observability/prometheus/helm-values.yaml

# Kube-Prometheus-Stack Helm Values
# Chart: https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack
# Includes: Prometheus Operator, Prometheus, Alertmanager, Grafana, and various exporters

# Global settings
fullnameOverride: ""

# CRD 관리
crds:
  enabled: false  # CRD annotation 크기 문제로 비활성화 (CRD는 이미 설치됨)

# Prometheus Operator
prometheusOperator:
  enabled: true
  # CRD 생성 비활성화
  createCustomResource: false

# Kubelet ServiceMonitor with cluster label
kubelet:
  enabled: true
  serviceMonitor:
    # cAdvisor metrics (container_memory_working_set_bytes 등)
    cAdvisorRelabelings:
      - targetLabel: cluster
        replacement: "mayne-cluster"
      - sourceLabels: [__metrics_path__]
        targetLabel: metrics_path
    # Resource metrics
    resourceRelabelings:
      - targetLabel: cluster
        replacement: "mayne-cluster"
    # Probes metrics
    probesRelabelings:
      - targetLabel: cluster
        replacement: "mayne-cluster"
  
# Prometheus
prometheus:
  enabled: true

  # Thanos Sidecar - for long-term storage and HA
  thanosService:
    enabled: true
  thanosServiceMonitor:
    enabled: true

  prometheusSpec:
    # HA: 2 replicas on different worker nodes
    replicas: 2
    replicaExternalLabelName: prometheus_replica

    # Pod anti-affinity for HA
    affinity:
      podAntiAffinity:
        preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 100
            podAffinityTerm:
              labelSelector:
                matchLabels:
                  app.kubernetes.io/name: prometheus
              topologyKey: kubernetes.io/hostname

    scrapeInterval: 60s       # 30s → 60s (메모리 절감)
    evaluationInterval: 60s   # 30s → 60s
    retention: 3d             # Local retention (S3 has longer retention via Thanos)

    # Thanos Sidecar configuration
    thanos:
      image: quay.io/thanos/thanos:v0.37.2
      objectStorageConfig:
        existingSecret:
          name: thanos-objstore-secret
          key: objstore.yml

    storageSpec:
      volumeClaimTemplate:
        spec:
          storageClassName: local-path
          accessModes: ["ReadWriteOnce"]
          resources:
            requests:
              storage: 5Gi

    resources:
      requests:
        cpu: 50m
        memory: 512Mi
      limits:
        memory: 512Mi
    
    # ServiceMonitor 자동 발견 - 모든 ServiceMonitor 선택
    serviceMonitorSelectorNilUsesHelmValues: false
    serviceMonitorSelector: {}
    podMonitorSelectorNilUsesHelmValues: false
    podMonitorSelector: {}
    probeSelectorNilUsesHelmValues: false
    ruleSelector: {}
    
    # Alertmanager 설정
    alertingEndpoints:
      - name: alertmanager
        namespace: alertmanager
        port: http
        scheme: http

    # External labels added to all metrics
    externalLabels:
      cluster: "mayne-cluster"

    additionalScrapeConfigs:
      # ArgoCD metrics
      - job_name: 'argocd-metrics'
        static_configs:
          - targets:
            - 'argocd-application-controller-metrics.argocd.svc.cluster.local:8082'
            labels:
              service: argocd-controller
          - targets:
            - 'argocd-server-metrics.argocd.svc.cluster.local:8083'
            labels:
              service: argocd-server
          - targets:
            - 'argocd-repo-server-metrics.argocd.svc.cluster.local:8084'
            labels:
              service: argocd-repo

      # Cert-Manager
      - job_name: 'cert-manager'
        static_configs:
          - targets:
            - 'cert-manager.cert-manager.svc.cluster.local:9402'

      # MinIO
      - job_name: 'minio-cluster'
        static_configs:
          - targets:
            - 'minio.minio.svc.cluster.local:9000'
        metrics_path: /minio/v2/metrics/cluster
        scheme: http

      - job_name: 'minio-node'
        static_configs:
          - targets:
            - 'minio.minio.svc.cluster.local:9000'
        metrics_path: /minio/v2/metrics/node
        scheme: http

      # Ingress NGINX
      - job_name: 'ingress-nginx'
        kubernetes_sd_configs:
          - role: pod
            namespaces:
              names:
                - ingress-nginx
        relabel_configs:
          - source_labels: [__meta_kubernetes_pod_label_app_kubernetes_io_name]
            action: keep
            regex: ingress-nginx
          - source_labels: [__meta_kubernetes_pod_label_app_kubernetes_io_component]
            action: keep
            regex: controller
          - source_labels: [__address__]
            action: replace
            regex: ([^:]+)(?::\d+)?
            replacement: $1:10254
            target_label: __address__
          - source_labels: [__meta_kubernetes_pod_name]
            action: replace
            target_label: pod
          - source_labels: [__meta_kubernetes_namespace]
            action: replace
            target_label: namespace

# API Server 메트릭 수집 비활성화 (메모리 절감: ~37k series)
kubeApiServer:
  enabled: false

# etcd 메트릭 수집 비활성화 (메모리 절감: ~26k series)
kubeEtcd:
  enabled: false

# Alertmanager (기존 alertmanager 사용)
alertmanager:
  enabled: false

# Grafana (기존 grafana 사용)
grafana:
  enabled: false

# Node Exporter (기존 node-exporter 사용)
nodeExporter:
  enabled: false
  
# Kube State Metrics (기존 kube-state-metrics 사용)
# 별도로 배포된 kube-state-metrics가 자체 ServiceMonitor 생성
kubeStateMetrics:
  enabled: false

# 기본 ServiceMonitors
defaultRules:
  create: true
  # 비활성화할 개별 alert rule
  disabled:
    KubeCPUOvercommit: true
    KubeMemoryOvercommit: true
  rules:
    alertmanager: true
    etcd: false
    configReloaders: true
    general: true
    k8s: true
    kubeApiserverAvailability: true
    kubeApiserverBurnrate: true
    kubeApiserverHistogram: true
    kubeApiserverSlos: true
    kubeControllerManager: false
    kubelet: true
    kubeProxy: false
    kubePrometheusGeneral: true
    kubePrometheusNodeRecording: true
    kubernetesApps: true
    kubernetesResources: true
    kubernetesStorage: true
    kubernetesSystem: true
    kubeSchedulerAlerting: false
    kubeSchedulerRecording: false
    kubeStateMetrics: true
    network: true
    node: true
    nodeExporterAlerting: true
    nodeExporterRecording: true
    prometheus: true
    prometheusOperator: true