NVIDIA/nvidia-docker issues and pull requests

#1658 - nvidia-docker installation GPG error

Issue - State: closed - Opened by enricoschroeder over 2 years ago - 3 comments

#1657 - nvidia-container-cli: initialization error: driver rpc error: timed out

Issue - State: closed - Opened by nightzjp over 2 years ago - 3 comments

#1652 - nvidia-container-cli: initialization error on Ubuntu22.04LTS

Issue - State: open - Opened by Lonitch over 2 years ago - 14 comments

#1650 - Failed to initialize NVML: Unknown Error after calling`systemctl daemon-reload`

Issue - State: open - Opened by cheyang over 2 years ago - 22 comments

#1648 - docker: Error response from daemon: failed to create shim task: OCI runtime create failed: runc create failed

Issue - State: closed - Opened by zyr-NULL over 2 years ago - 45 comments

#1642 - could not select device driver with capabilities

Issue - State: closed - Opened by Maryamtarazkar almost 3 years ago - 2 comments

#1638 - Nested container can't start

Issue - State: open - Opened by easeway almost 3 years ago - 7 comments

#1635 - Upgrade NVidia driver without the need to restart docker daemon

Issue - State: open - Opened by remoteweb almost 3 years ago - 9 comments

#1634 - No repository for Ubuntu 20.04

Issue - State: closed - Opened by vbalaji21 almost 3 years ago - 1 comment

#1632 - GPG error "public key is not available" in Ubuntu 20.04 CUDA 11.4.0 image while building

Issue - State: open - Opened by wilkesreid almost 3 years ago - 27 comments

#1631 - Public GPG key error

Issue - State: open - Opened by vamman almost 3 years ago - 57 comments

#1627 - Unable to use sudo apt-get - new issue

Issue - State: open - Opened by michaelklachko almost 3 years ago - 6 comments

#1626 - Unable to use sudo apt-get

Issue - State: closed - Opened by nikiabramov98 almost 3 years ago - 5 comments

#1616 - nvidia-container-cli: ldcache error: process /sbin/ldconfig.real failed with error code: 1: unknown

Issue - State: open - Opened by Hurricane-eye almost 3 years ago - 9 comments

#1613 - Request: official Fedora support

Issue - State: closed - Opened by locture about 3 years ago - 7 comments

#1606 - Request: Alpine Linux support

Issue - State: open - Opened by lizelive about 3 years ago - 3 comments

#1597 - "/usr/local/nvidia/bin" is not available in Docker image

Issue - State: open - Opened by somum about 3 years ago - 4 comments

#1593 - Volta MPS Server error "Receive command failed, assuming client exit. Client process disconnected"

Issue - State: open - Opened by amybachir about 3 years ago - 3 comments

#1589 - support Debian stable

Issue - State: closed - Opened by jonorthwash about 3 years ago - 8 comments

#1584 - NVIDIA repository to list did not update to 20.04

Issue - State: open - Opened by vikashg about 3 years ago - 8 comments

#1582 - Fails to return GPU info when I run nvidia-smi inside the container but it works as expected outside of the container

Issue - State: closed - Opened by JonShelley about 3 years ago

#1580 - AttributeError: module 'torch.cuda' has no attribute 'amp'

Issue - State: closed - Opened by suresh-s about 3 years ago

#1576 - Use mps on kubernetes

Issue - State: open - Opened by somelaoda about 3 years ago - 6 comments

#1575 - Conflict on daemon.json from nvidia-docker2 and docker

Issue - State: closed - Opened by zmiimz about 3 years ago - 11 comments

#1572 - Documentation about DinD is lacking and misleading

Issue - State: closed - Opened by awoimbee about 3 years ago - 2 comments

#1571 - No support for installing drivers on Rocky Linux 8.4

Issue - State: closed - Opened by DeepikaKrishnaiah about 3 years ago - 4 comments

#1570 - Tensor Cores utilization containts during an l4t container instantiation

Issue - State: closed - Opened by ktsakos about 3 years ago

#1567 - Is it possible to run nvidia-docker inside rootfs filesystems?

