[WIP] Deploy Unified Grafana Instance on the utility cluster

kubernetes/apps/monitoring.yaml

@@ -0,0 +1,68 @@
+apiVersion: argoproj.io/v1alpha1
+kind: ApplicationSet
+metadata:
+  name: monitoring
+  annotations:
+    argocd.argoproj.io/sync-wave: "-3"
+spec:
+  goTemplate: true
+  generators:
+    - clusters:
+        selector:
+          matchLabels:
+            clusterType: 'utility'
+  template:
+    metadata:
+      name: 'monitoring-{{ .name }}'
+      annotations:
+        argocd.argoproj.io/compare-options: ServerSideDiff=true,IncludeMutationWebhook=true
+    spec:
+      destination:
+        namespace: monitoring
+        server: "{{ .server }}"
+      project: default
+      sources:
+        - chart: kube-prometheus-stack
+          repoURL: https://prometheus-community.github.io/helm-charts
+          targetRevision: 62.3.1
+          helm:
+            releaseName: monitoring
+            valueFiles:
+              - $values/kubernetes/{{ .name }}/helm/monitoring.yaml
+        - repoURL: https://github.com/borg-land/k8s.io
+          targetRevision: observability-stack
+          ref: values
+        - path: kubernetes/{{ .name }}/monitoring/
+          repoURL: https://github.com/borg-land/k8s.io
+          targetRevision: observability-stack
+      syncPolicy:
+        automated:
+          prune: true
+          selfHeal: true
+        syncOptions:
+          - CreateNamespace=true
+          - ServerSideApply=true
+        managedNamespaceMetadata:
+          labels:
+            istio-injection: enabled
+---
+apiVersion: argoproj.io/v1alpha1
+kind: Application
+metadata:
+  name: thanos
+spec:
+  destination:
+    name: gke-utility
+    namespace: monitoring
+  project: default
+  sources:
+    - chart: thanos
+      repoURL: registry-1.docker.io/bitnamicharts
+      targetRevision: 15.7.27
+      helm:
+        releaseName: thanos
+        valueFiles:
+          - $values/kubernetes/gke-utility//helm/thanos.yaml
+    - repoURL: https://github.com/borg-land/k8s.io
+      targetRevision: observability-stack
+      ref: values

kubernetes/gke-utility/helm/cert-manager.yaml

@@ -1,8 +1,7 @@
 crds:
   enabled: true
 extraObjects:
-  - |
-    apiVersion: cert-manager.io/v1
+  - apiVersion: cert-manager.io/v1
     kind: ClusterIssuer
     metadata:
       name: letsencrypt-prod

kubernetes/gke-utility/helm/monitoring.yaml

@@ -0,0 +1,54 @@
+grafana:
+  enabled: true
+  sidecar:
+    datasources:
+      # point the default Promtheus Datasource to Thanos
+      url: http://thanos-query-frontend.monitoring:9090/
+  grafana.ini:
+    analytics:
+      reporting_enabled: false
+      check_for_updates: true
+    server:
+      root_url: https://monitoring.prow.k8s.io
+    auth:
+      oauth_auto_login: true
+      signout_redirect_url: https://oauth2-proxy.k8s.io/oauth2/sign_out
+    auth.proxy:
+      enabled: true
+      header_name: X-Auth-Request-Email
+      header_property: email
+      auto_sign_up: true
+      headers: Name:X-Auth-Request-User Email:X-Auth-Request-Email Groups:X-Auth-Request-Groups
+    users:
+      allow_sign_up: false
+      auto_assign_org: true
+      auto_assign_org_role: Editor
+
+prometheus:
+  thanosService:
+    enabled: true
+  prometheusSpec:
+    externalLabels:
+      cluster: gke-utility
+      cloud: gke
+    thanos:
+      image: quay.io/thanos/thanos:v0.36.1
+      listenLocal: false
+      logLevel: debug
+      objectStorageConfig:
+        secret:
+          type: GCS
+          config:
+            bucket: k8s-io-thanos
+    serviceMonitorSelectorNilUsesHelmValues: false
+    podMonitorSelectorNilUsesHelmValues: false
+    ruleSelectorNilUsesHelmValues: false
+    probeSelectorNilUsesHelmValues: false
+
+# These endpoints aren't scrapable on GKE
+coreDns:
+  enabled: false
+kubeControllerManager:
+  enabled: false
+kubeScheduler:
+  enabled: false

kubernetes/gke-utility/helm/thanos.yaml

@@ -0,0 +1,28 @@
+objstoreConfig: |-
+  type: GCS
+  config:
+    bucket: k8s-io-thanos
+storegateway:
+  enabled: true
+query:
+  dnsDiscovery:
+    sidecarsService: monitoring-kube-prometheus-thanos-discovery
+    sidecarsNamespace: monitoring
+bucketweb:
+  enabled: true
+compactor:
+  enabled: true
+ruler:
+  enabled: true
+  alertmanagers:
+    - http://monitoring-kube-prometheus-alertmanager.monitoring.svc.cluster.local:9093
+  config: |-
+    groups:
+      - name: "metamonitoring"
+        rules:
+          - alert: "PrometheusDown"
+            expr: absent(up{prometheus="monitoring/kube-prometheus"})
+metrics:
+  enabled: true
+  serviceMonitor:
+    enabled: true

kubernetes/gke-utility/monitoring/httproute.yaml

@@ -0,0 +1,15 @@
+apiVersion: gateway.networking.k8s.io/v1
+kind: HTTPRoute
+metadata:
+  name: grafana
+spec:
+  hostnames:
+    - monitoring.prow.k8s.io
+  parentRefs:
+    - name: istio-ingressgateway
+      namespace: istio-system
+      sectionName: https
+  rules:
+    - backendRefs:
+        - name: monitoring-grafana
+          port: 80

kubernetes/gke-utility/monitoring/kustomization.yaml

@@ -0,0 +1,5 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+namespace: monitoring
+resources:
+  - httproute.yaml