♻️ intel-device-plugin

kubernetes/apps/kube-system/intel-device-plugin/app/helmrelease.yaml

@@ -0,0 +1,32 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/fluxcd-community/flux2-schemas/main/helmrelease-helm-v2beta1.json
+apiVersion: helm.toolkit.fluxcd.io/v2beta1
+kind: HelmRelease
+metadata:
+  name: intel-device-plugin-operator
+  namespace: kube-system
+spec:
+  interval: 30m
+  chart:
+    spec:
+      chart: intel-device-plugins-operator
+      version: 0.27.1
+      sourceRef:
+        kind: HelmRepository
+        name: intel
+        namespace: flux-system
+  maxHistory: 2
+  install:
+    crds: CreateReplace
+    remediation:
+      retries: 3
+  upgrade:
+    cleanupOnFail: true
+    crds: CreateReplace
+    remediation:
+      retries: 3
+  uninstall:
+    keepHistory: false
+  dependsOn:
+    - name: node-feature-discovery
+      namespace: kube-system

kubernetes/apps/kube-system/intel-device-plugin/exporter/helmrelease.yaml

@@ -51,20 +51,13 @@ spec:
                 targetLabel: node
     securityContext:
       privileged: true
-    affinity:
+    nodeSelector:
-      nodeAffinity:
+      intel.feature.node.kubernetes.io/gpu: "true"
-        requiredDuringSchedulingIgnoredDuringExecution:
-          nodeSelectorTerms:
-            - matchExpressions:
-                - key: feature.node.kubernetes.io/custom-intel-gpu
-                  operator: In
-                  values:
-                    - "true"
     resources:
       requests:
-        gpu.intel.com/i915: 1
+        gpu.intel.com/i915_monitoring: 1
-        cpu: 15m
+        cpu: 100m
-        memory: 105Mi
+        memory: 100Mi
       limits:
-        gpu.intel.com/i915: 1
+        gpu.intel.com/i915_monitoring: 1
-        memory: 105Mi
+        memory: 500Mi

kubernetes/apps/kube-system/intel-device-plugin/gpu/helmrelease.yaml

@@ -0,0 +1,34 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/fluxcd-community/flux2-schemas/main/helmrelease-helm-v2beta1.json
+apiVersion: helm.toolkit.fluxcd.io/v2beta1
+kind: HelmRelease
+metadata:
+  name: intel-device-plugin-gpu
+  namespace: kube-system
+spec:
+  interval: 30m
+  chart:
+    spec:
+      chart: intel-device-plugins-gpu
+      version: 0.27.1
+      sourceRef:
+        kind: HelmRepository
+        name: intel
+        namespace: flux-system
+  maxHistory: 2
+  install:
+    remediation:
+      retries: 3
+  upgrade:
+    cleanupOnFail: true
+    remediation:
+      retries: 3
+  uninstall:
+    keepHistory: false
+  dependsOn:
+    - name: intel-device-plugin-operator
+      namespace: kube-system
+  values:
+    name: intel-device-plugin-gpu
+    sharedDevNum: 4
+    nodeFeatureRule: true

kubernetes/apps/kube-system/intel-device-plugin/gpu/kustomization.yaml

@@ -0,0 +1,7 @@
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/SchemaStore/schemastore/master/src/schemas/json/kustomization.json
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+namespace: kube-system
+resources:
+  - ./helmrelease.yaml

kubernetes/apps/kube-system/intel-device-plugin/ks.yaml

@@ -3,48 +3,52 @@
 apiVersion: kustomize.toolkit.fluxcd.io/v1
 kind: Kustomization
 metadata:
-  name: cluster-apps-intel-gpu-plugin
+  name: cluster-apps-intel-device-plugin
   namespace: flux-system
   labels:
     substitution.flux.home.arpa/enabled: "true"
 spec:
-  dependsOn:
+  path: ./kubernetes/apps/kube-system/intel-device-plugin/app
-    - name: cluster-apps-node-feature-discovery
-  path: ./kubernetes/apps/kube-system/intel-gpu/plugin
   prune: true
   sourceRef:
     kind: GitRepository
     name: home-ops-kubernetes
-  healthChecks:
-    - apiVersion: helm.toolkit.fluxcd.io/v2beta1
-      kind: HelmRelease
-      name: intel-gpu-plugin
-      namespace: kube-system
   interval: 30m
   retryInterval: 1m
-  timeout: 3m
+  timeout: 5m
 ---
 # yaml-language-server: $schema=https://raw.githubusercontent.com/fluxcd-community/flux2-schemas/main/kustomization-kustomize-v1.json
 apiVersion: kustomize.toolkit.fluxcd.io/v1
 kind: Kustomization
 metadata:
-  name: cluster-apps-intel-gpu-exporter
+  name: cluster-apps-intel-device-plugin-gpu
   namespace: flux-system
   labels:
     substitution.flux.home.arpa/enabled: "true"
 spec:
-  dependsOn:
+  path: ./kubernetes/apps/kube-system/intel-device-plugin/gpu
-    - name: cluster-apps-intel-gpu-plugin
-  path: ./kubernetes/apps/kube-system/intel-gpu/exporter
   prune: true
   sourceRef:
     kind: GitRepository
     name: home-ops-kubernetes
-  healthChecks:
-    - apiVersion: helm.toolkit.fluxcd.io/v2beta1
-      kind: HelmRelease
-      name: intel-gpu-exporter
-      namespace: kube-system
   interval: 30m
   retryInterval: 1m
-  timeout: 3m
+  timeout: 5m
+---
+# yaml-language-server: $schema=https://raw.githubusercontent.com/fluxcd-community/flux2-schemas/main/kustomization-kustomize-v1.json
+apiVersion: kustomize.toolkit.fluxcd.io/v1
+kind: Kustomization
+metadata:
+  name: cluster-apps-intel-device-plugin-exporter
+  namespace: flux-system
+  labels:
+    substitution.flux.home.arpa/enabled: "true"
+spec:
+  path: ./kubernetes/apps/kube-system/intel-device-plugin/exporter
+  prune: true
+  sourceRef:
+    kind: GitRepository
+    name: home-ops-kubernetes
+  interval: 30m
+  retryInterval: 1m
+  timeout: 5m

