CloudOps Orchestration Service:ACS-ECS-RepairImage

{
    "Version": "1",
    "Statement": [
        {
            "Action": [
                "ecs:CreateImage",
                "ecs:CreateSnapshot",
                "ecs:DeleteInstance",
                "ecs:DeleteSnapshot",
                "ecs:DescribeDisks",
                "ecs:DescribeImages",
                "ecs:DescribeInstances",
                "ecs:DescribeInvocationResults",
                "ecs:DescribeInvocations",
                "ecs:DescribeSnapshots",
                "ecs:InvokeCommand",
                "ecs:RebootInstance",
                "ecs:RunCommand",
                "ecs:RunInstances"
            ],
            "Resource": "*",
            "Effect": "Allow"
        }
    ]
}

FormatVersion: OOS-2019-06-01
Description:
  zh-cn: This template repairs an image offline based on image check results. For more information about image checks, see https://www.alibabacloud.com/help/zh/ecs/user-guide/overview-18. The repair process first creates a repair instance from an Alibaba Cloud Linux image. It then creates a disk from a snapshot of your image and attaches the disk to the repair instance. After the data disk is repaired offline, a new image is created from the repaired data disk. Limitation: If the source image is a public image, such as Ubuntu 22.04 64-bit, the repaired image is a custom image, such as ubuntu_64-bit.
  en: This template repairs an image offline based on image check results. For more information about image checks, see https://www.alibabacloud.com/help/en/ecs/user-guide/overview-18. The repair process creates a repair instance from an Alibaba Cloud Linux image. It then creates a disk from a snapshot of your image and attaches it to the repair instance. After the data disk is repaired offline, a new image is created from the repaired data disk.
  name-en: ACS-ECS-RepairImage
  name-zh-cn: Repair Image
  categories:
    - image_manage
Parameters:
  regionId:
    Type: String
    Label:
      en: Region ID
      zh-cn: Region ID
    AssociationProperty: RegionId
    Default: '{{ ACS::RegionId }}'
  zoneId:
    Type: String
    Label:
      en:  vSwitch Zone
      zh-cn: vSwitch Zone
    AssociationProperty: ALIYUN::ECS::ZoneId
    AssociationPropertyMetadata:
      RegionId: regionId
  repairImageId:
    Type: String
    Label:
      en: ID of Image to Repair
      zh-cn: ID of Image to Repair
    AssociationProperty: ALIYUN::ECS::Image::ImageId
    AssociationPropertyMetadata:
      RegionId: regionId
      SupportedImageOwnerAlias:
        - self
      OSType: linux
  baseImageId:
    Type: String
    Label:
      en: Base Image
      zh-cn: Base Image
    Description:
      en: '<font color=''red''>The base image is used to create the repair instance. Select a base image that has the same architecture as the target image. For example, if the custom image to repair is an x86_64 image, select an aliyun_3_x64 image.</font>'
      zh-cn: <font color='red'>The base image is used to create the repair instance. Select a base image that has the same architecture as the target image. For example, if the custom image to repair is an x86_64 image, select an aliyun_3_x64 image.</font>
    AllowedValues:
      - aliyun_3_x64_20G_alibase_20250629.vhd
      - aliyun_3_arm64_20G_alibase_20250629.vhd
  instanceType:
    Label:
      en: Instance Type
      zh-cn: Instance Type
    Type: String
    AssociationProperty: ALIYUN::ECS::Instance::InstanceType
    AssociationPropertyMetadata:
      RegionId: regionId
      ZoneId: zoneId
  securityGroupId:
    Label:
      en: Security Group ID
      zh-cn: Security Group ID
    Type: String
    AssociationProperty: ALIYUN::ECS::SecurityGroup::SecurityGroupId
    AssociationPropertyMetadata:
      RegionId: regionId
  vSwitchId:
    Label:
      en: vSwitch ID
      zh-cn: vSwitch ID
    Type: String
    AssociationProperty: ALIYUN::VPC::VSwitch::VSwitchId
    AssociationPropertyMetadata:
      RegionId: regionId
      ZoneId: zoneId
      Filters:
        - SecurityGroupId: securityGroupId
  systemDiskCategory:
    Label:
      en: System Disk Category
      zh-cn: System Disk Category
    Type: String
    AssociationProperty: ALIYUN::ECS::Disk::SystemDiskCategory
    AssociationPropertyMetadata:
      RegionId: regionId
      InstanceType: instanceType
    Default: cloud_essd
  repairItems:
    Label:
      en: Items to Repair
      zh-cn: Items to Repair
    Type: List
    AllowedValues:
      - all
      - assist
      - aegis
      - growpart
      - virtio
      - selinux
      - dhcp
      - grub
      - fstab
      - nvme
      - cloudinit
    Default:
      - assist
      - cloudinit
      - virtio
      - nvme
      - fstab
      - grub
      - dhcp
      - growpart
Tasks:
  - Name: checkImage
    Action: ACS::CheckFor
    Description:
      en: Checks if the image can be repaired.
      zh-cn: Checks if the image can be repaired.
