Nvidia-docker: unsupported version on ubuntu studio 18.04.1 LTS

Created on 5 Nov 2018 · 7Comments · Source: NVIDIA/nvidia-docker

Hello,

I have a fresh install and docker is working fine. Since I have a nvidia quadro p400 I would like to use the GPU for additional processing on the containers.

Any ideas?

Thanks

Carlos

curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list |

sudo tee /etc/apt/sources.list.d/nvidia-docker.list

Unsupported distribution!

Check https://nvidia.github.io/nvidia-docker

Source

carlozousa

👍1

All 7 comments

I have not been able to reproduce this issue.

Add cuda-repo and install the cuda-driver package.

https://developer.nvidia.com/cuda-gpus
https://github.com/NVIDIA/nvidia-docker/wiki/Frequently-Asked-Questions#how-do-i-install-the-nvidia-driver

CUDA SDK 10.0 support for compute capability 3.0 – 7.5 (Kepler, Maxwell, Pascal, Volta, Turing).

https://en.wikipedia.org/wiki/CUDA

2nd, Add nvidia-docker repo.

% curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | \
  sudo apt-key add -
% distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
% curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | \
  sudo tee /etc/apt/sources.list.d/nvidia-docker.list
% sudo apt-get update

3rd, Install the nvidia-docker2,nvidia-container-runtime packages.

$ apt-cache madison docker.io
 docker.io | 17.12.1-0ubuntu1 | http://archive.ubuntu.com/ubuntu bionic/universe amd64 Packages
$ apt-cache madison nvidia-docker2 nvidia-container-runtime|grep docker17.12.1
nvidia-docker2 | 2.0.3+docker17.12.1-1 | https://nvidia.github.io/nvidia-docker/ubuntu18.04/amd64  Packages
nvidia-container-runtime | 2.0.0+docker17.12.1-1 | https://nvidia.github.io/nvidia-container-runtime/ubuntu18.04/amd64  Packages
$ sudo apt-get install -y nvidia-docker2=2.0.3+docker17.12.1-1 nvidia-container-runtime=2.0.0+docker17.12.1-1

$ sudo docker run --runtime=nvidia --rm nvidia/cuda:9.0-base nvidia-smi
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 410.72       Driver Version: 410.72       CUDA Version: 10.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1080    Off  | 00000000:03:00.0 Off |                  N/A |
| 26%   35C    P0    39W / 180W |      0MiB /  8119MiB |      5%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

ytooyama on 6 Nov 2018

❤1

Hello!

What is the distribution you are trying to install nvidia-docker on?

RenaudWasTaken on 23 Nov 2018

ubuntu studio 18.04.1 lts ...

A sex, 23/11/2018, 20:54, Renaud Gaubert notifications@github.com
escreveu:

Hello!

What is the distribution you are trying to install nvidia-docker on?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
https://github.com/NVIDIA/nvidia-docker/issues/855#issuecomment-441316565,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ADwiIjgGde8h87Qae0AzZmSPLio2viSlks5uyGBzgaJpZM4YOVoD
.

carlozousa on 23 Nov 2018

Are you still reproducing that bug?

On an aws Ubuntu 18.04 that command is successful.

On Fri, Nov 23, 2018, 12:58 carlozousa notifications@github.com wrote:

ubuntu studio 18.04.1 lts ...

A sex, 23/11/2018, 20:54, Renaud Gaubert notifications@github.com
escreveu:

Hello!

What is the distribution you are trying to install nvidia-docker on?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<
https://github.com/NVIDIA/nvidia-docker/issues/855#issuecomment-441316565
,
or mute the thread
<
https://github.com/notifications/unsubscribe-auth/ADwiIjgGde8h87Qae0AzZmSPLio2viSlks5uyGBzgaJpZM4YOVoD

.

—
You are receiving this because you commented.

Reply to this email directly, view it on GitHub
https://github.com/NVIDIA/nvidia-docker/issues/855#issuecomment-441317026,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ACZyE37ErVoHdlyOAqYz1DXGs99md92Jks5uyGF6gaJpZM4YOVoD
.

RenaudWasTaken on 24 Nov 2018

Closing as no further comment from the user was given.
Feel free to reopen if you have the information given by the following command: cat /etc/os-release

RenaudWasTaken on 20 Dec 2018

I had the same problem with ubuntu 16.04. It think the problem may be the README not being clear enough about the installation command. I tried to copy & paste the whole instruction block "add the package repositories" and I had the same error, but when I copied the command separately it worked.

havetc on 11 Feb 2019

👍1

in curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | \ just replace $distribution with your distro ID and VERSION_ID which you can get from the command cat /etc/os-release for example ubuntu18.04