NVIDIA / nvidia-docker

Build and run Docker containers leveraging NVIDIA GPUs
Apache License 2.0
17.19k stars 2.03k forks source link

Big Issue with NVIDA-docker on Ubuntu Linux #1058

Closed tobimichigan closed 5 years ago

tobimichigan commented 5 years ago

The template below is mostly useful for bug reports and support questions. Feel free to remove anything which doesn't apply to you and add more information where it makes sense.

Also, before reporting a new issue, please make sure that:


1. Issue or feature description

The repository 'https://nvidia.github.io/nvidia-docker disco Release' does not have a Release file E: Unable to locate package nvidia-container-toolkit

2. Steps to reproduce the issue

configure docker successfully on Ubuntu 10.0.4 then

sudo apt-get install -y nvidia-container-toolkit

3. Information to attach (optional if deemed irrelevant)

Timestamp : Thu Aug 29 21:01:03 2019 Driver Version : 390.116

Attached GPUs : 1 GPU 00000000:02:00.0 Product Name : GeForce GT 630 Product Brand : GeForce

I think NVIDIA Developer team should give this issue better attension.

cdesiniotis commented 5 years ago

Hello!

We currently do not have any published packages specifically for Ubuntu 19.04 (disco), however the packages for 18.04 will work just fine. Refer to the installation instructions here for Ubuntu 18.04.

Instead of distribution=$(. /etc/os-release;echo $ID$VERSION_ID) simply replace it with distribution=ubuntu18.04

tobimichigan commented 5 years ago

Yea Developer CDE nvidia-container-cli -V version: 1.0.4 build date: 2019-08-23T22:49+00:00 build revision: 67b384e9768f692581d08aecb8088f4dd09ea59d build compiler: x86_64-linux-gnu-gcc-7 7.4.0 build platform: x86_64 build flags: -D_GNU_SOURCE -D_FORTIFY_SOURCE=2 -DNDEBUG -std=gnu11 -O2 -g -fdata-sections -ffunction-sections -fstack-protector -fno-strict-aliasing -fvisibility=hidden -Wall -Wextra -Wcast-align -Wpointer-arith -Wmissing-prototypes -Wnonnull -Wwrite-strings -Wlogical-op -Wformat=2 -Wmissing-format-attribute -Winit-self -Wshadow -Wstrict-prototypes -Wunreachable-code -Wconversion -Wsign-conversion -Wno-unknown-warning-option -Wno-format-extra-args -Wno-gnu-alignof-expression -Wl,-zrelro -Wl,-znow -Wl,-zdefs -Wl,--gc-sections

ok. ..Great job But before I close this issue, why do I get this error "docker: Error response from daemon: could not select device driver "" with capabilities: [[utility gpu]]." Is there a reason for this?

tobimichigan commented 5 years ago

Sorry, my fault, I forgot to run: "sudo systemctl restart docker", after installing NVIDIA container. After running the above code, I now ran "docker run --gpus all nvidia/cuda:9.0-base nvidia-smi" and it returned: +-----------------------------------------------------------------------------+ | NVIDIA-SMI 390.116 Driver Version: 390.116 | |-------------------------------+----------------------+----------------------+ | 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 GT 630 Off | 00000000:02:00.0 N/A | N/A | | 40% 45C P0 N/A / N/A | 185MiB / 1985MiB | N/A Default | +-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+ | Processes: GPU Memory | | GPU PID Type Process name Usage | |=============================================================================| | 0 Not Supported | +-----------------------------------------------------------------------------+

Why is it saying "NOT SUPPORTED"? My card's CC is 2.1. I hope I can test NVIDIA deep learning models with this card?

cdesiniotis commented 5 years ago

The output above is expected. It is not telling you that your graphics card is not supported. Rather, it is telling you that the display of processes which are using the GPU is not supported, which is expected for your GeForce GT 630.

tobimichigan commented 5 years ago

You mean the monitor right?