Description
What Happened?
minikube start --memory=max --driver=docker --container-runtime=docker --gpus=all --force --addons=nvidia-device-plugin
π minikube v1.34.0 on Nixos 25.05
βͺ MINIKUBE_WANTUPDATENOTIFICATION=false
β minikube skips various validations when --force is supplied; this may lead to unexpected behavior
β¨ Using the docker driver based on user configuration
π§― The requested memory allocation of 15875MiB does not leave room for system overhead (total system memory: 15875MiB). You may face stability issues.
π‘ Suggestion: Start minikube with less memory allocated: 'minikube start --memory=3900mb'
π Using Docker driver with root privileges
π Starting "minikube" primary control-plane node in "minikube" cluster
π Pulling base image v0.0.45 ...
β minikube was unable to download gcr.io/k8s-minikube/kicbase:v0.0.45, but successfully downloaded docker.io/kicbase/stable:v0.0.45 as a fallback image
π₯ Creating docker container (CPUs=2, Memory=15875MB) ...
π€¦ StartHost failed, but will try again: creating host: create: creating: create kic node: create container: docker run -d -t --privileged --security-opt seccomp=unconfined --tmpfs /tmp --tmpfs /run -v /run/current-system/kernel-modules/lib/modules:/lib/modules:ro --hostname minikube --name minikube --label created_by.minikube.sigs.k8s.io=true --label name.minikube.sigs.k8s.io=minikube --label role.minikube.sigs.k8s.io= --label mode.minikube.sigs.k8s.io=minikube --network minikube --ip 192.168.49.2 --gpus all --env NVIDIA_DRIVER_CAPABILITIES=all --volume minikube:/var --security-opt apparmor=unconfined --memory=15875mb -e container=docker --expose 8443 --publish=127.0.0.1::8443 --publish=127.0.0.1::22 --publish=127.0.0.1::2376 --publish=127.0.0.1::5000 --publish=127.0.0.1::32443 docker.io/kicbase/stable:v0.0.45@sha256:81df288595202a317b1a4dc2506ca2e4ed5f22373c19a441b88cfbf4b9867c85: exit status 125
stdout:
b42d8ba6ac445e1fd3de0dc12c79e9f3655b79b46376b6aa6d1d3bc99ddb1b9e
stderr:
docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]]
Run 'docker run --help' for more information
Attach the log file
None
Operating System
Other
Driver
Docker
Other infomation
In nixos's docker, --gpus=all is not supported, we can only use --device=nvidia.com/gpu=all
See also
https://github.com/NixOS/nixpkgs/issues/363505#issuecomment-2665190781
Example:
Ξ» ~/ docker run --rm -it --gpus=all ubuntu:latest nvidia-smi
docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]]
Run 'docker run --help' for more information
Ξ» ~/ docker run --rm -it --device=nvidia.com/gpu=all ubuntu:latest nvidia-smi
Mon Apr 7 09:50:01 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.133.07 Driver Version: 570.133.07 CUDA Version: 12.8 |
|-----------------------------------------+------------------------+----------------------+
| 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 NVIDIA GeForce RTX 2070 Off | 00000000:01:00.0 Off | N/A |
| 28% 38C P8 11W / 175W | 157MiB / 8192MiB | 3% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
+-----------------------------------------------------------------------------------------+
Question
Can we support
minikube start --memory=max --driver=docker --container-runtime=docker --device=nvidia.com/gpu=all --force --addons=nvidia-device-plugin
?
Metadata
Metadata
Assignees
Labels
No labels