API Aggregator metrics and alerts

Find the metrics APIA exposes and the alerts defined for APIA.

Related documentation:

• How Outbound works

Service	CRD or annotations?	Port	Endpoint/Selector	Metrics update interval
API Aggregator	ServiceMonitor	9102	/metrics	15 seconds

See details about:

• API Aggregator metrics
• API Aggregator alerts

Metrics

Here are some of the metrics exposed by API aggregator.

Metric and description	Metric details	Indicator of
cxc_api_aggregator_schedules_created_total Total schedules created.	Unit: Type: Counter Label: Sample value: 42
cxc_api_aggregator_schedules_removed_total Total schedules removed.	Unit: Type: Counter Label: Sample value: 42
cxc_api_aggregator_campaign_template_created_total Total campaign templates created.	Unit: Type: Counter Label: Sample value: 42
cxc_api_aggregator_campaign_template_removed_total Total campaign templates removed.	Unit: Type: Counter Label: Sample value: 42
cxc_api_aggregator_users_logged_in_total Total number of users who are logged in.	Unit: Type: Gauge Label: Sample value: 4.2
cxc_api_aggregator_users_logged_out_total Total number of users who are logged out.	Unit: Type: Gauge Label: Sample value: 4.2
cxc_api_aggregator_api_requests_total Total count of requests.	Unit: Type: Counter Label: "'ccid', 'tenant_name'" Sample value: 42
cxc_api_healthy_instance Healthy instance.	Unit: Type: Gauge Label: "'ccid', 'tenant_name'" Sample value: 4.2
cxc_api_aggregator_api_requests_processed_success Total count of success requests.	Unit: Type: Counter Label: "'ccid', 'tenant_name'" Sample value: 42
cxc_api_aggregator_top_api_requests Top api requests.	Unit: Type: Counter Label: "'path', 'method', 'id', 'name', 'ccid', 'tenant_name', 'code'" Sample value: 42
cxc_api_aggregator_redis_connection_failed Failed Redis connection.	Unit: Type: Gauge Label: "'ccid', 'tenant_name'" Sample value: 4.2
cxc_api_aggregator_request_count Total requests by verb and code.	Unit: Type: Counter Label: "'method', 'path', 'code'" Sample value: 42
cxc_api_aggregator_request_latencies_ms Request latencies histogram by verb, in milliseconds.	Unit: Type: Histogram Label: "'method', 'path', 'code'" Sample value: [1, 2, 3]
cxc_api_aggregator_request_out_count Total out requests by verb, destination and code.	Unit: Type: Counter Label: "'method', 'destination', 'code'" Sample value: 42
cxc_api_aggregator_request_out_latencies_ms Out Request latencies histogram by verb, destination and code, in milliseconds.	Unit: Type: Histogram Label: "'method', 'destination', 'code'" Sample value: [1, 2, 3]
cxc_api_aggregator_elasticsearch_service_latencies_ms Elasticsearch Request latencies histogram by verb, destination and code, in milliseconds.	Unit: Type: Histogram Label: "'method', 'destination', 'code'" Sample value: [1, 2, 3]

Alerts

The following alerts are defined for API Aggregator.

Alert	Severity	Description	Based on	Threshold
CXC-API-LatencyHigh	HIGH	Triggered when the latency for API responses is beyond the defined threshold.		2500ms for 5m
CXC-API-Redis-Connection-Failed	HIGH	Triggered when the connection to redis fails for more than 1 minute.		1m
CXC-EXT-Ingress-Error-Rate	HIGH	Triggered when the Ingress error rate is above the specified threshold.		20% for 5m
cxc_api_too_many_errors_from_auth	HIGH	Triggered when there are too many error responses from the auth service for more than the specified time threshold.		1m
CXC-CPUUsage	HIGH	Triggered when the CPU utilization of a pod is beyond the threshold		300% for 5m
CXC-MemoryUsage	HIGH	Triggered when the memory utilization of a pod is beyond the threshold.		70% for 5m
CXC-PodNotReadyCount	HIGH	Triggered when the number of pods ready for a CX Contact deployment is less than or equal to the threshold.		1 for 5m
CXC-PodRestartsCount	HIGH	Triggered when the restart count for a pod is beyond the threshold.		1 for 5m
CXC-MemoryUsagePD	HIGH	Triggered when the memory usage of a pod is above the critical threshold.		90% for 5m
CXC-PodRestartsCountPD	HIGH	Triggered when the restart count is beyond the critical threshold.		5 for 5m
CXC-PodsNotReadyPD	HIGH	Triggered when there are no pods ready for CX Contact deployment.		0 for 1m