Skip to content

Latest commit

 

History

History
96 lines (71 loc) · 4.21 KB

metrics.md

File metadata and controls

96 lines (71 loc) · 4.21 KB

Metrics

K8GB generates Prometheus-compatible metrics. Metrics endpoints are exposed via -metrics service in operator namespace and can be scraped by 3rd party tools:

spec:
...
  ports:
  - name: http-metrics
    port: 8383
    protocol: TCP
    targetPort: 8383
  - name: cr-metrics
    port: 8686
    protocol: TCP
    targetPort: 8686

Metrics can be also automatically discovered and monitored by Prometheus Operator via automatically generated ServiceMonitor CRDs , in case if Prometheus Operator is deployed into the cluster.

General metrics

controller-runtime standard metrics, extended with K8GB operator-specific metrics listed below:

healthy_records

Number of healthy records observed by K8GB.

Example:

# HELP k8gb_gslb_healthy_records Number of healthy records observed by K8GB.
# TYPE k8gb_gslb_healthy_records gauge
k8gb_gslb_healthy_records{name="test-gslb",namespace="test-gslb"} 6

ingress_hosts_per_status

Number of ingress hosts per status (NotFound, Healthy, Unhealthy), observed by K8GB.

Example:

# HELP k8gb_gslb_ingress_hosts_per_status Number of managed hosts observed by K8GB.
# TYPE k8gb_gslb_ingress_hosts_per_status gauge
k8gb_gslb_ingress_hosts_per_status{name="test-gslb",namespace="test-gslb",status="Healthy"} 1
k8gb_gslb_ingress_hosts_per_status{name="test-gslb",namespace="test-gslb",status="NotFound"} 1
k8gb_gslb_ingress_hosts_per_status{name="test-gslb",namespace="test-gslb",status="Unhealthy"} 2

Served on 0.0.0.0:8383/metrics endpoint

Custom resource specific metrics

Info metrics, automatically exposed by operator based on the number of the current instances of an operator's custom resources in the cluster.

Example:

# HELP gslb_info Information about the Gslb custom resource.
# TYPE gslb_info gauge
gslb_info{namespace="test-gslb",gslb="test-gslb"} 1

Served on 0.0.0.0:8686/metrics endpoint

Metrics

The k8gb exposes several metrics to help you monitor the health and behavior.

Metric Type Description Labels
k8gb_gslb_errors_total Counter Number of errors namespace, name
k8gb_gslb_healthy_records Gauge Number of healthy records observed by k8gb. namespace, name
k8gb_gslb_reconciliation_loops_total Counter Number of successful reconciliation loops. namespace, name
k8gb_gslb_service_status_num Gauge Number of managed hosts observed by k8gb. namespace, name, status
k8gb_gslb_status_count_for_failover Gauge Gslb status count for Failover strategy. namespace, name, status
k8gb_gslb_status_count_for_geoip Gauge Gslb status count for GeoIP strategy. namespace, name, status
k8gb_gslb_status_count_for_roundrobin Gauge Gslb status count for RoundRobin strategy. namespace, name, status
k8gb_infoblox_heartbeat_errors_total Counter Number of k8gb Infoblox TXT record errors. namespace, name
k8gb_infoblox_heartbeats_total Counter Number of k8gb Infoblox heartbeat TXT record updates. namespace, name
k8gb_infoblox_request_duration Histogram Duration of the HTTP request to Infoblox API in seconds. request, success
k8gb_infoblox_zone_update_errors_total Counter Number of k8gb Infoblox zone update errors. namespace, name
k8gb_infoblox_zone_updates_total Counter Number of k8gb Infoblox zone updates. namespace, name
k8gb_endpoint_status_num Gauge Number of targets in DNS endpoint. namespace, name, dns_name
k8gb_runtime_info Gauge K8gb runtime info. namespace, k8gb_version,
go_version, arch, os, git_sha

OpenTracing

Optionally k8gb operator can expose traces in OpenTelemetry format to any available OTEL compliant tracing solution. Consult the following page for more details.