    Properties:
      Service: ECS
      API: DescribeImages
      PropertySelector: Images.Image[0].DetectionOptions.Items.Item[]? | select(.Name == "SystemImage") | .Value
      DesiredValues: []
      Parameters:
        RegionId: '{{ regionId }}'
        ImageId: '{{ repairImageId }}'
      NotDesiredValues:
        - Invalid
    Outputs:
      architecture:
        Type: String
        ValueSelector: Images.Image[0].Architecture
      snapshotId:
        Type: String
        ValueSelector: Images.Image[0].DiskDeviceMappings.DiskDeviceMapping[0].SnapshotId
      imageName:
        Type: String
        ValueSelector: Images.Image[0].ImageName
      osType:
        Type: String
        ValueSelector: Images.Image[0].OSType
      platform:
        Type: String
        ValueSelector: Images.Image[0].Platform
      bootMode:
        Type: String
        ValueSelector: Images.Image[0].BootMode
  - Name: checkImageArchitecture
    Action: ACS::Choice
    Description:
      en: Checks if the image architecture is x86_64 or arm64.
      zh-cn: Checks if the image architecture is x86_64 or arm64.
    Properties:
      DefaultTask: ACS::END
      Choices:
        - When:
            Fn::Or:
              - Fn::Equals:
                  - x86_64
                  - '{{checkImage.architecture}}'
              - Fn::Equals:
                  - arm64
                  - '{{checkImage.architecture}}'
          NextTask: runInstances
  - Name: runInstances
    Action: ACS::ExecuteAPI
    Description:
      en: Creates an ECS instance.
      zh-cn: Creates an ECS instance.
    Properties:
      Service: ECS
      API: RunInstances
      Parameters:
        RegionId: '{{ regionId }}'
        Amount: 1
        ImageId: '{{ baseImageId }}'
        InstanceType: '{{ instanceType }}'
        InstanceName: image_repair_please_no_delete
        SecurityGroupId: '{{ securityGroupId }}'
        VSwitchId: '{{ vSwitchId }}'
        InternetMaxBandwidthIn: 200
        InternetMaxBandwidthOut: 0
        SystemDiskCategory: '{{ systemDiskCategory }}'
        DataDisk:
          - SnapshotId: '{{ checkImage.snapshotId }}'
            DiskName: image_repair_{{ checkImage.snapshotId }}
            Category: '{{ systemDiskCategory }}'
        InstanceChargeType: PostPaid
    Outputs:
      instanceId:
        Type: String
        ValueSelector: InstanceIdSets.InstanceIdSet[0]
  - Name: untilInstanceReady
    Action: ACS::WaitFor
    Description:
      en: Waits for the instance to enter the Running state.
      zh-cn: Waits for the instance to enter the Running state.
    Properties:
      Service: ECS
      API: DescribeInstances
      PropertySelector: Instances.Instance[0].Status
      DesiredValues:
        - Running
      NotDesiredValues: []
      StopRetryValues: []
      Parameters:
        RegionId: '{{ regionId }}'
        InstanceIds:
          - '{{ runInstances.instanceId }}'
    Retries: 10
    DelayType: Exponential
    Delay: 2
    BackOff: 2
  - Name: checkForCreatedDiskReady
    Action: ACS::CheckFor
    Description:
      en: Checks if the disk is in the In_use state.
      zh-cn: Checks if the disk is in the In_use state.
    OnError: stopInstances
    Properties:
      Service: ECS
      API: DescribeDisks
      Parameters:
        RegionId: '{{ regionId }}'
        InstanceId: '{{ runInstances.instanceId }}'
      DesiredValues:
        - In_use
      PropertySelector: Disks.Disk[0].Status
    Outputs:
      diskSN:
        Type: String
        ValueSelector: Disks.Disk[0].SerialNumber
      diskId:
        Type: String
        ValueSelector: Disks.Disk[0].DiskId
  - Name: runCommandOpenselinux
    Action: ACS::ECS::RunCommand
    Description:
      en: Runs a command on the instance.
      zh-cn: Runs a command on the instance.
    OnError: stopInstances
    Properties:
      regionId: '{{ regionId }}'
      commandContent: sed -i s/^SELINUX=.*/SELINUX=permissive/g /etc/selinux/config
      commandType: RunShellScript
      instanceId: '{{ runInstances.instanceId }}'
    Outputs:
      invocationOutput:
        Type: String
        ValueSelector: invocationOutput
  - Name: rebootInstance
    Action: ACS::ECS::RebootInstance
    Description:
      en: Reboots the instance.
      zh-cn: Reboots the instance.
