Skip to main content

​
Schema

​
Required

  • admin_username (String) MLflow admin username.
  • network_id (String) ID of the vpc network.
  • parent_id (String) Identifier of the parent resource to which the resource belongs.
  • service_account_id (String) Id of the service account that will be used to access S3 bucket (and create one if not provided).

​
Optional

  • admin_password (String, Sensitive) MLflow admin password.
  • description (String) Description of the cluster.
  • labels (Map of String) : Labels associated with the resource.
  • metadata (Attributes) :

    ​
    Inner value description

    Common resource metadata. (see below for nested schema)
  • name (String) Human readable name for the resource.
  • public_access (Boolean) Either make cluster public accessible or accessible only via private VPC.
  • sensitive (Attributes, Sensitive) This object mirrors the structure of the resource, providing view on the resource’s write-only fields. It is a preferred way to define such fields as they are not stored in the state file. To use any field in write-only mode, do not set it in the main resource, instead set it in the sensitive object. Setting the field in the main resource will overwrite the field in the sensitive object at merging. (see below for nested schema)
  • size (String) : Size defines how much resources will be allocated to mlflow See supported sizes in the documentation. Default size is the smallest available in the region.
  • storage_bucket_name (String) Name of the Nebius S3 bucket for MLflow artifacts. If not provided, will be created under the same parent.

​
Read-Only

  • created_at (String) : Timestamp indicating when the resource was created. A string representing a timestamp in ISO 8601 format: YYYY-MM-DDTHH:MM:SSZ or YYYY-MM-DDTHH:MM:SS.SSS±HH:MM
  • id (String) Identifier for the resource, unique for its resource type.
  • resource_version (Number) : Version of the resource for safe concurrent modifications and consistent reads. Positive and monotonically increases on each resource spec change (but not on each change of the resource’s container(s) or status). Service allows zero value or current.
  • status (Attributes) (see below for nested schema)
  • updated_at (String) : Timestamp indicating when the resource was last updated. A string representing a timestamp in ISO 8601 format: YYYY-MM-DDTHH:MM:SSZ or YYYY-MM-DDTHH:MM:SS.SSS±HH:MM

​
Nested Schema for metadata

​
Nested Schema for sensitive

Optional:
  • admin_password (String, Sensitive, Write-only) MLflow admin password.
  • version (String) The version of the write-only fields combined. You can use a hash here, some number, or a version of the secret from which you pull the value, or you can concatenate all of that together. This field is not sent to the server, but it is stored in the state, so don’t store raw secrets here. Changing this field will trigger an update of the resource even if no other fields are changed. Use of this field is not required, but if you do not use it, changes in write-only fields will not trigger updates of the resource by themselves and therefore might not be reflected.

​
Nested Schema for status

Read-Only:
  • effective_storage_bucket_name (String) Name of the Nebius S3 bucket for MLflow artifacts.
  • experiments_count (Number) Count of experiments in the MLflow cluster
  • mlflow_version (String) MLflow version
  • phase (String) : Current phase of the cluster.

    ​
    Supported values

    Possible values:
    • PHASE_UNSPECIFIED
    • PHASE_PROVISIONING
    • PHASE_RUNNING
    • PHASE_UPDATING
    • PHASE_DELETING
    • PHASE_DELETED[deprecated]
    • PHASE_PURGING
    • PHASE_STOPPING
    • PHASE_RESUMING
  • state (String) : State reflects substatus of the phase to define whether it’s healthy or not.

    ​
    Supported values

    Possible values:
    • STATE_UNSPECIFIED
    • STATE_IN_PROGRESS
    • STATE_FINISHED
    • STATE_ERROR
    • STATE_DEGRADED
    • STATE_SCHEDULED
  • tracking_endpoint (String) : Tracking endpoint url. Will be removed soon in favor of private_tracking_endpoint and public_tracking_endpoint.
  • tracking_endpoints (Attributes) Public and private tracking endpoints (see below for nested schema)

​
Nested Schema for status.tracking_endpoints

Read-Only:
  • private (String) Private endpoint
  • public (String) Public endpoint