oos-timing-management-of-ecs.yml

ROSTemplateFormatVersion: '2015-09-01'
Description:
  zh-cn: 该模板用于在选定ECS实例上定时执行云助手命令，展示CPU使用率最高的10个进程，通过OOS自动运维实现，并配置资源清理。
  en: This template is designed to schedule the execution of Cloud Assistant commands
    on selected ECS instances, showcasing the top 10 processes with the highest CPU
    utilization. It leverages OOS (Operation Orchestration Service) for automated
    operations and includes configuration for resource cleanup.
Parameters:
  EcsInstanceIds:
    Type: CommaDelimitedList
    Label:
      en: Select the ecs instance to execute the cloud assistant command(Used to view
        the 10 processes with the highest CPU usage on the instance)
      zh-cn: 选择执行云助手命令的ecs实例(用于查看实例上CPU占用最高的10个进程)
    Description:
      en: Please select running ECS instances.
      zh-cn: 请选择运行中状态的ECS实例。
    AssociationProperty: ALIYUN::ECS::Instance::InstanceId
  ExecuteTime:
    Type: String
    Label:
      en: Timing type
      zh-cn: 定时类型
    AssociationProperty: ALIYUN::OOS::Component::TimerTrigger
Resources:
  OosRole:
    Type: ALIYUN::RAM::Role
    Properties:
      RoleName:
        Fn::Sub: OosRole-${ALIYUN::StackId}
      Policies:
      - PolicyName:
          Fn::Sub: OosPolicy-${ALIYUN::StackId}
        PolicyDocument:
          Version: '1'
          Statement:
          - Action:
            - ecs:RunCommand
            - ecs:DescribeInstances
            - ecs:DescribeInvocationResults
            - ecs:DescribeInvocations
            Resource:
            - '*'
            Effect: Allow
      AssumeRolePolicyDocument:
        Version: '1'
        Statement:
        - Action: sts:AssumeRole
          Effect: Allow
          Principal:
            RAM:
            - Fn::Sub: acs:ram::${ALIYUN::TenantId}:root
            Service:
            - oos.aliyuncs.com
  Execution:
    Type: ALIYUN::OOS::Execution
    Properties:
      Parameters:
        timerTrigger:
          Ref: ExecuteTime
        OOSAssumeRole:
          Fn::GetAtt:
          - OosRole
          - RoleName
        templateParameters:
          username: ''
          rateControl:
            MaxErrors: 0
            Mode: Concurrency
            Concurrency: 10
          regionId:
            Ref: ALIYUN::Region
          workingDir: /root
          commandType: RunShellScript
          windowsPasswordName: ''
          commandContent: "#!/bin/bash\n# Show top processes sorted by cpu usage\n\
            # [ECS] ACS-ECS-CpuTopNProcessList-for-linux.sh - Provided by Alibaba\
            \ Cloud ECS\n# Version: 1.0\n# \n# The Count of Processes Listed: '{{PROCESS_NUM}}'\n\
            TOPN=10\nif [ \"$TOPN\" == \"\" ]\nthen\nTOPN='10'\nfi\nps auxw | head\
            \ -1; ps auxw | grep -v PID | sort -rn -k3 | head -$TOPN"
          enableParameter: false
          timeout: 600
          targets:
            ResourceIds:
              Ref: EcsInstanceIds
            RegionId:
              Ref: ALIYUN::Region
            Type: ResourceIds
        templateName: ACS-ECS-BulkyRunCommand
      TemplateName: ACS-ECS-ScheduleToExecuteTemplate
  ResourceCleaner:
    Type: ALIYUN::ROS::ResourceCleaner
    Properties:
      Action: Scan+CleanUp
      CleanUpTimeout: 120
      Resources:
      - ResourceType: RAM:Role
        ResourceId:
          Fn::GetAtt:
          - OosRole
          - RoleId
        RegionId: center
      - ResourceType: RAM:ManagedPolicy
        ResourceId:
          Fn::Sub: OosPolicy-${ALIYUN::StackId}
        RegionId: center
    DependsOn: Execution
Outputs:
  Execute:
    Description:
      en: The template ACS-ECS-BulkyRunCommand has been used to run Bulky Assistant
        commands on multiple ECS instances in batches. Please go to the OOS console
        to confirm
      zh-cn: 已使用模板ACS-ECS-BulkyRunCommand批量在多台ECS实例上运行云助手命令。请前往OOS控制台，在左侧导航栏选择自动化运维>执行管理进行验证，控制台链接：https://oos.console.aliyun.com/。
    Value: ACS-ECS-BulkyRunCommand
Metadata:
  ALIYUN::ROS::Interface:
    ParameterGroups:
    - Parameters:
      - EcsInstanceIds
      - ExecuteTime
    TemplateTags:
    - acs:document:试用教程:OOS实现ECS的定时管理