Automate installing NVIDIA Container Toolkit --container-runtime

[spowelljr]

$ minikube start --container-runtime nvidia-docker
😄  minikube v1.31.2 on Debian rodete
✨  Automatically selected the docker driver
📌  Using Docker driver with root privileges
👍  Starting control plane node minikube in cluster minikube
🚜  Pulling base image ...
🔥  Creating docker container (CPUs=2, Memory=32100MB) ...
🛠️   Installing the NVIDIA Container Toolkit...
🐳  Preparing Kubernetes v1.28.2 on Docker 24.0.6 ...
    ▪ Generating certificates and keys ...
    ▪ Booting up control plane ...
    ▪ Configuring RBAC rules ...
🔗  Configuring CNI (Container Networking Interface) ...
🔎  Verifying Kubernetes components...
    ▪ Using image nvcr.io/nvidia/k8s-device-plugin:v0.14.1
    ▪ Using image gcr.io/k8s-minikube/storage-provisioner:v5
🌟  Enabled addons: nvidia-device-plugin, storage-provisioner, default-storageclass
🏄  Done! kubectl is now configured to use "minikube" cluster and "default" namespace by default

$ cat << EOF | kubectl create -f -
apiVersion: v1
kind: Pod
metadata:
  name: nvidia-version-check
spec:
  restartPolicy: OnFailure
  containers:
  - name: nvidia-version-check
    image: "nvidia/cuda:11.0.3-base-ubuntu20.04"
    command: ["nvidia-smi"]
    resources:
      limits:
         nvidia.com/gpu: "1"
EOF
pod/nvidia-version-check created

$ kubectl logs nvidia-version-check
Fri Sep 22 18:45:31 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 525.125.06   Driver Version: 525.125.06   CUDA Version: 12.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Quadro P1000        On   | 00000000:65:00.0 Off |                  N/A |
| 34%   26C    P8    N/A /  47W |     15MiB /  4096MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

$ minikube start --driver kvm --container-runtime nvidia-docker
😄  minikube v1.31.2 on Debian rodete (kvm/amd64)
✨  Using the kvm2 driver based on user configuration

❌  Exiting due to MK_USAGE: The nvidia-docker container-runtime can only be run with the docker driver

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: spowelljr

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [spowelljr]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[spowelljr]

/ok-to-test

[minikube-pr-bot]

kvm2 driver with docker runtime

+----------------+----------+---------------------+
|    COMMAND     | MINIKUBE | MINIKUBE (PR 17287) |
+----------------+----------+---------------------+
| minikube start | 50.2s    | 50.3s               |
| enable ingress | 27.0s    | 28.0s               |
+----------------+----------+---------------------+

Times for minikube start: 48.2s 51.3s 50.3s 50.4s 51.0s
Times for minikube (PR 17287) start: 50.1s 51.7s 50.8s 51.8s 47.0s

Times for minikube ingress: 27.1s 25.2s 27.7s 26.7s 28.1s
Times for minikube (PR 17287) ingress: 27.6s 27.7s 27.2s 29.2s 28.6s

docker driver with docker runtime

+----------------+----------+---------------------+
|    COMMAND     | MINIKUBE | MINIKUBE (PR 17287) |
+----------------+----------+---------------------+
| minikube start | 24.4s    | 23.9s               |
| enable ingress | 21.0s    | 21.0s               |
+----------------+----------+---------------------+

Times for minikube start: 24.7s 25.1s 22.1s 24.1s 25.7s
Times for minikube (PR 17287) start: 25.1s 22.3s 25.1s 21.7s 25.1s

Times for minikube ingress: 20.8s 21.3s 21.3s 20.8s 20.8s
Times for minikube (PR 17287) ingress: 20.8s 20.8s 20.9s 22.8s 19.8s

docker driver with containerd runtime

+----------------+----------+---------------------+
|    COMMAND     | MINIKUBE | MINIKUBE (PR 17287) |
+----------------+----------+---------------------+
| minikube start | 23.3s    | 22.8s               |
| enable ingress | 32.0s    | 29.1s               |
+----------------+----------+---------------------+

Times for minikube start: 23.4s 20.8s 23.8s 24.7s 23.7s
Times for minikube (PR 17287) start: 24.3s 21.4s 24.2s 23.3s 21.0s

Times for minikube ingress: 31.3s 31.3s 47.3s 18.4s 31.4s
Times for minikube (PR 17287) ingress: 31.4s 20.3s 31.3s 31.3s 31.3s

[minikube-pr-bot]

These are the flake rates of all failed tests.

Environment	Failed Tests	Flake Rate (%)
KVM_Linux_crio	TestStartStop/group/no-preload/serial/Pause (gopogh)	n/a
KVM_Linux_crio	TestStartStop/group/no-preload/serial/VerifyKubernetesImages (gopogh)	n/a
Docker_Linux_containerd	TestKVMDriverInstallOrUpdate (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/CertSync (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/DockerEnv/powershell (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageBuild (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageListJson (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageListShort (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageListTable (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageListYaml (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageLoadDaemon (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageLoadFromFile (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageReloadDaemon (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageSaveToFile (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ImageCommands/ImageTagAndLoadDaemon (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/MySQL (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/NodeLabels (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/PersistentVolumeClaim (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmdConnect (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/DeployApp (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/Format (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/HTTPS (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/JSONOutput (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/List (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/ServiceCmd/URL (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/StatusCmd (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/TunnelCmd/serial/RunSecondTunnel (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/TunnelCmd/serial/WaitService/Setup (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/parallel/UpdateContextCmd/no_changes (gopogh)	0.00 (chart)
Hyper-V_Windows	TestFunctional/serial/CacheCmd/cache/cache_reload (gopogh)	0.00 (chart)
More tests...	Continued...

Too many tests failed - See test logs for more details.

To see the flake rates of all tests by environment, click here.

[medyagh]

lets try to make another PR that will keep container runtimes 3 and have a flag called --enable-nvidia that way if we later have amd-gpu and also containerd and crio we could enable it fo rthem

[k8s-ci-robot]

PR needs rebase.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.