Skip to main content

elastic_cluster

note

The resource's API is deprecated and may cause unknown behaviors. It is recommended to deploy a multi-warehouse elastic CelerData cluster using Multi-warehouse Based elastic cluster.

Deploy a single-warehouse elastic CelerData cluster on AWS EC2 instances、Azure virtual machines or GCP virtual machines.

The implementation of this resource is part of the whole cluster deployment procedure and depends on the implementation of a data credential, a deployment credential, and a network configuration. For detailed procedures of cluster deployments on AWS、Azure and GCP, see Provision CelerData Cloud BYOC on AWS and Provision CelerData Cloud BYOC on Azure.

Supported Node Sizes​

For information about the instance types supported by CelerData, see Supported instance types.

Example Usage​

# The celerdatabyoc_elastic_cluster resource API is deprecated
# It is recommended to deploy a multi-warehouse elastic CelerData cluster using resource `celerdatabyoc_elastic_cluster_v2`.

resource "celerdatabyoc_elastic_cluster" "elastic_cluster_1" {
  deployment_credential_id = "<deployment_credential_resource_ID>"
  data_credential_id = "<data_credential_resource_ID>"
  network_id = "<network_configuration_resource_ID>"

  cluster_name = "<cluster_name>"
  coordinator_node_size = "<coordinator_node_instance_type>"
  coordinator_node_count = <coordinator_node_number>
  
  // optional
  coordinator_node_volume_config {
    vol_size = <vol_size>
    iops = <iops>
    throughput = <throughput>
  }
  // optional
  coordinator_node_configs = {
    <key> = <value>
  }

  compute_node_size = "<compute_node_instance_type>"
  compute_node_count = <compute_node_number>
  
  // optional
  compute_node_volume_config {
    vol_number = <vol_number>
    vol_size = <vol_size>
    iops = <iops>
    throughput = <throughput>
  }
  // optional
  compute_node_configs = {
    <key> = <value>
  }

  // optional
  scheduling_policy {
    policy_name = "auto-resume-suspend"
    description = "Auto resume/suspend"
    active_days = ["TUESDAY"]
    time_zone = "UTC" // IANA Time-Zone
    resume_at   = "09:00"
    suspend_at  = "18:00"
    enable      = true
  }

  default_admin_password = "<SQL_user_initial_password>"

  expected_cluster_state = "{Suspended | Running}"
    ldap_ssl_certs = [
    "<ssl_cert_s3_path>"
  ]
  resource_tags = {
    celerdata = "<tag_name>"
  }
  csp = "aws"
  region = "<AWS_VPC_region>"

  init_scripts {
      logs_dir    = "<log_s3_path>"
      script_path = "<script_s3_path>"
  }
  run_scripts_parallel = false
  query_port = 9030
  idle_suspend_interval = 64
}

Argument Reference​

The celerdatabyoc_elastic_cluster resource contains the following required arguments and optional arguments:

Required:

  • cluster_name: (Not allowed to modify) The desired name for the cluster. Enter a unique name.

  • coordinator_node_size: The instance type for coordinator nodes in the cluster. Select a coordinator node instance type from the table "Supported Node Sizes". For example, you can set this argument to m6i.4xlarge.

  • deployment_credential_id: (Not allowed to modify) Set the value to celerdatabyoc_aws_deployment_role_credential.deployment_role_credential.id.

  • data_credential_id: (Not allowed to modify) Set the value to celerdatabyoc_aws_data_credential.data_credential.id.

  • network_id: (Not allowed to modify) Set the value to celerdatabyoc_aws_network.network.id.

  • compute_node_size: The instance type for compute nodes in the cluster. Select a compute node instance type from the table "Supported Node Sizes". For example, you can set this argument to r6id.4xlarge.

  • default_admin_password: (Not allowed to modify) The initial password of the cluster admin user.

  • expected_cluster_state: When creating a cluster, you need to declare the status of the cluster you are creating. Cluster states are categorized as Suspended and Running. If you want the cluster to start after provisioning, set this argument to Running. If you do not do so, the cluster will be suspended after provisioning.

  • csp: (Not allowed to modify) The cloud service provider of the cluster. Only AWS is supported.

  • region: (Not allowed to modify) The ID of the cloud provider region to which the network hosting the cluster belongs. See Supported cloud platforms and regions.

Optional:

  • coordinator_node_count: The number of coordinator nodes in the cluster. Valid values: 1, 3, and 5. Default value: 1.

  • coordinator_node_volume_config: The coordinator nodes volume configuration.

