hashicorp/terraform-provider-google-beta v7.42.0

on GitHub

NOTES:

• compute: migrated google_compute_region_instance_template resource to use direct HTTP rather than a client library (#12752)

DEPRECATIONS:

• vertexai: deprecated google_vertex_ai_schedule, an accidentally-added duplicate resource; use google_colab_schedule instead. (#12739)

FEATURES:

• New List Resource: google_service_account_key (#12750)
• New Resource: google_agent_identity_auth_provider (#12762)
• New Resource: google_apihub_runtime_project_attachment (#12764)
• New Resource: google_chronicle_big_query_export (#12737)
• New Resource: google_vector_search_data_object (#12755)
• New Resource: google_vertex_ai_persistent_resource (#12756)

IMPROVEMENTS:

• bigquery: added table_type field to google_bigquery_routine resource (#12761)
• colab: added catch_up, create_pipeline_job_request, create_time, last_pause_time, last_resume_time, last_scheduled_run_response, max_concurrent_active_run_count, next_run_time, started_run_count, update_time fields, and sub-fields under create_notebook_execution_job_request.notebook_execution_job (create_time, custom_environment_spec, encryption_spec, job_state, kernel_name, labels, name, schedule_resource_name, workbench_runtime) and under create_notebook_execution_job_request (notebook_execution_job_id, parent) to google_colab_schedule resource (#12739)
• composer: added config.node_config.traffic_routing_config field to google_composer_environment resource (#12733)
• compute: added effective_location field to google_compute_interconnect resource (#12742)
• compute: added request_headers and response_headers fields to log_config on google_compute_backend_service and google_compute_region_backend_service resources (#12746)
• compute: added identity support to google_compute_instance, allowing resource import using an identity block (#12754)
• compute: changed location field to mutable for google_compute_interconnect resource (#12742)
• container: added addons_config.node_readiness_config field to google_container_cluster resource (#12743)
• container: added rollback_safe_upgrade, desired_emulated_version, and emulated_version fields to google_container_cluster resource (#12757)
• container: increased default timeout to 2 hours for google_container_node_pool resource (#12727)
• dataproc: added confidential_instance_type field to google_dataproc_cluster resource (#12725)
• gkehub: added min_control_plane_version, min_node_version, target_control_plane_version, target_node_version, and operational_state fields to google_gke_hub_rollout_sequence resource (#12749)
• hypercomputecluster: increased default timeouts for google_hypercomputecluster_cluster to 120 minutes (#12763)
• modelarmor: added template_metadata.filter_version_selector field to google_model_armor_template resource (#12736)
• sql: added identity support to google_sql_user for terraform query support (#12748)

BUG FIXES:

• bigtable: fixed an issue where bigtable_custom_endpoint and universe_domain were ignored when creating Bigtable resources (#12738)
• compute: fixed an issue where diffs in google_compute_security_policy were not detected (#12745)
• gkehub: fixed rollout_creation_scope and upgrade_types fields in google_gke_hub_rollout_sequence resource (#12749)
• osconfig: added client-side validation to ensure resource_hierarchy_selector and location_selector are not set at the same time in google_os_config_v2_policy_orchestrator, google_os_config_v2_policy_orchestrator_for_folder, and google_os_config_v2_policy_orchestrator_for_organization (#12740)
• secretmanager: fixed an issue where google_secret_manager_secret_version would fail at apply time if neither secret_data nor secret_data_wo was set (#12744)
• sql: fixed issue where updates to settings.ip_configuration.psc_config.allowed_consumer_projects in google_sql_database_instance were silently ignored on in-place updates (#12759)
• vertexai: fixed google_vertex_ai_endpoint_with_model_garden_deployment destroying and recreating the endpoint when min_replica_count, max_replica_count, required_replica_count, or autoscaling_metric_specs changed (#12735)