    OnError: stopInstances
    Properties:
      regionId: '{{ regionId }}'
      instanceId: '{{ runInstances.instanceId }}'
  - Name: runCommand
    Action: ACS::ECS::InvokeCommand
    Description:
      en: Runs a command on the instance.
      zh-cn: Runs a command on the instance.
    OnError: stopInstances
    Properties:
      regionId: '{{ regionId }}'
      commandId: ACS-ECS-ImageRepair-for-linux.sh
      parameters:
        Fn::Join:
          - ''
          - - '{"repairitems":'
            - '"'
            - Fn::Join:
                - ' '
                - '{{repairItems}}'
            - '"'
            - ',"disk_serial":"'
            - '{{checkForCreatedDiskReady.diskSN}}'
            - '"}'
      instanceId: '{{ runInstances.instanceId }}'
    Outputs:
      invocationOutput:
        Type: String
        ValueSelector: invocationOutput
  - Name: waitDiskFlush
    Action: ACS::Sleep
    Description:
      en: Waits for I/O operations to be written to the disk.
      zh-cn: Waits for I/O operations to be written to the disk.
    Properties:
      Duration: 1M
  - Name: createSnapshot
    Action: ACS::ECS::CreateSnapshot
    Description:
      en: Creates a snapshot for the disk.
      zh-cn: Creates a snapshot for the disk.
    OnError: deleteInstance
    Properties:
      regionId: '{{ regionId }}'
      snapshotName: image_repair_{{ checkForCreatedDiskReady.diskId }}
      diskId: '{{ checkForCreatedDiskReady.diskId }}'
      retentionDays: 1
    Outputs:
      snapshotId:
        Type: String
        ValueSelector: snapshotId
  - Name: createImage
    Action: ACS::ExecuteAPI
    Description:
      en: Creates an image from the snapshot.
      zh-cn: Creates an image from the snapshot.
    OnError: deleteSnapshot
    OnSuccess: deleteInstance
    Properties:
      Service: ECS
      API: CreateImage
      Parameters:
        RegionId: '{{ regionId }}'
        SnapshotId: '{{ createSnapshot.snapshotId }}'
        ImageName:
          Fn::Join:
            - _
            - - repaired
              - '{{ checkImage.imageName }}'
              - Fn::FormatUTCTime:
                  - '{{ACS::CurrentUTCTime}}'
                  - '%Y%m%d%H%M%S'
        DetectionStrategy: Standard
        Platform:
          Fn::Select:
            - '{{ checkImage.platform }}'
            - Aliyun: Aliyun
              Anolis: Anolis
              CentOS: CentOS
              Ubuntu: Ubuntu
              SUSE: SUSE
              Debian: Debian
              OpenSUSE: OpenSUSE
              Red Hat: RedHat
              Kylin: Kylin
              UOS: UOS
              Fedora: Fedora
              CentOS Stream: CentOS Stream
              AlmaLinux: AlmaLinux
              Rocky Linux: Rocky Linux
              Customized Linux: Customized Linux
              Others Linux: Others Linux
        BootMode: '{{ checkImage.bootMode }}'
        Architecture: '{{ checkImage.architecture }}'
    Outputs:
      imageId:
        Type: String
        ValueSelector: ImageId
  - Name: deleteSnapshot
    Action: ACS::ExecuteAPI
    Description:
      en: Releases the snapshot.
      zh-cn: Releases the snapshot.
    OnError: deleteInstance
    Properties:
      Service: ECS
      API: DeleteSnapshot
      Parameters:
        RegionId: '{{ regionId }}'
        SnapshotId: '{{ createSnapshot.snapshotId }}'
        Force: true
  - Name: stopInstances
    Action: ACS::Sleep
    Description:
      en: Waits for the instance protection to expire.
      zh-cn: Waits for the instance protection to expire.
    Properties:
      Duration: 1M
  - Name: deleteInstance
    Action: ACS::ExecuteAPI
    Description:
      en: Releases the ECS instance.
      zh-cn: Releases the ECS instance.
    Properties:
      Service: ECS
      API: DeleteInstance
      Parameters:
        RegionId: '{{ regionId }}'
        InstanceId: '{{ runInstances.instanceId }}'
        Force: true
Outputs:
  imageId:
    Type: String
    Value: '{{ createImage.imageId }}'
Metadata:
  ALIYUN::OOS::Interface:
    ParameterGroups:
      - Parameters:
          - regionId
          - repairImageId
          - baseImageId
        Label:
          default:
            zh-cn: Select Image
            en: Select Image
      - Parameters:
          - zoneId
          - instanceType
          - securityGroupId
          - vSwitchId
          - systemDiskCategory
        Label:
          default:
            zh-cn: Instance Configurations
            en: Instance Configurations
      - Parameters:
          - repairItems
        Label:
          default:
            zh-cn: Run Command
            en: Run Command