♻️ kube-tools

cluster/apps/kube-tools/intel-gpu-exporter/helm-release.yaml

@@ -0,0 +1,56 @@
+---
+apiVersion: helm.toolkit.fluxcd.io/v2beta1
+kind: HelmRelease
+metadata:
+  name: &app intel-gpu-exporter
+  namespace: default
+spec:
+  interval: 15m
+  chart:
+    spec:
+      chart: app-template
+      version: 0.1.1
+      sourceRef:
+        kind: HelmRepository
+        name: bjw-s-charts
+        namespace: flux-system
+  install:
+    createNamespace: true
+    remediation:
+      retries: 5
+  upgrade:
+    remediation:
+      retries: 5
+  dependsOn:
+    - name: intel-gpu-plugin
+      namespace: default
+  values:
+    controller:
+      type: daemonset
+    image:
+      repository: ghcr.io/onedr0p/intel-gpu-exporter
+      tag: rolling@sha256:1c84020b442e0f95e2e6a46281d3bfc6199902d3d91b10515fc000e43c9a7421
+    service:
+      main:
+        ports:
+          http:
+            port: 8080
+    securityContext:
+      privileged: true
+    affinity:
+      nodeAffinity:
+        requiredDuringSchedulingIgnoredDuringExecution:
+          nodeSelectorTerms:
+            - matchExpressions:
+                - key: feature.node.kubernetes.io/custom-intel-gpu
+                  operator: In
+                  values:
+                    - "true"
+    resources:
+      requests:
+        gpu.intel.com/i915: 1
+        cpu: 100m
+        memory: 100Mi
+      limits:
+        gpu.intel.com/i915: 1
+        memory: 500Mi

cluster/apps/kube-tools/intel-gpu-exporter/kustomization.yaml

@@ -0,0 +1,6 @@
+---
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+resources:
+  - helm-release.yaml
+  - service-monitor.yaml

cluster/apps/kube-tools/intel-gpu-exporter/service-monitor.yaml

@@ -0,0 +1,21 @@
+---
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: &app intel-gpu-exporter
+  namespace: default
+  labels: &labels
+    app.kubernetes.io/instance: *app
+    app.kubernetes.io/name: *app
+spec:
+  selector:
+    matchLabels:
+      <<: *labels
+  endpoints:
+    - port: http
+      interval: 1m
+      scrapeTimeout: 10s
+      path: /metrics
+      relabelings:
+        - sourceLabels: [__meta_kubernetes_pod_node_name]
+          targetLabel: node