Control: GCP > Vertex AI > Endpoint > Active > Request-Response Logging

Determine the action to take when a GCP Vertex AI endpoint does not have request-response logging enabled.

Request-response logging captures a sample of prediction requests and responses to a BigQuery table. When enabled it provides an audit trail of model traffic. This control checks the predictRequestResponseLoggingConfig.enabled field and raises an alarm when the policy requires logging to be enabled but it is not.

Resource Types

This control targets the following resource types:

GCP > Vertex AI > Endpoint

Policies

The following policies can be used to configure this control:

GCP > Vertex AI > Endpoint > Active > Request-Response Logging

In Your Workspace

Developers

Control Type URI

tmod:@turbot/gcp-vertexai#/control/types/endpointActiveRequestResponseLogging

Category URI

tmod:@turbot/turbot#/control/categories/resourceActive

GraphQL

query controlType(id: "tmod:@turbot/gcp-vertexai#/control/types/endpointActiveRequestResponseLogging") { … }

query controls(filter: "controlTypeId:'tmod:@turbot/gcp-vertexai#/control/types/endpointActiveRequestResponseLogging'") { … }

CLI

Get Controls

turbot graphql controls --filter "controlTypeId:tmod:@turbot/gcp-vertexai#/control/types/endpointActiveRequestResponseLogging"