Deploy monitoring stack

2026-05-18 21:42:32 -05:00
parent dbf0b18f07
commit 2ed70b4964
8 changed files with 386 additions and 25 deletions
--- a/cluster/applications/monitoring/values.yaml
+++ b/cluster/applications/monitoring/values.yaml
@@ -0,0 +1,170 @@
+# ------------------------------------------------------------------------------
+# kube-prometheus-stack — Helm Values
+# Chart: https://prometheus-community.github.io/helm-charts
+# Version: 70.4.2
+# ------------------------------------------------------------------------------
+
+# ─── Global ───────────────────────────────────────────────────────────────────
+fullnameOverride: ""
+
+# ─── Alertmanager ─────────────────────────────────────────────────────────────
+# Deployed but unconfigured — wire into n8n later
+alertmanager:
+  enabled: true
+  alertmanagerSpec:
+    storage:
+      volumeClaimTemplate:
+        spec:
+          storageClassName: nfs-emporium
+          accessModes:
+            - ReadWriteOnce
+          resources:
+            requests:
+              storage: 5Gi
+
+# ─── Prometheus ───────────────────────────────────────────────────────────────
+prometheus:
+  prometheusSpec:
+    # Retention
+    retention: 30d
+    retentionSize: "45GB"
+
+    # Storage — NFS via csi-driver-nfs
+    storageSpec:
+      volumeClaimTemplate:
+        spec:
+          storageClassName: nfs-emporium
+          accessModes:
+            - ReadWriteOnce
+          resources:
+            requests:
+              storage: 50Gi
+
+    # Scrape all ServiceMonitors and PodMonitors regardless of namespace
+    serviceMonitorSelectorNilUsesHelmValues: false
+    podMonitorSelectorNilUsesHelmValues: false
+    ruleSelectorNilUsesHelmValues: false
+
+    # Resource limits — fastpass workers have reasonable headroom
+    resources:
+      requests:
+        cpu: 200m
+        memory: 512Mi
+      limits:
+        cpu: 1000m
+        memory: 2Gi
+
+# ─── Grafana ──────────────────────────────────────────────────────────────────
+grafana:
+  enabled: true
+
+  # Stateless — dashboards managed as ConfigMaps, no PVC needed
+  persistence:
+    enabled: false
+
+  # Admin credentials via ESO from 1Password
+  admin:
+    existingSecret: grafana-admin-secret
+    userKey: admin-user
+    passwordKey: admin-password
+
+  # Grafana URL — used for redirects and OAuth callbacks
+  grafana.ini:
+    server:
+      root_url: https://grafana.local.mk-labs.cloud
+    auth:
+      disable_login_form: false
+
+  # Default dashboards
+  defaultDashboardsEnabled: true
+  defaultDashboardsTimezone: browser
+
+  # Node Exporter Full — dashboard 1860
+  dashboardProviders:
+    dashboardproviders.yaml:
+      apiVersion: 1
+      providers:
+        - name: default
+          orgId: 1
+          folder: ""
+          type: file
+          disableDeletion: false
+          editable: true
+          options:
+            path: /var/lib/grafana/dashboards/default
+
+  dashboards:
+    default:
+      node-exporter-full:
+        gnetId: 1860
+        revision: 37
+        datasource: Prometheus
+      kubernetes-cluster:
+        gnetId: 7249
+        revision: 1
+        datasource: Prometheus
+      kubernetes-pods:
+        gnetId: 6336
+        revision: 1
+        datasource: Prometheus
+
+  # Sidecar for ConfigMap-based dashboards
+  sidecar:
+    dashboards:
+      enabled: true
+      searchNamespace: ALL
+      label: grafana_dashboard
+      labelValue: "1"
+
+  resources:
+    requests:
+      cpu: 100m
+      memory: 256Mi
+    limits:
+      cpu: 500m
+      memory: 512Mi
+
+# ─── Node Exporter ────────────────────────────────────────────────────────────
+nodeExporter:
+  enabled: true
+
+prometheus-node-exporter:
+  tolerations:
+    - key: node-role.kubernetes.io/control-plane
+      operator: Exists
+      effect: NoSchedule
+
+# ─── kube-state-metrics ───────────────────────────────────────────────────────
+kubeStateMetrics:
+  enabled: true
+
+# ─── Prometheus Operator ──────────────────────────────────────────────────────
+prometheusOperator:
+  enabled: true
+
+# ─── Component scraping ───────────────────────────────────────────────────────
+kubeApiServer:
+  enabled: true
+
+kubelet:
+  enabled: true
+
+kubeControllerManager:
+  enabled: true
+
+kubeScheduler:
+  enabled: true
+
+kubeProxy:
+  enabled: false  # Cilium replaces kube-proxy
+
+kubeEtcd:
+  enabled: true
+  endpoints:
+    - 10.1.71.66
+    - 10.1.71.67
+    - 10.1.71.68
+  service:
+    enabled: true
+    port: 2381
+    targetPort: 2381