Add KubeDNS/CoreDNS support in cluster metrics (#892)

Signed-off-by: Pete Wall <[email protected]>

charts/feature-cluster-metrics/README.md

@@ -201,7 +201,7 @@ Actual integration testing in a live environment should be done in the main [k8s
 
 | Key | Type | Default | Description |
 |-----|------|---------|-------------|
-| controlPlane.enabled | bool | `false` | enable all Kubernetes Control Plane metrics sources. This includes api-server, kube-scheduler, kube-controller-manager, and etcd. |
+| controlPlane.enabled | bool | `false` | enable all Kubernetes Control Plane metrics sources. This includes api-server, kube-scheduler, kube-controller-manager, and KubeDNS. |
 
 ### General settings
 
@@ -263,6 +263,18 @@ Actual integration testing in a live environment should be done in the main [k8s
 | kubeControllerManager.port | int | `10257` | Port number used by the Kube Controller Manager, set by `--secure-port.` |
 | kubeControllerManager.scrapeInterval | string | 60s | How frequently to scrape metrics from the Kube Controller Manager Overrides metrics.scrapeInterval |
 
+### KubeDNS
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| kubeDNS.enabled | bool | `false` | Scrape metrics from KubeDNS |
+| kubeDNS.extraDiscoveryRules | string | `""` | Rule blocks to be added to the discovery.relabel component for KubeDNS. These relabeling rules are applied pre-scrape against the targets from service discovery. Before the scrape, any remaining target labels that start with `__` (i.e. `__meta_kubernetes*`) are dropped. ([docs](https://grafana.com/docs/alloy/latest/reference/components/discovery/discovery.relabel/#rule-block)) |
+| kubeDNS.extraMetricProcessingRules | string | `""` | Rule blocks to be added to the prometheus.relabel component for KubeDNS. These relabeling rules are applied post-scrape against the metrics returned from the scraped target, no `__meta*` labels are present. ([docs](https://grafana.com/docs/alloy/latest/reference/components/prometheus/prometheus.relabel/#rule-block)) |
+| kubeDNS.maxCacheSize | string | `nil` | Sets the max_cache_size for cadvisor prometheus.relabel component. This should be at least 2x-5x your largest scrape target or samples appended rate. ([docs](https://grafana.com/docs/alloy/latest/reference/components/prometheus/prometheus.relabel/#arguments)) Overrides metrics.maxCacheSize |
+| kubeDNS.metricsTuning.excludeMetrics | list | `[]` | Metrics to drop. Can use regular expressions. |
+| kubeDNS.metricsTuning.includeMetrics | list | `[]` | Metrics to keep. Can use regular expressions. An empty list means keep all. |
+| kubeDNS.scrapeInterval | string | 60s | How frequently to scrape metrics from KubeDNS Overrides metrics.scrapeInterval |
+
 ### Kube Proxy
 
 | Key | Type | Default | Description |

charts/feature-cluster-metrics/templates/_api_server.alloy.tpl

@@ -1,9 +1,3 @@
-{{ define "feature.clusterMetrics.apiServer.allowList" }}
-{{ if .Values.apiServer.metricsTuning.includeMetrics }}
-{{ .Values.apiServer.metricsTuning.includeMetrics | toYaml }}
-{{ end }}
-{{ end }}
-
 {{- define "feature.clusterMetrics.apiServer.alloy" }}
 {{- if or .Values.apiServer.enabled (and .Values.controlPlane.enabled (not (eq .Values.apiServer.enabled false))) }}
 {{- $metricAllowList := .Values.apiServer.metricsTuning.includeMetrics }}
@@ -17,8 +11,8 @@ kubernetes.apiserver "scrape" {
 {{- if $metricDenyList }}
   drop_metrics = {{ $metricDenyList | join "|" | quote }}
 {{- end }}
-  scrape_interval = {{ .Values.cadvisor.scrapeInterval | default .Values.global.scrapeInterval | int }}
-  max_cache_size = {{ .Values.cadvisor.maxCacheSize | default .Values.global.maxCacheSize | int }}
+  scrape_interval = {{ .Values.apiServer.scrapeInterval | default .Values.global.scrapeInterval | quote }}
+  max_cache_size = {{ .Values.apiServer.maxCacheSize | default .Values.global.maxCacheSize | int }}
 {{- if .Values.apiServer.extraMetricProcessingRules }}
   forward_to = [prometheus.relabel.apiServer.receiver]
 }

