aistudiooutboundrules

Azure AI Studio Hub Outbound Rules Terraform Module

Info for user-defined outbound rule management

This module has been created to centrally manage user-defined outbound rules for an Azure AI Hub.

New user-defined outbound rules can technically already be added individually via the Microsoft.MachineLearningServices/workspaces/outboundRules resource type. However, the next outbound rule in a list can only be added after the previous rule has been added successfully. Since, Terraform does not support a batch size parameter when creating resources using the for_each meta-argument, user-defined outbound rules must be added in batches and cannot be created individually. If you don't follow the batch approach, you will see unexpected behaviors and configuration results when adding a large number of rules individually via the for_each meta-argument.

When working on larger projects, the outbound rules may also depend on other Terraform resources to be created. Due to these internal dependencies you may also require creating these outbound rules after creating the AI Hub. Hence, the outbound rules cannot be defined when creating the initial AI Hub resource. or you may end up with cyclic dependencies in your Terraform configuration. This can be avoided by using this Terraform module.

This module has been created to manage all user-defined outbound rules centrally and apply the configuration in batches and overcome all aforementioned issues.

Documentation

Requirements

The following requirements are needed by this module:

• terraform (>=0.12)

• azapi (~> 2.0)

• null (>= 3.2.2)

Modules

No modules.

Required Inputs

The following input variables are required:

ai_studio_hub_id

Description: Specifies the resource id of the ai studio hub.

Type: string

ai_studio_hub_storage_account_id

Description: Specifies the id of the storage account that is connected to the ai studio hub.

Type: string

Optional Inputs

The following input variables are optional (have default values):

ai_studio_hub_approve_private_endpoints

Description: Specifies whether the managed private endpoints should be approved as part of the ai studio hub deployment.

Type: bool

Default: false

ai_studio_hub_outbound_rules_fqdns

Description: Specifies the outbound FQDN rules that should be added to the AI Studio Hub. Only provide FQDNs without specific paths such as 'microsoft.com' or '*.microsoft.com' but NOT 'microsoft.com/mypath'.

Type: map(string)

Default: {}

ai_studio_hub_outbound_rules_private_endpoints

Description: Specifies the private endpoint rules that should be added to the AI Studio Hub.

Type:

map(object({
    private_connection_resource_id = string
    subresource_name               = string
  }))

Default: {}

ai_studio_hub_outbound_rules_service_endpoints

Description: Specifies the service endpoint rules that should be added to the AI Studio Hub.

Type:

map(object({
    service_tag = string
    protocol    = optional(string, "TCP")
    port_range  = optional(string, "443")
  }))

Default: {}

ai_studio_hub_provision_managed_network

Description: Specifies whether the managed vnet should be provisioned as part of the ai studio hub deployment.

Type: bool

Default: false

Outputs

No outputs.