The Deckhouse Platform installs CRDs but does not remove them when a module is disabled. If you no longer need the created CRDs, delete them.

The module lifecycle stage: General Availability

YandexInstanceClass

Scope: Cluster

  • v1
  • v1alpha1
  • metadata
    object
  • spec
    object

    Required value

    YandexInstanceClassSpec defines the desired state of the YandexInstanceClass.

    • spec.additionalLabels
      object

      Additional labels to be attached to the Yandex Compute Instance resource.

      Example:

      additionalLabels: "{project: cms-production, severity: critical}"
      
    • spec.additionalSubnets
      array of strings

      Subnet IDs that VirtualMachines’ secondary NICs will connect to.

      For CloudEphemeral nodes, every subnet in the list is attached as a separate network interface.

      For CloudPermanent nodes, a single subnet is selected from the list by the node index, so the list must contain at least as many subnets as there are nodes in the group.

      Example:

      additionalSubnets: "[b0csh41c1or82vuch89v, e2lgddi5svochh5fbq96]"
      
    • spec.assignPublicIPAddress
      boolean

      Should a public external IPv4 address be assigned to a provisioned Yandex Compute Instance.

      Default: false

      Example:

      assignPublicIPAddress: 'false'
      
    • spec.coreFraction
      integer

      Percent of reserved CPU capacity on a Yandex Compute Instance.

      More about performance levels.

      Default: 100

      Allowed values: 5, 20, 50, 100

      Example:

      coreFraction: '20'
      
    • spec.cores
      integer

      Required value

      Amount of CPU cores to provision on a Yandex Compute Instance.

      Example:

      cores: '4'
      
    • spec.diskSizeGB
      integer

      Yandex Compute Instance disk size in gibibytes.

      Default: 50

      Example:

      diskSizeGB: '20'
      
    • spec.diskType
      string

      Instance disk type.

      Size of network-ssd-nonreplicated and network-ssd-io-m3 disks must be a multiple of 93 GB.

      Default: "network-hdd"

      Allowed values: network-ssd, network-hdd, network-ssd-nonreplicated, network-ssd-io-m3

      Example:

      diskType: network-hdd
      
    • spec.etcdDiskSizeGB
      integer

      Disk size for etcd is set only for InstanceClass intended for master NodeGroup. The value is specified in GiB.

      Example:

      etcdDiskSizeGB: '20'
      
    • spec.gpus
      integer

      Number of GPUs on a Yandex Compute Instance.

      Default: 0

      Example:

      gpus: '4'
      
    • spec.imageID
      string

      Image ID to use while provisioning Yandex Compute Instances.

      If the parameter is not set, the image of the instance class of the master node group is used. In a cluster that has not been migrated to the ModuleConfig settings, the masterNodeGroup.instanceClass.imageID parameter of the YandexClusterConfiguration resource is used instead.

      Example:

      imageID: fd83ica41cade1mj35sr
      
    • spec.mainSubnet
      string

      Subnet ID that VirtualMachines’ primary NIC will connect to.

      If the parameter is not specified, Deckhouse Platform selects the network. When the existingZoneToSubnetIDMap parameter is set, the network is taken from it, otherwise the network created during the cluster deployment is used.

      Example:

      mainSubnet: e9bnc7g9mu9mper9clk4
      
    • spec.memory
      integer

      Required value

      Amount of primary memory in MB provision on a Yandex Compute Instance.

      Example:

      memory: '8192'
      
    • spec.networkType
      string

      Network type: Standard or SoftwareAccelerated.

      Default: Standard

      Allowed values: Standard, SoftwareAccelerated

    • spec.platformID
      string

      Default: standard-v3

    • spec.preemptible
      boolean

      Should a provisioned Yandex Compute Instance be preemptible.

      For more information about preemptible virtual machines, read the provider’s documentation.

      Default: false

  • status
    object

    YandexInstanceClassStatus stores information about the YandexInstanceClass resource.

    • status.nodeGroupConsumers
      array of strings

      NodeGroupConsumers lists the names of NodeGroups that use this instance class.

  • metadata
    object
  • spec
    object

    Required value

    InstanceClassSpec defines the desired state of the YandexInstanceClass.

    • spec.additionalLabels
      object

      Additional labels to be attached to the Yandex Compute Instance resource.

      Example:

      additionalLabels: "{project: cms-production, severity: critical}"
      
    • spec.additionalSubnets
      array of strings

      Subnet IDs that VirtualMachines’ secondary NICs will connect to.

      For CloudEphemeral nodes, every subnet in the list is attached as a separate network interface.

      For CloudPermanent nodes, a single subnet is selected from the list by the node index, so the list must contain at least as many subnets as there are nodes in the group.

      Example:

      additionalSubnets: "[b0csh41c1or82vuch89v, e2lgddi5svochh5fbq96]"
      
    • spec.assignPublicIPAddress
      boolean

      Should a public external IPv4 address be assigned to a provisioned Yandex Compute Instance.

      Default: false

      Example:

      assignPublicIPAddress: 'false'
      
    • spec.coreFraction
      integer

      Percent of reserved CPU capacity on a Yandex Compute Instance.

      More about performance levels.

      Default: 100

      Allowed values: 5, 20, 50, 100

      Example:

      coreFraction: '20'
      
    • spec.cores
      integer

      Required value

      Amount of CPU cores to provision on a Yandex Compute Instance.

      Example:

      cores: '4'
      
    • spec.diskSizeGB
      integer

      Yandex Compute Instance disk size in gibibytes.

      Default: 20

      Example:

      diskSizeGB: '50'
      
    • spec.diskType
      string

      Instance disk type.

      Size of network-ssd-nonreplicated and network-ssd-io-m3 disks must be a multiple of 93 GB.

      For more information about possible disk types, read the provider’s documentation.

      Default: "network-hdd"

      Allowed values: network-ssd, network-hdd, network-ssd-nonreplicated

      Example:

      diskType: network-hdd
      
    • spec.gpus
      integer

      Number of GPUs on a Yandex Compute Instance.

      Default: 0

      Example:

      gpus: '4'
      
    • spec.imageID
      string

      Image ID to use while provisioning Yandex Compute Instances.

      If the parameter is not set, the image of the instance class of the master node group is used. In a cluster that has not been migrated to the ModuleConfig settings, the masterNodeGroup.instanceClass.imageID parameter of the YandexClusterConfiguration resource is used instead.

      Example:

      imageID: fd83ica41cade1mj35sr
      
    • spec.mainSubnet
      string

      Subnet ID that VirtualMachines’ primary NIC will connect to.

      If the parameter is not specified, Deckhouse Platform selects the network. When the existingZoneToSubnetIDMap parameter is set, the network is taken from it, otherwise the network created during the cluster deployment is used.

      Example:

      mainSubnet: e9bnc7g9mu9mper9clk4
      
    • spec.memory
      integer

      Required value

      Amount of primary memory in MB provision on a Yandex Compute Instance.

      Example:

      memory: '8192'
      
    • spec.networkType
      string

      Network type: STANDARD or SOFTWARE_ACCELERATED

      Default: STANDARD

      Allowed values: STANDARD, SOFTWARE_ACCELERATED

    • spec.platformID
      string

      Default: standard-v2

    • spec.preemptible
      boolean

      Should a provisioned Yandex Compute Instance be preemptible.

      For more information about preemptible virtual machines, read the provider’s documentation.

      Default: false