downgrade nvidia driver to work with 12.8 cuda[ci skip]

modules/kubernetes/nvidia/main.tf

@@ -49,7 +49,7 @@ resource "helm_release" "nvidia-gpu-operator" {
   #   version    = "0.9.3"
   timeout = 6000
 
-  # values = [templatefile("${path.module}/values.yaml", {})]
+  values = [templatefile("${path.module}/values.yaml", {})]
 }
 
 resource "kubernetes_deployment" "nvidia-exporter" {

modules/kubernetes/nvidia/values.yaml

@@ -1,48 +1,15 @@
-# values-cuda12.8.yaml
 driver:
   enabled: true
-  repository: nvcr.io/nvidia/driver
+  # repository: nvcr.io/nvidia/driver
   # choose a driver version compatible with your GPU + CUDA 12.x (example)
-  version: "535.113.01"
-
-toolkit:
-  enabled: true
-  image:
-    repository: nvcr.io/nvidia/cuda
-    # use a CUDA 12.8 toolkit image; adjust OS tag as needed (e.g. ubuntu22.04, ubuntu20.04)
-    tag: "12.8.1-devel-ubuntu22.04"
-  # (Optional) If you have containerd, you might need environment for containerd config
-  # env:
-  #   - name: CONTAINERD_CONFIG
-  #     value: /etc/containerd/config.toml
-  #   - name: CONTAINERD_SOCKET
-  #     value: /run/containerd/containerd.sock
-  #   - name: RUNTIME_CONFIG_SOURCE
-  #     value: "command, file"
-
-devicePlugin:
-  enabled: true
-  image:
-    repository: nvcr.io/nvidia/k8s-device-plugin
-    # pick a device-plugin build compatible with CUDA 12.8
-    tag: "1.14.0-12.8-ubuntu22.04"
-
-dcgmExporter:
-  enabled: true
-  image:
-    repository: nvcr.io/nvidia/k8s/dcgm-exporter
-    tag: "2.12.13-12.8-ubuntu22.04"
-
-# Optional: if you use MIG or want MIG Manager
-# migManager:
-#   enabled: false
-
-# If you want GPU‑Feature‑Discovery (GFD) — optional
-# gfd:
-#   enabled: false
-
-# Other optional overrides (keep defaults or adjust as needed)
-# ccManager:
-#   enabled: false
-# cdi:
-#   enabled: true
+  # NVIDIA GPU driver - https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/platform-support.html#known-issue
+  # https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/
+  # 13.x >= 580
+  # 12.x >= 525, <580
+  # 11.x >= 450, <525
+  #
+  # Delete the cluster policy before each change
+  # version: "575.57.08" # CUDA 12.9
+  version: "570.195.03" # CUDA 12.8
+  upgradePolicy:
+    autoUpgrade: false