feat: Bump gpu-operator to v24.6.2

licenses.d2iq.yaml

@@ -385,32 +385,32 @@ resources:
       - license_path: LICENSE
         ref: ${image_tag}
         url: https://github.com/stakater/Reloader
-  - container_image: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia/cloud-native/gpu-operator-validator:v24.3.0-d2iq.0
+  - container_image: nvcr.io/nvidia/cloud-native/gpu-operator-validator:v24.6.2
     sources:
       - license_path: validator/LICENSE
-        ref: ${image_tag%-d2iq.0}
+        ref: ${image_tag}
         url: https://github.com/NVIDIA/gpu-operator
-  - container_image: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia/gpu-operator:v24.3.0-d2iq.0
+  - container_image: nvcr.io/nvidia/gpu-operator:v24.6.2
     sources:
       - license_path: LICENSE
-        ref: ${image_tag%-d2iq.0}
+        ref: ${image_tag}
         url: https://github.com/NVIDIA/gpu-operator
-  - container_image: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia/k8s-device-plugin:v0.15.0-ubi8-d2iq.0
+  - container_image: nvcr.io/nvidia/k8s-device-plugin:v0.16.2
     sources:
       - license_path: LICENSE
-        ref: ${image_tag%-ubi8-d2iq.0}
+        ref: ${image_tag%}
         url: https://github.com/NVIDIA/k8s-device-plugin
-  - container_image: nvcr.io/nvidia/k8s/container-toolkit:v1.15.0-ubuntu20.04
+  - container_image: nvcr.io/nvidia/k8s/container-toolkit:v1.16.2-ubuntu20.04
     sources:
       - license_path: LICENSE
         ref: ${image_tag%-ubuntu20.04}
         url: https://github.com/NVIDIA/nvidia-container-toolkit
-  - container_image: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia/k8s/container-toolkit:v1.15.0-ubi8-d2iq.0
+  - container_image: nvcr.io/nvidia/k8s/container-toolkit:v1.16.2-ubi8
     sources:
       - license_path: LICENSE
-        ref: ${image_tag%-ubi8-d2iq.0}
+        ref: ${image_tag%-ubi8}
         url: https://github.com/NVIDIA/nvidia-container-toolkit
-  - container_image: nvcr.io/nvidia/k8s/dcgm-exporter:3.3.5-3.4.1-ubuntu22.04
+  - container_image: nvcr.io/nvidia/k8s/dcgm-exporter:3.3.7-3.5.0-ubuntu22.04
     sources:
       - license_path: LICENSE
         ref: ${image_tag%-ubuntu22.04}
@@ -553,7 +553,7 @@ resources:
       - url: https://github.com/fluent/fluentd
         ref: ${image_tag%-full-build.122}
         license_path: LICENSE
-  - container_image: nvcr.io/nvidia/cloud-native/dcgm:3.3.5-1-ubuntu22.04
+  - container_image: nvcr.io/nvidia/cloud-native/dcgm:3.3.7-1-ubuntu22.04
     sources:
       - url: https://github.com/NVIDIA/DCGM
         ref: v${image_tag%-1-ubuntu22.04}

services/nvidia-gpu-operator/24.3.2/defaults/cm.yaml → services/nvidia-gpu-operator/24.6.2/defaults/cm.yaml

@@ -1,7 +1,7 @@
 apiVersion: v1
 kind: ConfigMap
 metadata:
-  name: nvidia-gpu-operator-24.3.2-d2iq-defaults
+  name: nvidia-gpu-operator-24.6.2-d2iq-defaults
   namespace: ${releaseNamespace}
 data:
   values.yaml: |
@@ -13,29 +13,29 @@ data:
       config:
         # Create a ConfigMap (default: false)
         create: false
-      repository: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia
-      version: v0.15.0-ubi8-d2iq.0
+      repository: nvcr.io/nvidia
+      version: v0.16.2
     toolkit:
       # toolkit needs to be set on per OS
       # see: https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/getting-started.html#bare-metal-passthrough-with-default-configurations-on-centos
       # this comment explains the dependency on the hosts
       # version of libc.so
       # https://github.com/NVIDIA/gpu-operator/issues/72#issuecomment-742023528
-      version: v1.15.0-ubuntu20.04
+      version: v1.16.2-ubuntu20.04
     gfd:
       # gfd is no longer published a standalone helm chart or image and instead uses
       # the k8s-device-plugin image.
       enabled: true
-      version: v0.15.0-ubi8
+      version: v0.16.2-ubi8
     dcgm:
       enabled: true
-      version: 3.3.5-1-ubuntu22.04
+      version: 3.3.7-1-ubuntu22.04
     dcgmExporter:
       enabled: true
-      version: 3.3.5-3.4.1-ubuntu22.04
+      version: 3.3.7-3.5.0-ubuntu22.04
     validator:
-      repository: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia/cloud-native
-      version: v24.3.0-d2iq.0
+      repository: nvcr.io/nvidia/cloud-native
+      version: v24.6.2
     operator:
-      repository: ghcr.io/mesosphere/dkp-container-images/nvcr.io/nvidia
-      version: v24.3.0-d2iq.0
+      repository: nvcr.io/nvidia
+      version: v24.6.2

services/nvidia-gpu-operator/24.3.2/helmrelease.yaml → services/nvidia-gpu-operator/24.6.2/helmrelease.yaml

@@ -9,7 +9,7 @@ spec:
   wait: true
   interval: 6h
   retryInterval: 1m
-  path: ./services/nvidia-gpu-operator/24.3.2/helmrelease
+  path: ./services/nvidia-gpu-operator/24.6.2/helmrelease
   sourceRef:
     kind: GitRepository
     name: management

services/nvidia-gpu-operator/24.6.2/helmrelease/extra-images.txt

@@ -0,0 +1,7 @@
+nvcr.io/nvidia/k8s/container-toolkit:{{ regexReplaceAllLiteral "-.+$" .Values.toolkit.version "" }}-ubuntu20.04
+nvcr.io/nvidia/k8s/container-toolkit:{{ regexReplaceAllLiteral "-.+$" .Values.toolkit.version "" }}-ubi8
+nvcr.io/nvidia/cloud-native/gpu-operator-validator:{{ .Values.validator.version }}
+nvcr.io/nvidia/cloud-native/dcgm:{{ .Values.dcgm.version }}
+nvcr.io/nvidia/k8s/dcgm-exporter:{{ .Values.dcgmExporter.version }}
+nvcr.io/nvidia/k8s-device-plugin:{{ .Values.devicePlugin.version }}
+nvcr.io/nvidia/k8s/cuda-sample:vectoradd-cuda12.5.0

services/nvidia-gpu-operator/24.3.2/helmrelease/nvidia.yaml → services/nvidia-gpu-operator/24.6.2/helmrelease/nvidia.yaml

@@ -11,7 +11,7 @@ spec:
         kind: HelmRepository
         name: helm.ngc.nvidia.com-nvidia
         namespace: kommander-flux
-      version: v24.3.0
+      version: v24.6.2
   interval: 15s
   install:
     crds: CreateReplace
@@ -24,5 +24,5 @@ spec:
   releaseName: nvidia-gpu-operator
   valuesFrom:
     - kind: ConfigMap
-      name: nvidia-gpu-operator-24.3.2-d2iq-defaults
+      name: nvidia-gpu-operator-24.6.2-d2iq-defaults
   targetNamespace: ${releaseNamespace}