, click Create Resource Group to create one.

After selecting a resource group from the drop-down list, you can click View Resource Details in a Group to view the details of resources in the group. After an alarm rule is configured, the group cannot be modified.

-

Dimension

When you set Monitoring Scope to Specific resources, select a dimension of the resources to connect the event data of the selected cloud service to Cloud Eye by dimension.

Document Database Instances

Instance

When Monitoring Scope is set to Specific resources, you need to select the monitored objects for the alarm rule.

Click Select Specific Resources to select desired resources.

-

Method

Method used to create an alarm rule. If you select System event for Event Type, Method can be Associate template or Configure manually. If you select Custom event for Event Type, only Configure manually is supported.

  • Configure manually: You can create a custom alarm policy as needed.
  • Associate template: If you need to configure the same alarm rule for multiple groups of resources under the same cloud product, you can use an alarm template to simplify operations.

Configure manually

Template

When you set Method to Associate template, you need to select a template.

You can select a default or custom template.

-

Alarm Policy

When you set Method to Configure manually, you need to configure alarm policies. You can add up to 50 alarm policies to an alarm rule. For details about alarm policy parameters, see Configuring Alarm Policies for Events.

You can customize alarm policy parameters or select Use Template to select an alarm template and modify the alarm policies in that template. Changes made will not affect the original policies in that template.

The event that triggers an alarm is an instant operation. For example, an alarm is triggered if a VM is restarted.

-

Operation

If there are multiple alarm policies, you can click Delete to delete one.

NOTE:

You can also click Create above the alarm policy list to add more alarm policies.

Delete

Alarm Notification Parameters

If no notification is received after alarm notification is enabled, rectify the fault by referring to Why Do Subscribers Fail to Receive Messages After I Publish Messages to a Topic?

Notification policies, contact groups, and notification templates are available only in some regions. These features will be available in other regions later. For details, check the console.

Table 4 Configuring alarm notifications

Parameter

Description

Example Value

Alarm Notifications

Whether to send alarm notifications by SMS, email, FunctionGraph (function), FunctionGraph (Workflow), WeCom, DingTalk, Lark, WeLink Chatbot, HTTP, or HTTPS. This parameter is enabled by default.

Enabled

Notified By

Notification method. Select Notification policies, Contact groups, or Topic subscriptions.

  • Notification policies: If several O&M owners handle service maintenance, you can set up different notification scopes in a notification policy to alert specific owners by Alarm Severity or at a specified schedule. By configuring multiple notification scopes in a notification policy, you can implement alarm severity-based notification and personnel scheduling. For details, see Creating a Notification Policy.
  • Contact groups: Cloud Eye provides alarm notification capabilities. Select this option if you do not need to send alarm notifications by severity or schedule. To send alarm notifications to a group of contacts, you can create a contact group and add contacts to it in Cloud Eye for easier management. For details, see Creating a Contact/Contact Group.
  • Topic subscriptions: If the current region does not support alarm notifications or the recipients have been configured in SMN, you can select a topic created in SMN or your cloud account contact as required.

    For details, see Creating a Topic and Adding Subscriptions.

NOTE:

Cloud Eye uses SMN for alarm notifications. If SMN has major delays, the alarm notifications may also be delayed.

Contact groups

Notification Window

If Notified By is set to Contact groups or Topic subscriptions, you need to set the notification window.

Cloud Eye sends notifications only within the validity period specified in the alarm rule.

For example, if the notification window is set to 08:00 to 20:00, notifications are sent only within this specified time range when a metric reaches the specified threshold or a specified event occurs.

08:00-20:00

Time Zone

Time zone for the alarm notification window. By default, it matches the time zone of the client browser, but can be manually configured.

(GMT+08:00) Beijing, Chongqing, Hong Kong, Urumqi, Kuala Lumpur, Singapore, Perth, Taipei, Irkutsk, Ulaanbaatar

Trigger Condition

This parameter is required when you set Notified By to Contact groups or Topic subscriptions.

  • If you set Alarm Type to Metric, select Generated alarm, Cleared alarm, or both for this parameter.
  • If you set Alarm Type to Event, you can only select Generated alarm for this parameter.

Generated alarm

Notification Template

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

NOTE:

Some cloud services do not support fields such as the resource name, enterprise project, resource tag, private IP address, and public IP address in the notification template. When you select a system template as the notification template, these fields will appear blank in the alarm notifications even if you have them configured.

-

(Optional) Parameters of Advanced Settings

Table 5 Advanced settings

Parameter

Description

Enterprise Project

Enterprise project that the alarm rule belongs to. Only users with the enterprise project permissions can manage the alarm rule. For details about how to create an enterprise project, see Creating an Enterprise Project.

Tags

Key-value pairs that you can use to easily categorize and search for cloud resources. You are advised to create predefined tags in TMS. For details, see Creating Predefined Tags.

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

  • A key can contain up to 128 characters, and a value can contain up to 255 characters.
  • You can create up to 20 tags.

No-Data Processing

If Alarm Type is set to Metric, you can configure whether to trigger alarms when no data is reported. This parameter is selected by default. When enabled, an insufficient data alarm record will be generated if the configured metric reports no data for three consecutive hours.

Reference

If Notified By is set to Notification policies and System template is selected for Notification Template, SMN sends alarms in non-JSON format. To receive JSON-formatted alarms, select a notification template whose Protocol is HTTP/HTTPS and Format is JSON under Notification Policies. For details, see Creating, Deleting, Copying, or Modifying a Notification Template and Creating, Modifying, or Deleting a Notification Policy.

If Notified By is set to Contact groups or Topic subscriptions and a default template is selected, SMN sends alarms in a fixed format. JSON-formatted alarms are unavailable. Table 6 lists related alarm fields.

