def sample_create_detection_config(): # [START create_anomaly_detection_config] from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorAdministrationClient from azure.ai.metricsadvisor.models import (ChangeThresholdCondition, HardThresholdCondition, SmartDetectionCondition, SuppressCondition, MetricDetectionCondition, AnomalyDetectionConfiguration) service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT") subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY") api_key = os.getenv("METRICS_ADVISOR_API_KEY") metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID") client = MetricsAdvisorAdministrationClient( service_endpoint, MetricsAdvisorKeyCredential(subscription_key, api_key)) change_threshold_condition = ChangeThresholdCondition( anomaly_detector_direction="Both", change_percentage=20, shift_point=10, within_range=True, suppress_condition=SuppressCondition(min_number=5, min_ratio=2)) hard_threshold_condition = HardThresholdCondition( anomaly_detector_direction="Up", upper_bound=100, suppress_condition=SuppressCondition(min_number=2, min_ratio=2)) smart_detection_condition = SmartDetectionCondition( anomaly_detector_direction="Up", sensitivity=10, suppress_condition=SuppressCondition(min_number=2, min_ratio=2)) anomaly_detection_configuration = AnomalyDetectionConfiguration( name="my_detection_config", metric_id=metric_id, description="anomaly detection config for metric", whole_series_detection_condition=MetricDetectionCondition( cross_conditions_operator="OR", change_threshold_condition=change_threshold_condition, hard_threshold_condition=hard_threshold_condition, smart_detection_condition=smart_detection_condition)) detection_config = client.create_metric_anomaly_detection_configuration( anomaly_detection_configuration) return detection_config
class TestMetricsAdvisorAdministrationClientBase(AzureTestCase): FILTER_HEADERS = ReplayableTest.FILTER_HEADERS + ['Ocp-Apim-Subscription-Key', 'x-api-key'] def __init__(self, method_name): super(TestMetricsAdvisorAdministrationClientBase, self).__init__(method_name) self.vcr.match_on = ["path", "method", "query"] if self.is_live: service_endpoint = self.get_settings_value("METRICS_ADVISOR_ENDPOINT") subscription_key = self.get_settings_value("METRICS_ADVISOR_SUBSCRIPTION_KEY") api_key = self.get_settings_value("METRICS_ADVISOR_API_KEY") self.sql_server_connection_string = self.get_settings_value("METRICS_ADVISOR_SQL_SERVER_CONNECTION_STRING") self.azure_table_connection_string = self.get_settings_value("METRICS_ADVISOR_AZURE_TABLE_CONNECTION_STRING") self.azure_blob_connection_string = self.get_settings_value("METRICS_ADVISOR_AZURE_BLOB_CONNECTION_STRING") self.azure_cosmosdb_connection_string = self.get_settings_value("METRICS_ADVISOR_COSMOS_DB_CONNECTION_STRING") self.http_request_get_url = self.get_settings_value("METRICS_ADVISOR_HTTP_GET_URL") self.http_request_post_url = self.get_settings_value("METRICS_ADVISOR_HTTP_POST_URL") self.application_insights_api_key = self.get_settings_value("METRICS_ADVISOR_APPLICATION_INSIGHTS_API_KEY") self.azure_data_explorer_connection_string = self.get_settings_value("METRICS_ADVISOR_AZURE_DATA_EXPLORER_CONNECTION_STRING") self.influxdb_connection_string = self.get_settings_value("METRICS_ADVISOR_INFLUX_DB_CONNECTION_STRING") self.influxdb_password = self.get_settings_value("METRICS_ADVISOR_INFLUX_DB_PASSWORD") self.azure_datalake_account_key = self.get_settings_value("METRICS_ADVISOR_AZURE_DATALAKE_ACCOUNT_KEY") self.mongodb_connection_string = self.get_settings_value("METRICS_ADVISOR_AZURE_MONGO_DB_CONNECTION_STRING") self.mysql_connection_string = self.get_settings_value("METRICS_ADVISOR_MYSQL_CONNECTION_STRING") self.postgresql_connection_string = self.get_settings_value("METRICS_ADVISOR_POSTGRESQL_CONNECTION_STRING") self.elasticsearch_auth_header = self.get_settings_value("METRICS_ADVISOR_ELASTICSEARCH_AUTH_HEADER") self.anomaly_detection_configuration_id = self.get_settings_value("METRICS_ADVISOR_ANOMALY_DETECTION_CONFIGURATION_ID") self.data_feed_id = self.get_settings_value("METRICS_ADVISOR_DATA_FEED_ID") self.metric_id = self.get_settings_value("METRICS_ADVISOR_METRIC_ID") self.scrubber.register_name_pair( self.sql_server_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.azure_table_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.azure_blob_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.azure_cosmosdb_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.http_request_get_url, "connectionstring" ) self.scrubber.register_name_pair( self.http_request_post_url, "connectionstring" ) self.scrubber.register_name_pair( self.application_insights_api_key, "connectionstring" ) self.scrubber.register_name_pair( self.azure_data_explorer_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.influxdb_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.influxdb_password, "connectionstring" ) self.scrubber.register_name_pair( self.azure_datalake_account_key, "connectionstring" ) self.scrubber.register_name_pair( self.mongodb_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.mysql_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.postgresql_connection_string, "connectionstring" ) self.scrubber.register_name_pair( self.elasticsearch_auth_header, "connectionstring" ) self.scrubber.register_name_pair( self.metric_id, "metric_id" ) self.scrubber.register_name_pair( self.data_feed_id, "data_feed_id" ) self.scrubber.register_name_pair( self.anomaly_detection_configuration_id, "anomaly_detection_configuration_id" ) else: service_endpoint = "https://endpointname.cognitiveservices.azure.com" subscription_key = "METRICS_ADVISOR_SUBSCRIPTION_KEY" api_key = "METRICS_ADVISOR_API_KEY" self.sql_server_connection_string = "SQL_SERVER_CONNECTION_STRING" self.azure_table_connection_string = "AZURE_TABLE_CONNECTION_STRING" self.azure_blob_connection_string = "AZURE_BLOB_CONNECTION_STRING" self.azure_cosmosdb_connection_string = "COSMOS_DB_CONNECTION_STRING" self.http_request_get_url = "METRICS_ADVISOR_HTTP_GET_URL" self.http_request_post_url = "METRICS_ADVISOR_HTTP_POST_URL" self.application_insights_api_key = "METRICS_ADVISOR_APPLICATION_INSIGHTS_API_KEY" self.azure_data_explorer_connection_string = "METRICS_ADVISOR_AZURE_DATA_EXPLORER_CONNECTION_STRING" self.influxdb_connection_string = "METRICS_ADVISOR_INFLUXDB_CONNECTION_STRING" self.influxdb_password = "******" self.azure_datalake_account_key = "METRICS_ADVISOR_AZURE_DATALAKE_ACCOUNT_KEY" self.mongodb_connection_string = "METRICS_ADVISOR_AZURE_MONGODB_CONNECTION_STRING" self.mysql_connection_string = "METRICS_ADVISOR_MYSQL_CONNECTION_STRING" self.postgresql_connection_string = "METRICS_ADVISOR_POSTGRESQL_CONNECTION_STRING" self.elasticsearch_auth_header = "METRICS_ADVISOR_ELASTICSEARCH_AUTH" self.anomaly_detection_configuration_id = "anomaly_detection_configuration_id" self.metric_id = "metric_id" self.data_feed_id = "data_feed_id" self.admin_client = MetricsAdvisorAdministrationClient(service_endpoint, MetricsAdvisorKeyCredential(subscription_key, api_key)) def _create_data_feed(self, name): name = create_random_name(name) return self.admin_client.create_data_feed( DataFeed( name=name, source=SQLServerDataFeed( connection_string=self.sql_server_connection_string, query='select * from adsample2 where Timestamp = @StartTime' ), granularity=DataFeedGranularity( granularity_type="Daily", ), schema=DataFeedSchema( metrics=[ DataFeedMetric(name="cost"), DataFeedMetric(name="revenue") ], dimensions=[ DataFeedDimension(name="category"), DataFeedDimension(name="city") ], timestamp_column="Timestamp" ), ingestion_settings=DataFeedIngestionSettings( ingestion_begin_time=datetime.datetime(2019, 10, 1), ) ) ) def _create_data_feed_and_anomaly_detection_config(self, name): data_feed = self._create_data_feed(name) detection_config_name = create_random_name(name) detection_config = self.admin_client.create_metric_anomaly_detection_configuration( AnomalyDetectionConfiguration( name=detection_config_name, metric_id=data_feed.metric_ids[0], description="testing", whole_series_detection_condition=MetricDetectionCondition( smart_detection_condition=SmartDetectionCondition( sensitivity=50, anomaly_detector_direction="Both", suppress_condition=SuppressCondition( min_number=50, min_ratio=50 ) ) ) ) ) return detection_config, data_feed def _create_data_feed_for_update(self, name): data_feed_name = create_random_name(name) return self.admin_client.create_data_feed( DataFeed( name=data_feed_name, source=SQLServerDataFeed( connection_string=self.sql_server_connection_string, query='select * from adsample2 where Timestamp = @StartTime' ), granularity=DataFeedGranularity( granularity_type="Daily", ), schema=DataFeedSchema( metrics=[ DataFeedMetric(name="cost", display_name="display cost", description="the cost"), DataFeedMetric(name="revenue", display_name="display revenue", description="the revenue") ], dimensions=[ DataFeedDimension(name="category", display_name="display category"), DataFeedDimension(name="city", display_name="display city") ], timestamp_column="Timestamp" ), ingestion_settings=DataFeedIngestionSettings( ingestion_begin_time=datetime.datetime(2019, 10, 1), data_source_request_concurrency=0, ingestion_retry_delay=-1, ingestion_start_offset=-1, stop_retry_after=-1, ), options=DataFeedOptions( admin_emails=["*****@*****.