/

Metrics

Akv2k8s metrics

Both the Controller and Env-Injector have Prometheus metrics than can be enabled through the Helm chart.

--set global.metrics.enabled=true

The above setting will enable a Prometheus endpoint that can be scraped for metrics.

If the Prometheus Operator is available in the cluster, a ServiceMonitor can be enabled to allow Prometheus to scrape the metrics.

--set global.metrics.serviceMonitor.enabled=true/false  # default false
--set global.metrics.serviceMonitor.interval=[timespan] # default 30s

The same settings are available for each component and can override global settings by replacing global with controller or env_injector. For other available settings, see Helm chart. To enable metrics for the Env-Injector only:

--set env_injector.metrics.enabled=true

Available Env-Injector Metrics

akv2k8s_pod_mutations_total                # total number of pods mutated 
akv2k8s_pod_inspections_total              # total number of pods inspected, including mutated
akv2k8s_pod_mutations_failed_total         # total number of attempted pod mutations that failed
akv2k8s_auth_requests_total                # total number of successful auth requests
akv2k8s_auth_requests_failed_total         # total number failed auth requests
akv2k8s_auth_validations_total             # total number of successful auth validations
akv2k8s_auth_validations_failed_total      # total number of failed auth validations
akv2k8s_container_inspections_total        # total number of inspected container images
akv2k8s_container_inspections_failed_total # total number of failed container images inspections

Available Controller Metrics

akv2k8s_syncs_total        # total number of syncs
akv2k8s_syncs_failed_total # total number of sync failures
Edit on GitHub