Help Center/ Cloud Eye/ Best Practices/ Scenario-specific Cases/ Live Streaming Monitoring/ Server Monitoring
Updated on 2025-12-23 GMT+08:00
View PDF
Share

Server Monitoring

ECSs are the core of live video streaming. Slight changes in ECS performance may greatly affect other cloud services. To monitor more fine-grained metrics, you can install the Agents on ECSs. For details, see Agent Installation and Configuration.

This section describes how to create alarm rules for ECS CPU usage, memory usage, and disk usage. It also includes how to configure an AS policy. When the ECS CPU usage reaches 90% for five consecutive times within 5 minutes, AS automatically adds an ECS to ensure stable service running.

After the alarm rule and AS policy are created, if the service volume soars and the specified threshold is reached, AS automatically adds an ECS and Cloud Eye immediately informs you of the resource exception.

Background

CPU Usage, Memory Usage, and Disk Usage are metrics for basic ECS monitoring. (Agent) CPU Usage, (Agent) Memory Usage, and (Agent) Disk Usage are metrics for fine-grained minute-level monitoring.

Creating an Alarm Rule

  1. Log in to the management console.
  2. In the upper left corner, select a region and a project.
  3. Choose Service List > Cloud Eye.
  4. In the navigation pane, choose Server Monitoring > Elastic Cloud Server.
  5. Locate the target ECS. In the Operation column, choose More > Create Alarm Rule.
  6. On the Create Alarm Rule page, set Method to Configure manually and configure alarm policies for the monitored instance.
    Figure 1 Creating an alarm rule
    Table 1 Alarm policy description

    Parameter

    Example Value

    Description

    Metric Name

    Alarm policy 1: ECSs / (Agent) CPU Usage

    Alarm policy 2: ECSs / (Agent) Memory Usage

    Alarm policy 3: Mount Point / (Agent) Disk Usage

    Metric for triggering alarms.

    Statistic

    Avg. > 1 hour

    Metric value type. Cloud Eye supports the following statistics for metrics: Raw data, Avg., Max., Min., Variance, and Sum.

    Consecutive Triggers

    3 times (consecutively)

    Number of consecutive times that an alarm is triggered.

    Operator

    >=

    Operator used to compare metric value and the threshold.

    Threshold

    Major: 80%; Critical: 90%

    Alarm threshold and unit.

    Frequency

    Daily

    How often alarms are repeatedly notified when there is already an alarm.
  7. Configure alarm notification parameters. For details, see Creating an Alarm Rule and Notifications.
    Table 2 Parameters for configuring alarm notifications

    Parameter

    Example Value

    Description

    Alarm Notifications

    Enabled

    Whether to send an alarm notification when an alarm is triggered.

    Notified By

    Topic subscriptions

    Select an alarm notification mode.

    Recipient

    Account contact
    • If you select Topic subscriptions for Notified By, select one or more notification recipients.
    • If Recipient is set to Account contact, notifications will be sent to the registered phone number and email address.

    Notification Window

    00:00–23:59
    • If Notified By is set to Topic subscriptions, you need to set the notification window.
    • Cloud Eye sends notifications only within the specified time period.

    Trigger Condition

    Generated alarm, Cleared alarm

    This parameter is available when Notified By is set to Topic subscriptions or Notification groups. You can select Generated alarm, Cleared alarm, or both.

Configuring an AS Policy

  1. Choose Computing > Auto Scaling.
  2. Click Create AS Group. For details, see Creating an AS Group.
  3. In the AS group list, locate the created AS group and click View AS Policy in the Operation column.
  4. On the AS Policies tab page, click Add. In the displayed Add AS Policy dialog box, set parameters based on Figure 2.
    Figure 2 Adding an AS policy