kubernetes/apps/kube-system/intel-gpu/plugin/helmrelease.yaml

@@ -1,83 +0,0 @@
----
-# yaml-language-server: $schema=https://raw.githubusercontent.com/fluxcd-community/flux2-schemas/main/helmrelease-helm-v2beta1.json
-apiVersion: helm.toolkit.fluxcd.io/v2beta1
-kind: HelmRelease
-metadata:
-  name: &app intel-gpu-plugin
-  namespace: kube-system
-spec:
-  interval: 30m
-  chart:
-    spec:
-      chart: app-template
-      version: 1.5.1
-      sourceRef:
-        kind: HelmRepository
-        name: bjw-s
-        namespace: flux-system
-  maxHistory: 2
-  install:
-    createNamespace: true
-    remediation:
-      retries: 3
-  upgrade:
-    cleanupOnFail: true
-    remediation:
-      retries: 3
-  uninstall:
-    keepHistory: false
-  values:
-    controller:
-      type: daemonset
-      strategy: RollingUpdate
-    image:
-      repository: docker.io/intel/intel-gpu-plugin
-      tag: 0.27.1
-      pullPolicy: IfNotPresent
-    args:
-      - -shared-dev-num
-      - "4"
-    service:
-      main:
-        enabled: false
-    # TODO(intel-gpu-plugin): Write probes to check for something to tell if it's working
-    probes:
-      liveness:
-        enabled: false
-      readiness:
-        enabled: false
-      startup:
-        enabled: false
-    persistence:
-      devfs:
-        enabled: true
-        type: hostPath
-        hostPath: /dev/dri
-        hostPathType: Directory
-        readOnly: true
-      sysfs:
-        enabled: true
-        type: hostPath
-        hostPath: /sys/class/drm
-        hostPathType: Directory
-        readOnly: true
-      kubeletsockets:
-        enabled: true
-        type: hostPath
-        hostPathType: Directory
-        hostPath: /var/lib/kubelet/device-plugins
-    affinity:
-      nodeAffinity:
-        requiredDuringSchedulingIgnoredDuringExecution:
-          nodeSelectorTerms:
-            - matchExpressions:
-                - key: feature.node.kubernetes.io/custom-intel-gpu
-                  operator: In
-                  values:
-                    - "true"
-    resources:
-      requests:
-        cpu: 15m
-        memory: 105Mi
-      limits:
-        memory: 105Mi

kubernetes/apps/kube-system/kustomization.yaml

@@ -9,7 +9,7 @@ resources:
   - ./cilium/ks.yaml
   - ./descheduler/ks.yaml
   - ./external-secrets/ks.yaml
-  - ./intel-gpu/ks.yaml
+  - ./intel-device-plugin/ks.yaml
   - ./kubelet-csr-approver/ks.yaml
   - ./metrics-server/ks.yaml
   - ./node-feature-discovery/ks.yaml

kubernetes/flux/repositories/helm/intel.yaml

@@ -0,0 +1,10 @@
+---
+# yaml-language-server: $schema=https://kubernetes-schemas.devbu.io/helmrepository_v1beta2.json
+apiVersion: source.toolkit.fluxcd.io/v1beta2
+kind: HelmRepository
+metadata:
+  name: intel
+  namespace: flux-system
+spec:
+  interval: 2h
+  url: https://intel.github.io/helm-charts

kubernetes/flux/repositories/helm/kustomization.yaml

@@ -21,6 +21,7 @@ resources:
   - ./grafana.yaml
   - ./hajimari.yaml
   - ./ingress-nginx.yaml
+  - ./intel.yaml
   - ./jetstack.yaml
   - ./kyverno.yaml
   - ./metrics-server.yaml