  • The alarm fields in the following table are common fields. If there are special fields, submit a service ticket to contact technical support.
  • AlarmNotificationType, LastAlarmLevel, IngressMaxBandwidthPerSec, EgressMaxBandwidthPerSec, RelatedResourceName, IsCycleTrigger, RecommendedSolution, ImpactScope, DataPointTime, and ResourceDimensionInfos are available only in some regions. They will be available in other regions later. For details, see the notification you receive.
Table 6 Alarm fields

Field

Description

message_type

The value is alarm.

alarm_id

Alarm rule ID.

alarm_name

Alarm rule name.

alarm_status

Alarm status.

Range:

  • ok: The alarm status is normal.
  • alarm: An alarm is generated.

time

Time when an alarm was generated.

namespace

Service namespace.

The format is service.item. The values of service and item must be character strings and start with a letter.

metric_name

Metric name.

dimension
  • name: monitoring dimension. For example, the ECS dimension is instance_id. name must start with a letter and contain only letters, digits, underscores (_), and hyphens (-). name can be 1 to 32 characters.
  • value: dimension value, for example, an ECS ID. Start with a letter or a digit. Enter 1 to 256 characters. Only digits, letters, underscores (_), and hyphens (-) are allowed.

period

Aggregation granularity of the monitoring data. For details about aggregation, see What Is Aggregation?

Range:

  • 1: Cloud Eye performs no aggregation and displays raw data.
  • 300: Cloud Eye aggregates data every 5 minutes.
  • 1200: Cloud Eye aggregates data every 20 minutes.
  • 3600: Cloud Eye aggregates data every 1 hour.
  • 14400: Cloud Eye aggregates data every 4 hours.
  • 86400: Cloud Eye aggregates data every 24 hours.

filter

Data aggregation method. Range:

  • average: Cloud Eye calculates the average value of metric data within an aggregation period.
  • max: Cloud Eye calculates the maximum value of metric data within an aggregation period.
  • min: Cloud Eye calculates the minimum value of metric data within an aggregation period.
  • sum: Cloud Eye calculates the sum of metric data within an aggregation period.
  • variance: Cloud Eye calculates the variance value of metric data within an aggregation period.

comparison_operator

Operator of alarm thresholds. Range: >, =, <, >=, and <=.

value

Alarm threshold. Supported range: 0 to Number.MAX_VALUE (1.7976931348623157e+108)

unit

Data unit. Enter up to 32 characters.

count

Number of consecutive times that the alarm policy was met. Supported range: 1 to 5

alarmValue
  • time: Data point time.
  • value: Value corresponding to time, which ranges from 0 to Number.MAX_VALUE.

sms_content

SMS message content.

template_variable

Variable required by the template. For details, see Table 7.
Table 7 Descriptions of template_variable parameters

Parameter

Description

AccountName

Account name.

Namespace

Service name.

DimensionName

Dimension name.

MetricName

Metric name.

IsAlarm

Whether an alarm was generated. Range:

  • true
  • false

IsCycleTrigger

Whether to continuously trigger alarms. Range:

  • true
  • false

AlarmLevel

Alarm severity. Range: Critical, Major, Minor, and Warning.

Region

Region name.

ResourceId

Resource ID.

CurrentData

Current metric value.

AlarmTime

Time when an alarm was generated.

DataPoint

Metric value when an alarm is triggered.

AlarmRuleName

Alarm rule name.

AlarmId

Alarm rule ID.

AlarmDesc

Alarm description.

MonitoringRange

Monitoring scope.

Range:

  • Specific resources
  • Resource groups
  • All resources

Filter

Data aggregation method. Range:

  • average: Cloud Eye calculates the average value of metric data within an aggregation period.
  • max: Cloud Eye calculates the maximum value of metric data within an aggregation period.
  • min: Cloud Eye calculates the minimum value of metric data within an aggregation period.
  • sum: Cloud Eye calculates the sum of metric data within an aggregation period.
  • variance: Cloud Eye calculates the variance value of metric data within an aggregation period.

ComparisonOperator

Operator of alarm thresholds. Range: >, =, <, >=, and <=.

Value

Alarm threshold. Supported range: 0 to Number.MAX_VALUE (1.7976931348623157e+108)

Unit

Data unit. Enter up to 32 characters.

Count

Number of consecutive times that the alarm policy was met. Supported range: 1 to 5

EventContent

Additional information about event monitoring.

AlarmNotificationType

Alarm notification type.

Range:

  • Metric
  • Event

LastAlarmLevel

Severity of the last alarm. Range: Critical, Major, Minor, and Warning.

When an alarm policy has several thresholds and the metric data triggers them, the alarm severity updates and this parameter appears in the notification.

IngressMaxBandwidthPerSec

Inbound peak bandwidth.

This parameter is supported only by some event alarms. If it is not supported, it is left empty in the notification.

EgressMaxBandwidthPerSec

Outbound peak bandwidth.

This parameter is supported only by some event alarms. If it is not supported, it is left empty in the notification.

RelatedResourceName

Related resource.

This parameter is supported only by some event alarms. If it is not supported, it is left empty in the notification.

IsCycleTrigger

Trigger condition. For event alarms, the value can be:

true: cumulative.

false: instant.

For metric or WAN quality alarms, the value is false.

RecommendedSolution

Recommended solution.

This parameter is supported only by some event alarms. If it is not supported, it is left empty in the notification.

ImpactScope

Scope of impact.

This parameter is supported only by some event alarms. If it is not supported, it is left empty in the notification.

DataPointTime

Time when the event occurred.

This parameter has a value only for event alarms.

ResourceDimensionInfos

Dimension information.

It is available for metric and WAN quality alarms.
Parent Topic: Alarm Rules