Bump nvidia device plugin to 0.11 and nvidia driver to 510

2022-04-30 09:20:02 +02:00 · 2022-04-30 09:20:02 +02:00 · e622736992
parent 653aa6c606
commit e622736992
11 changed files with 20 additions and 18 deletions
--- a/docs/releases/1.24-NOTES.md
+++ b/docs/releases/1.24-NOTES.md
@ -27,6 +27,8 @@ kOps will directly manage the Karpenter Provisioner resources. Read more about h

 * The minimum version for the Terraform AWS Provider has been bumped to 4.0.0 to address the deprecation of the aws_s3_bucket_object resource and its replacement with the aws_s3_object resource. Such resources will be destroyed and recreated without downtime when applying the changes. 

+* ARM64 support for nvidia device driver. Nvidia nodes on ARM64 requires Ubuntu 22.04 AMIs.
+
 # Breaking changes

 ## Control plane taints and labels
--- a/pkg/apis/kops/containerdconfig.go
+++ b/pkg/apis/kops/containerdconfig.go
@ -17,7 +17,7 @@ limitations under the License.
 package kops

 // NvidiaDefaultDriverPackage is the nvidia driver default version
-const NvidiaDefaultDriverPackage = "nvidia-headless-460-server"
+const NvidiaDefaultDriverPackage = "nvidia-headless-510-server"

 // ContainerdConfig is the configuration for containerd
 type ContainerdConfig struct {
@ -45,7 +45,7 @@ type ContainerdConfig struct {

 type NvidiaGPUConfig struct {
 	// Package is the name of the nvidia driver package that will be installed.
-	// Default is "nvidia-headless-460-server".
+	// Default is "nvidia-headless-510-server".
 	DriverPackage string `json:"package,omitempty"`
 	// Enabled determines if kOps will install the Nvidia GPU runtime and drivers.
 	// They will only be installed on intances that has an Nvidia GPU.
--- a/tests/integration/update_cluster/nvidia/cloudformation.json.extracted.yaml
+++ b/tests/integration/update_cluster/nvidia/cloudformation.json.extracted.yaml
@ -133,7 +133,7 @@ Resources.AWSEC2LaunchTemplatemasterustest1amastersminimalexamplecom.Properties.
    logLevel: info
    nvidiaGPU:
      enabled: true
-      package: nvidia-headless-460-server
+      package: nvidia-headless-510-server
    version: 1.4.12
  docker:
    skipInstall: true
@ -248,7 +248,7 @@ Resources.AWSEC2LaunchTemplatemasterustest1amastersminimalexamplecom.Properties.
  ConfigBase: memfs://clusters.example.com/minimal.example.com
  InstanceGroupName: master-us-test-1a
  InstanceGroupRole: Master
-  NodeupConfigHash: XfW6vDRNRfTHJPOGeKVjmi0X8oyJNK2bnzKsVZy00K8=
+  NodeupConfigHash: Pxds5Al5CozIY+FCKHuqBqRAPIeptslqIdIdX8lPnkE=

  __EOF_KUBE_ENV

@ -389,7 +389,7 @@ Resources.AWSEC2LaunchTemplatenodesminimalexamplecom.Properties.LaunchTemplateDa
    logLevel: info
    nvidiaGPU:
      enabled: true
-      package: nvidia-headless-460-server
+      package: nvidia-headless-510-server
    version: 1.4.12
  docker:
    skipInstall: true
@ -422,7 +422,7 @@ Resources.AWSEC2LaunchTemplatenodesminimalexamplecom.Properties.LaunchTemplateDa
  ConfigBase: memfs://clusters.example.com/minimal.example.com
  InstanceGroupName: nodes
  InstanceGroupRole: Node
-  NodeupConfigHash: rXMJNOQF0/HDGwb7T9kem3+Ydeau/+I6oiA8ankZdEM=
+  NodeupConfigHash: me4mBnPl+0J3noivly6hq8f5geHok77lvs0noMjPxHQ=

  __EOF_KUBE_ENV

--- a/tests/integration/update_cluster/nvidia/data/aws_launch_template_master-us-test-1a.masters.minimal.example.com_user_data
+++ b/tests/integration/update_cluster/nvidia/data/aws_launch_template_master-us-test-1a.masters.minimal.example.com_user_data
@ -132,7 +132,7 @@ containerd:
  logLevel: info
  nvidiaGPU:
    enabled: true
-    package: nvidia-headless-460-server
+    package: nvidia-headless-510-server
  version: 1.4.12
 docker:
  skipInstall: true
@ -247,7 +247,7 @@ CloudProvider: aws
 ConfigBase: memfs://clusters.example.com/minimal.example.com
 InstanceGroupName: master-us-test-1a
 InstanceGroupRole: Master
-NodeupConfigHash: XfW6vDRNRfTHJPOGeKVjmi0X8oyJNK2bnzKsVZy00K8=
+NodeupConfigHash: Pxds5Al5CozIY+FCKHuqBqRAPIeptslqIdIdX8lPnkE=

