Skip to content

Latest commit

 

History

History
119 lines (110 loc) · 4.85 KB

README.MD

File metadata and controls

119 lines (110 loc) · 4.85 KB
Raw

CDP Prerequiste module.

A terraform module to create Cloudera CDP Public Cloud prerequisite resources.

Usuage:

Azure

module "cdp-prerequisite" {
  source = "${Path to this repo}//azure?ref=main"
  create_resource_group = false                         # Optional. Default to true.
  resource_group_name = "prerequisite-test"             # Mandatory. 
  managed_id = {                                        # Mandatory
    assumer    = "cdp-assumer"
    dataaccess = "cdp-dataaccess"
    logger     = "cdp-logger"
    ranger     = "cdp-ranger"
  }

  tags = {                                             # Optional. Default to null
    owner = "cdp user"
    usecase = "testing"
  }

  obj_storage = {                                      # Mandatory
    storage_account_name   = "cdpstorage"
    data_container_name    = "data"
    logs_container_name    = "logs"
    backups_container_name = "backups"
  }
  file_storage = {                                     # Optional. For ML only.
    storage_account_name = "cdpfilestorage"
    file_share_name      = "ml-fileshare"
    subnet_ids           = [
      data.azurerm_subnet.default.id,
      data.azurerm_subnet.snet1.id
    ]
  }

  cmk = {                                             # Optional
    kv_name = "mycdpkeyvault"
    key_name = "cdpkey"
    spn_object_id = "a37a577a-4502-4c7d-b6d7-8c717ee6e561"
  }
  raz_mi_name = "my-cdp-raz"                         # optional
  dw_mi       = {                                     # Optional
    managed_identiy_name = "cdp-dw-identity"
    custom_role_name     = "CDP DW Role"
  }
}

AWS

Variables

  • region: Optional. AWS region to deploy the resources. Default to us-west-2
  • cdp_bucket_name: Mandatory. AWS bucket name.
  • folders: Optional. The folders in the AWS bucket for CDP to use. Default to data, logs, backups.
  • ssh_key: Optional. The public key will be used to create an SSH public key in the AWS account. If not provided, the ssh key will not be created.
  • cross_account_role: Mandatory. The name of the cross account role and whether it is an existing role or need to be created.
    • if the create_role attribute is false, a role with this name must exist. Prerequisite resources will create necessary trust relationship with this role.
    • if the create_role attribute is true, a role with this name will be created, and proper permission will be granted.
  • aws_sso_user_arn_keyword: Optional. This is a key word to create a trust relationship for the cross account role to the administrator. So that the administrator can assume the cross account role and conduct administrative tasks on the provisioned CDP resources like EKS. The default value is a random string to avoid everyone being able to assume the role.
  • cmk: Optional. This is to indicate whether to use KMS in the provisioning. If null, no CMK will be used. If create_key is true, a key will be created, and proper key policy will be attached to the key_alias. If create_key is false, proper key policy will be attached to the key with the key_alias.
  • use_raz: not yet supported. default to null.
  • tags: Optional. tags to be attached to the new create resources
  • role_names: Optional. The names of the custom roles for CDP to conduct its job. Default to
    idbroker       = "CDP_IDBROKER"
    logger         = "CDP_LOG_ROLE"
    ranger         = "CDP_RANGER_AUDIT"
    datalake_admin = "CDP_DATALAKE_ADMIN"
  • policy_names: Optional. The names of the policies to be used in CDP provisioning. Default to:
    cross_account_policy        = "cdp-cross-account-policy"
    ec2-kms-policy              = "cdp-ec2-kms-policy"
    sse-kms-read-only-policy    = "cdp-sse-kms-read-only-policy"
    sse-kms-read-write-policy   = "cdp-sse-kms-read-write-policy"
    idbroker-assume-role-policy = "cdp-idbroker-assume-role-policy"
    log-policy                  = "cdp-log-policy"
    datalake-restore-policy     = "cdp-datalake-restore-policy"
    backup-policy               = "cdp-backup-policy"
    ranger-audit-s3-policy      = "cdp-ranger-audit-s3-policy"
    bucket-access-policy        = "cdp-bucket-access-policy"
    datalake-backup-policy      = "cdp-datalake-backup-policy"
    datalake-admin-s3-policy    = "cdp-datalake-admin-s3-policy"
  • instance_profile_names: Optional. The name of the instance profiles to be used in CDP provisioning. Default to
    data_access = "cdp-data-access-instance-profile"
    log_access  = "cdp-log-access-instance-profile"

Example

module "prerequisite" {
  source = "${Path to this repo}//aws?ref=main"
  cross_account_role = {
    name        = "cdp-poc-role"
    create_role = false
  }
  cdp_bucket_name = "cdp-poc-bucket"
  ssh_key = {
    name = "my-ssh-key"
    key = file("~/.ssh/id_rsa.pub")
  }
  cmk = {
    key_alias = "my-cdp-key"
    create_key = true
  }
  tags = {
    owner = "My Name"
  }
}

output "prerequisite" {
  value = module.prerequisite
}