    • vol_size: The size per disk for each coordinator node. Unit: GB. Default value: 150. You can only increase the value of this parameter.
    • iops: Disk iops.
    • throughput: Disk throughput.

  • coordinator_node_configs: The coordinator node static configuration.

  • compute_node_count: The number of compute nodes in the cluster. Valid values: any non-zero positive integer. Default value: 3.

  • compute_node_volume_config: The compute nodes volume configuration.

    • vol_number: (Not allowed to modify) The number of disks for each compute node. Valid values: [1,24]. Default value: 2.
    • vol_size: The size per disk for each compute node. Unit: GB. Default value: 100. You can only increase the value of this parameter.
    • iops: Disk iops.
    • throughput: Disk throughput. ~> You can use the vol_number and vol_size arguments to specify the disk space. The total disk space provisioned to a compute node is equal to vol_number * vol_size.

  • compute_node_configs: The compute node static configuration.

  • global_session_variables: Global session variables of the cluster. You can find all configurable variables by select VARIABLE_NAME from information_schema.global_variables;.

  • ldap_ssl_certs: The path in the AWS S3 bucket that stores the LDAP SSL certificates. Multiple paths must be separated by commas (,). CelerData supports using LDAP over SSL by uploading the LDAP SSL certificates from S3. To allow CelerData to successfully fetch the certificates, you must grant the ListObject and GetObject permissions to CelerData. To delete the certificates uploaded, you only need to remove this argument.

  • ranger_certs_dir: The parent dir path in the AWS S3 bucket that stores the Ranger SSL certificates. CelerData supports using Ranger over SSL by uploading the Ranger SSL certificates from S3. To allow CelerData to successfully fetch the certificates, you must grant the ListObject and GetObject permissions to CelerData. To delete the certificates uploaded, you only need to remove this argument.

note

You can only upload or delete LDAP or Ranger SSL certificates while the cluster's expected_cluster_state is set to Running.

  • resource_tags: The tags to be attached to the cluster (Please note that resource_tags is a concept in ClelerData. For AWS and Azure, it will be added as a tag to the corresponding resources. For GCP Cloud, it will be added as a label to the corresponding GCP resources).

  • init_scripts: The configuration block to specify the paths to which scripts and script execution results are stored. The maximum number of executable scripts is 20. For information about the formats supported by these arguments, see scripts.logs_dir and scripts.script_path in Run scripts.

    • logs_dir: The path in the AWS S3 bucket to which script execution results are stored. This S3 bucket can be the same as or different from the S3 bucket you specify in the celerdatabyoc_aws_data_credential resource.
    • script_path: The path in the AWS S3 bucket that stores the scripts to run via Terraform. This S3 bucket must be the one you specify in the celerdatabyoc_aws_data_credential resource.

  • run_scripts_parallel: Whether to execute the scripts in parallel. Valid values: true and false. Default value: false.

  • run_scripts_timeout: The amount of time after which the script execution times out. Unit: Seconds. Default: 3600 (1 hour). The maximum value of this item is 21600 (6 hours).

  • query_port: The query port, which must be within the range of 1-65535 excluding 443. The default query port is port 9030. Note that this argument can be specified only at cluster deployment, and cannot be modified once it is set.

  • idle_suspend_interval: The amount of time (in minutes) during which the cluster can stay idle. After the specified time period elapses, the cluster will be automatically suspended. The Auto Suspend feature is disabled by default. To enable the Auto Suspend feature, set this argument to an integer with the range of 15-999999. To disable this feature again, remove this argument from your Terraform configuration.

  • scheduling_policy:(Optional, List) When specified. CelerData will automatically suspend the cluster to save the majority of costs on EC2 (only EBS costs will be incurred) and resume the cluster for usage as scheduled.

    • policy_name: (Required) Policy name.
    • description: (Optional) Explanation of this policy strategy.
    • active_days: (Required) Configure the date when the cluster scheduling policy is triggered. Available values:
      • MONDAY
      • TUESDAY
      • WEDNESDAY
      • THURSDAY
      • FRIDAY
      • SATURDAY
      • SUNDAY
    • time_zone: (Optional) Specify your IANA Time-Zone. Default: UTC.
    • resume_at: (Optional) Cluster auto resume time. resume_at and suspend_at cannot both be empty.
    • suspend_at: (Optional) Cluster auto suspend time.
    • enable: (Required) Whether to enable this scheduling policy. When specified as true, the system will perform cluster scheduling according to this policy.

See Also​

AWS​

Azure​

GCP​