Skip to main content
Version: 2.6.0

Common

This document describes the Prometheus metrics generated both by Aperture Agents and the Aperture Controller.

Go & Process Metrics

Go & Process metrics can be exposed by enabling the enable_go_metrics flag and enable_process_collector in Agent's MetricsConfig and Controller's MetricsConfig. See

collector.NewGoCollector for more information.

HTTP Server Metrics

Metrics

NameTypeLabelsUnitDescription
http_requests_totalCounteragent_group, instance, job, process_uuid, handler_name, http_method, http_status_codecount (no unit)Total number of requests received
http_errors_totalCounteragent_group, instance, job, process_uuid, handler_name, http_method, http_status_codecount (no unit)Total number of errors that occurred
http_requests_latency_msHistogramagent_group, instance, job, process_uuid, handler_name, http_method, http_status_codemsLatency of the requests processed by the server

Labels

NameExampleDescription
agent_groupdefaultAgent Group of the policy that Flux Meter belongs to
instanceaperture-agent-cbfnpHost instance of the Aperture Agent
jobaperture-selfThe configured job name that the target belongs to
process_uuiddc0e82af-6730-4f70-8228-ee91da53ac5fHost instance's UUID
handler_namedefaultHTTP handler name
http_methodGET, POSTHTTP method
http_status_code200, 503HTTP status code

gRPC Server Metrics

gRPC server metrics are exposed by default. See grpc-ecosystem/go-grpc-prometheus server_metrics for more information.