**"], data_feed_description="my first data feed", missing_data_point_fill_settings=DataFeedMissingDataPointFillSettings( fill_type="SmartFilling" ), rollup_settings=DataFeedRollupSettings( rollup_type="NoRollup", rollup_method="None", ), viewer_emails=["viewers"], access_mode="Private", action_link_template="action link template" ) ) ) def _create_anomaly_alert_config_for_update(self, name): detection_config, data_feed = self._create_data_feed_and_anomaly_detection_config(name) alert_config_name = create_random_name(name) alert_config = self.admin_client.create_anomaly_alert_configuration( AnomalyAlertConfiguration( name=alert_config_name, cross_metrics_operator="AND", metric_alert_configurations=[ MetricAlertConfiguration( detection_configuration_id=detection_config.id, alert_scope=MetricAnomalyAlertScope( scope_type="TopN", top_n_group_in_scope=TopNGroupScope( top=5, period=10, min_top_count=9 ) ), alert_conditions=MetricAnomalyAlertConditions( metric_boundary_condition=MetricBoundaryCondition( direction="Both", companion_metric_id=data_feed.metric_ids[0], lower=1.0, upper=5.0 ) ) ), MetricAlertConfiguration( detection_configuration_id=detection_config.id, alert_scope=MetricAnomalyAlertScope( scope_type="SeriesGroup", series_group_in_scope={'city': 'Shenzhen'} ), alert_conditions=MetricAnomalyAlertConditions( severity_condition=SeverityCondition( min_alert_severity="Low", max_alert_severity="High" ) ) ), MetricAlertConfiguration( detection_configuration_id=detection_config.id, alert_scope=MetricAnomalyAlertScope( scope_type="WholeSeries" ), alert_conditions=MetricAnomalyAlertConditions( severity_condition=SeverityCondition( min_alert_severity="Low", max_alert_severity="High" ) ) ) ], hook_ids=[] ) ) return alert_config, data_feed, detection_config def _create_detection_config_for_update(self, name): data_feed = self._create_data_feed(name) detection_config_name = create_random_name("testupdated") detection_config = self.admin_client.create_metric_anomaly_detection_configuration( AnomalyDetectionConfiguration( name=detection_config_name, metric_id=data_feed.metric_ids[0], description="My test metric anomaly detection configuration", whole_series_detection_condition=MetricDetectionCondition( cross_conditions_operator="AND", smart_detection_condition=SmartDetectionCondition( sensitivity=50, anomaly_detector_direction="Both", suppress_condition=SuppressCondition( min_number=50, min_ratio=50 ) ), hard_threshold_condition=HardThresholdCondition( anomaly_detector_direction="Both", suppress_condition=SuppressCondition( min_number=5, min_ratio=5 ), lower_bound=0, upper_bound=100 ), change_threshold_condition=ChangeThresholdCondition( change_percentage=50, shift_point=30, within_range=True, anomaly_detector_direction="Both", suppress_condition=SuppressCondition( min_number=2, min_ratio=2 ) ) ), series_detection_conditions=[MetricSingleSeriesDetectionCondition( series_key={"city": "Shenzhen", "category": "Jewelry"}, smart_detection_condition=SmartDetectionCondition( anomaly_detector_direction="Both", sensitivity=63, suppress_condition=SuppressCondition( min_number=1, min_ratio=100 ) ) )], series_group_detection_conditions=[MetricSeriesGroupDetectionCondition( series_group_key={"city": "Sao Paulo"}, smart_detection_condition=SmartDetectionCondition( anomaly_detector_direction="Both", sensitivity=63, suppress_condition=SuppressCondition( min_number=1, min_ratio=100 ) ) )] ) ) return detection_config, data_feed def _create_email_hook_for_update(self, name): return self.admin_client.create_hook( hook=EmailNotificationHook( name=name, emails_to_alert=["*****@*****.**"], description="my email hook", external_link="external link" ) ) def _create_web_hook_for_update(self, name): return self.admin_client.create_hook( hook=WebNotificationHook( name=name, endpoint="https://httpbin.org/post", description="my web hook", external_link="external link", username="******", password="******" ) )