 __EOF_KUBE_ENV

--- a/tests/integration/update_cluster/nvidia/data/aws_launch_template_nodes.minimal.example.com_user_data
+++ b/tests/integration/update_cluster/nvidia/data/aws_launch_template_nodes.minimal.example.com_user_data
@ -132,7 +132,7 @@ containerd:
  logLevel: info
  nvidiaGPU:
    enabled: true
-    package: nvidia-headless-460-server
+    package: nvidia-headless-510-server
  version: 1.4.12
 docker:
  skipInstall: true
@ -165,7 +165,7 @@ CloudProvider: aws
 ConfigBase: memfs://clusters.example.com/minimal.example.com
 InstanceGroupName: nodes
 InstanceGroupRole: Node
-NodeupConfigHash: rXMJNOQF0/HDGwb7T9kem3+Ydeau/+I6oiA8ankZdEM=
+NodeupConfigHash: me4mBnPl+0J3noivly6hq8f5geHok77lvs0noMjPxHQ=

 __EOF_KUBE_ENV

--- a/tests/integration/update_cluster/nvidia/data/aws_s3_object_cluster-completed.spec_content
+++ b/tests/integration/update_cluster/nvidia/data/aws_s3_object_cluster-completed.spec_content
@ -22,7 +22,7 @@ spec:
    logLevel: info
    nvidiaGPU:
      enabled: true
-      package: nvidia-headless-460-server
+      package: nvidia-headless-510-server
    version: 1.4.12
  dnsZone: Z1AFAKE1ZON3YO
  docker:
--- a/tests/integration/update_cluster/nvidia/data/aws_s3_object_minimal.example.com-addons-bootstrap_content
+++ b/tests/integration/update_cluster/nvidia/data/aws_s3_object_minimal.example.com-addons-bootstrap_content
@ -47,7 +47,7 @@ spec:
    version: 9.99.0
  - id: k8s-1.16
    manifest: nvidia.addons.k8s.io/k8s-1.16.yaml
-    manifestHash: cee96cdd2bdf02a76bd8ed78e56ddac52339d9f03a5722184bf5716be4342e9d
+    manifestHash: 0b5963253c1cf13686226a8893b9b9e3ffd1373d5e2fb0699588d8714a4ba78a
    name: nvidia.addons.k8s.io
    selector:
      k8s-addon: nvidia.addons.k8s.io
--- a/tests/integration/update_cluster/nvidia/data/aws_s3_object_minimal.example.com-addons-nvidia.addons.k8s.io-k8s-1.16_content
+++ b/tests/integration/update_cluster/nvidia/data/aws_s3_object_minimal.example.com-addons-nvidia.addons.k8s.io-k8s-1.16_content
@ -22,7 +22,7 @@ spec:
      containers:
      - args:
        - --fail-on-init-error=false
-        image: docker.io/nvidia/k8s-device-plugin:v0.10.0
+        image: docker.io/nvidia/k8s-device-plugin:v0.11.0
        name: nvidia-device-plugin-ctr
        securityContext:
          allowPrivilegeEscalation: false
--- a/tests/integration/update_cluster/nvidia/data/aws_s3_object_nodeupconfig-master-us-test-1a_content
+++ b/tests/integration/update_cluster/nvidia/data/aws_s3_object_nodeupconfig-master-us-test-1a_content
@ -255,7 +255,7 @@ KubeletConfig:
  shutdownGracePeriodCriticalPods: 10s
 NvidiaGPU:
  enabled: true
-  package: nvidia-headless-460-server
+  package: nvidia-headless-510-server
 UpdatePolicy: automatic
 channels:
 - memfs://clusters.example.com/minimal.example.com/addons/bootstrap-channel.yaml
@ -263,7 +263,7 @@ containerdConfig:
  logLevel: info
  nvidiaGPU:
    enabled: true
-    package: nvidia-headless-460-server
+    package: nvidia-headless-510-server
  version: 1.4.12
 etcdManifests:
 - memfs://clusters.example.com/minimal.example.com/manifests/etcd/main.yaml
--- a/tests/integration/update_cluster/nvidia/data/aws_s3_object_nodeupconfig-nodes_content
+++ b/tests/integration/update_cluster/nvidia/data/aws_s3_object_nodeupconfig-nodes_content
@ -61,7 +61,7 @@ KubeletConfig:
  - nvidia.com/gpu:NoSchedule
 NvidiaGPU:
  enabled: true
-  package: nvidia-headless-460-server
+  package: nvidia-headless-510-server
 UpdatePolicy: automatic
 channels:
 - memfs://clusters.example.com/minimal.example.com/addons/bootstrap-channel.yaml
@ -69,5 +69,5 @@ containerdConfig:
  logLevel: info
  nvidiaGPU:
    enabled: true
-    package: nvidia-headless-460-server
+    package: nvidia-headless-510-server
  version: 1.4.12
--- a/upup/models/cloudup/resources/addons/nvidia.addons.k8s.io/k8s-1.16.yaml
+++ b/upup/models/cloudup/resources/addons/nvidia.addons.k8s.io/k8s-1.16.yaml
@ -17,7 +17,7 @@ spec:
        name: nvidia-device-plugin-ds
    spec:
      containers:
-      - image: docker.io/nvidia/k8s-device-plugin:v0.10.0
+      - image: docker.io/nvidia/k8s-device-plugin:v0.11.0
        name: nvidia-device-plugin-ctr
        args: ["--fail-on-init-error=false"]
        securityContext: