Stack Overflow
Stack Overflow

Questions tagged [nvidia-docker]

nvidia-docker is a thin wrapper on top of docker and act as a drop-in replacement for the docker command line interface. This is required as a convenience to automatically detect and setup GPU containers leveraging NVIDIA hardware. nvidia-docker calls docker and relies on the NVIDIA Docker plugin to discover driver files and GPU devices.

nvidia-docker is a thin wrapper on top of docker and act as a drop-in replacement for the docker command line interface. This is required as a convenience to automatically detect and setup GPU containers leveraging NVIDIA hardware. nvidia-docker calls docker and relies on the NVIDIA Docker plugin to discover driver files and GPU devices.

Source: Nvidia

264 questions
0
votes
1 answer

Kubernetes add GPU support, does the cpu node need install nvidia driver and nvidia docker?

I am new to Kubernetes. Now I need to add 2 nodes with GPU. The origin environment is 5 cpu node. Should I install nvidia driver and nvidia docker in the 5 cpu node too?
Alex Xu
  • 29
  • 1
  • 9
0
votes
0 answers

How are the docker containers scheduled?

I am trying to figure out how the docker containers are scheduled. Does anyone know which files I should read or what factors are considered when the containers are scheduled? I am also interested in nvidia-docker and experimenting with Rodinia…
minji
  • 1
0
votes
1 answer

Can I map gpu drivers of host machine (windows) inside docker container?

I am trying to execute a codebase (detect and track by facebook) inside the docker container that requires GPU access. My docker image is linux based with CUDA toolkit installed however I cannot see any GPU devices in proc filesystem. The host…
Rachit Tayal
  • 1,190
  • 14
  • 20
0
votes
1 answer

First run of a nvidia-docker2 container very slow

When running a GPU-enabled docker container on an EC2 p2.xlarge instance I experience a delay of between 30 and 90 seconds before the container starts running. Subsequent containers start fast (1 second delay). The EC2 is running ubuntu 18.04 with…
lenawal
  • 143
  • 7
0
votes
0 answers

Why docker run in daemon mode (-d) get a permission denied (selinux) and not in interactive mode (-ti)

Question: Can somebody explain to me why selinux rules are applied in case of running container in daemon mode but not in interactive mode ? Use case: I am running a docker container with nvidia-gpu support. When I am trying to run it interactive…
Minux
  • 56
  • 5
0
votes
1 answer

Prevent Kops from replacing docker installation when provisioning nodes

I use custom images (AMIs) configured for machine learning on GPU-enabled EC2 instances. This means cuda, libcudnn6, nvidia-docker etc are all properly setup on them. However when Kops starts new nodes from these AMIs (I use cluster-autoscaler) it…
MasterScrat
  • 7,090
  • 14
  • 48
  • 80
0
votes
2 answers

How to increase the timeout limit of nvidia-docker initialization

I would like to know how to increase the timeout limit of nvidia-docker at initialization. When 2 or more of my 4-GPU server are busy, I always get a timeout error: nvidia-container-cli: initialization error: driver error: timed out when…
f10w
  • 1,524
  • 4
  • 24
  • 39
0
votes
0 answers

Failed to run a code using GPU in nvidia-docker container. (nightly-gpu-py3)

I'm teaching myself what is docker and how to use it. I'm really new to docker, so hope to learn some basics here. I installed nvidia-docker (following installation guide) and tensorflow/tensorflow:nightly-gpu-py3 (nightly-gpu, start GPU (CUDA)…
displayname12345
  • 1
  • 1
0
votes
1 answer

The system is running in low-graphics mode

Im trying to repurpose a Cryptocoin mining rig for AI research. The operating system is Ubuntu 16.04 LTS. In order to utilize the graphics cards I need to install Nvidia drivers and Cuda. However, whenever I restart the box I always get the…
Magick
  • 4,603
  • 22
  • 66
  • 103
0
votes
1 answer

Error response from daemon: get nvidia_driver_375.66: no such volume: nvidia_driver_375.66

I proceeded trouble shooting for tensorflow can't use GPU in docker. I find nvidia driver DSO files is 375.66, incompatible with my current version 375.26. So I deleted the dir /var/lib/nvidia-docker/volumes/nvidia_driver/367.66. But when I closed…
semon
  • 15
  • 5
0
votes
1 answer

Use nvidia-docker instead of docker with Ansible

I'm trying to figure out how to use nvidia-docker (https://github.com/NVIDIA/nvidia-docker) using https://docs.ansible.com/ansible/latest/docker_container_module.html#docker-container. Problem My current Ansible playbook execute my container using…
PERPO
  • 3,812
  • 1
  • 13
  • 20
0
votes
1 answer

adapt vim configs to nvidia-docker container

I am using nvidia-docker to train few deep learning models. Every time I attach to my running container I realised, vim settings are not adapting into container's vim. So I installed vundle, and copied host's .vimrc options to containers…
enRIKO
  • 96
  • 6
0
votes
2 answers

nvidia-smi executable file not found

I have went through 3 different issues in the nvidia-docker repo about this exact problem but actually couldn't figure out what's wrong. I'm a heavy docker user but I don't understand much of the terminology and solution used in those issues. When…
bluesummers
  • 11,365
  • 8
  • 72
  • 108
0
votes
1 answer

TensorFlow won't detect GPU with custom docker image and python 3.6

(Posting here before submitting an issue to tensorflow as their issue template suggests) I'm trying to build a tensorflow docker image with python 3.6, I have the following Dockerfile FROM nvidia/cuda:8.0-cudnn5-devel-ubuntu16.04 RUN apt-get update…
SColvin
  • 11,584
  • 6
  • 57
  • 71
0
votes
1 answer

Nvidia-docker - cant find libnvcuvid.so on building process

I wrote a Dockerfile that builds my QT application and I'm having some problems on the build. If the build command is on the Dockerfile, it trhows this error: ninja: error: '/usr/lib/x86_64-linux-gnu/libnvcuvid.so', needed by…
Leonardo Galani
  • 184
  • 5
  • 22
1 2 3
…
17
18