charts/feature-cluster-metrics/templates/_kepler.alloy.tpl

@@ -1,5 +1,5 @@
 {{ define "feature.clusterMetrics.kepler.allowList" }}
-{{ if .Values.cadvisor.metricsTuning.useDefaultAllowList }}
+{{ if .Values.kepler.metricsTuning.useDefaultAllowList }}
 {{ "default-allow-lists/kepler.yaml" | .Files.Get }}
 {{ end }}
 {{ if .Values.kepler.metricsTuning.includeMetrics }}

charts/feature-cluster-metrics/templates/_kube_dns.alloy.tpl

@@ -0,0 +1,27 @@
+{{- define "feature.clusterMetrics.kubeDNS.alloy" }}
+{{- if or .Values.kubeDNS.enabled (and .Values.controlPlane.enabled (not (eq .Values.kubeDNS.enabled false))) }}
+{{- $metricAllowList := .Values.kubeDNS.metricsTuning.includeMetrics }}
+{{- $metricDenyList := .Values.kubeDNS.metricsTuning.excludeMetrics }}
+
+kubernetes.kube_dns "scrape" {
+  clustering = true
+{{- if $metricAllowList }}
+  keep_metrics = "up|{{ $metricAllowList | join "|" }}"
+{{- end }}
+{{- if $metricDenyList }}
+  drop_metrics = {{ $metricDenyList | join "|" | quote }}
+{{- end }}
+  scrape_interval = {{ .Values.kubeDNS.scrapeInterval | default .Values.global.scrapeInterval | quote }}
+  max_cache_size = {{ .Values.kubeDNS.maxCacheSize | default .Values.global.maxCacheSize | int }}
+{{- if .Values.kubeDNS.extraMetricProcessingRules }}
+  forward_to = [prometheus.relabel.kube_dns.receiver]
+}
+
+prometheus.relabel "kube_dns" {
+  max_cache_size = {{ .Values.kubeDNS.maxCacheSize | default .Values.global.maxCacheSize | int }}
+  {{ .Values.kubeDNS.extraMetricProcessingRules | indent 2 }}
+{{- end }}
+  forward_to = argument.metrics_destinations.value
+}
+{{- end }}
+{{- end }}

charts/feature-cluster-metrics/templates/_module.alloy.tpl

@@ -1,17 +1,25 @@
 {{- define "feature.clusterMetrics.module" }}
+{{- $includeKubernetesModule := false }}
+{{- $includeKubernetesModule = or $includeKubernetesModule .Values.cadvisor.enabled }}
+{{- $includeKubernetesModule = or $includeKubernetesModule .Values.kubelet.enabled }}
+{{- $includeKubernetesModule = or $includeKubernetesModule .Values.kubeletResource.enabled }}
+{{- $includeKubernetesModule = or $includeKubernetesModule .Values.apiServer.enabled }}
+{{- $includeKubernetesModule = or $includeKubernetesModule .Values.kubeDNS.enabled }}
+{{- $includeKubernetesModule = or $includeKubernetesModule (and .Values.controlPlane.enabled (or (not (eq .Values.apiServer.enabled false)) (not (eq .Values.kubeDNS.enabled false)))) }}
 declare "cluster_metrics" {
   argument "metrics_destinations" {
     comment = "Must be a list of metric destinations where collected metrics should be forwarded to"
   }
 
-  {{- if or .Values.cadvisor.enabled .Values.kubelet.enabled .Values.kubeletResource.enabled (or .Values.apiServer.enabled (and .Values.controlPlane.enabled (not (eq .Values.apiServer.enabled false)))) }}
+  {{- if $includeKubernetesModule }}
   {{- include "alloyModules.load" (deepCopy $ | merge (dict "name" "kubernetes" "path" "modules/kubernetes/core/metrics.alloy")) | nindent 2 }}
   {{- end }}
   {{- include "feature.clusterMetrics.kubelet.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.kubeletResource.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.cadvisor.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.apiServer.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.kubeControllerManager.alloy" . | indent 2 }}
+  {{- include "feature.clusterMetrics.kubeDNS.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.kubeProxy.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.kubeScheduler.alloy" . | indent 2 }}
   {{- include "feature.clusterMetrics.kube_state_metrics.alloy" . | indent 2 }}