Issue - State: closed - Opened by itomilin about 3 years ago

#1565 - CentOS 8.4 rootless podman: [nvidia-container-cli: initialization error: driver error: failed to process request]

Issue - State: closed - Opened by SoloGao over 3 years ago - 2 comments

#1562 - CUDA_MPS_ACTIVE_THREAD_PERCENTAGE did not support 3090 and A100

Issue - State: open - Opened by MoFHeka over 3 years ago

#1559 - How to build a image independent of the host machine.

Issue - State: closed - Opened by TiankaiHang over 3 years ago

#1558 - NO CMAKE_CUDA_COMPILER could be found when building CUDA app on x86

Issue - State: closed - Opened by Nir-Az over 3 years ago

#1557 - Cannot open /dev/tty0

Issue - State: closed - Opened by SpringDay6 over 3 years ago - 3 comments

#1554 - WSL2 + Docker + OpenGL + NVIDIA not working (uses llvmpipe)

Issue - State: open - Opened by rr-robert over 3 years ago - 32 comments

#1553 - Cannot kill container running with failed GPU

Issue - State: closed - Opened by cswinter over 3 years ago

#1551 - WSL2: nvidia-container-cli mount error, libnvidia-ml.so.1: file exists: unknown.

Issue - State: open - Opened by Mihawk2022 over 3 years ago - 35 comments

#1548 - "The NVIDIA Driver was not detected. GPU functionality will not be available." while using the latest drivers and nvidia-docker

Issue - State: closed - Opened by PQLLUX over 3 years ago - 3 comments

#1546 - Nvidia/CUDA Docker containers nested inside LXD container fail if running the LXD container as unprivileged (OK, if privileged)

Issue - State: open - Opened by waldekkot over 3 years ago - 13 comments

#1543 - mps could not work both in container and host in the same time

Issue - State: closed - Opened by archwolf118 over 3 years ago - 2 comments

#1540 - Unable to load NVML on SLES 15.3 based kubernetes with microk8s

Issue - State: closed - Opened by kbegiedza over 3 years ago - 2 comments

#1532 - Please add support for the Rocky Linux 8.x

Issue - State: closed - Opened by arm2arm over 3 years ago - 1 comment

#1531 - how to use CUDA_MPS_ACTIVE_THREAD_PERCENTAGE for multi-gpus in one pod situation

Issue - State: closed - Opened by MC17 over 3 years ago

#1529 - cannot start nvidia-docker runtime on jetson tx2

Issue - State: closed - Opened by HoangTienDuc over 3 years ago

#1524 - The public GPG key has expired

Issue - State: closed - Opened by Sentient07 over 3 years ago - 9 comments

#1523 - nvml error: insufficient permissions

Issue - State: closed - Opened by ethanabrooks over 3 years ago - 2 comments

#1521 - Unable to build ubuntu20.04 docker

Issue - State: closed - Opened by gnthibault over 3 years ago - 2 comments

#1520 - Missing EGL ICD configuration file

Issue - State: closed - Opened by velovix over 3 years ago - 5 comments

#1518 - Unable to run docker nvcr.io/nvidia/k8s/cuda-sample:nbody, error: nvml error: driver not loaded: unknown

Issue - State: closed - Opened by codeastra2 over 3 years ago - 5 comments

#1517 - how is `nvidia-docker` or `docker --gpus all` different with `docker -v /path/libcuda.so -v /path/other-nvidia-driver.so -v... --device /dev/nvidia0:/dev/nvidia0 --device...`

Issue - State: closed - Opened by melyseengit over 3 years ago - 2 comments

#1515 - The latest nvidia-container-toolkit caused inconsistent cuda version and 804 error.

Issue - State: open - Opened by gemfield over 3 years ago - 18 comments

#1512 - I installed nvidia/cuda:10.2-base, but still requirement error: unsatisfied condition: cuda>=11.2 when running tensorflow

Issue - State: closed - Opened by hnqmz over 3 years ago - 1 comment

#1511 - Add RHEL 8.4 support & repos

Issue - State: closed - Opened by shurik-io over 3 years ago - 7 comments

#1509 - how k8s+docker(19.03+)+nvidia-docker2 dectect to use --gpus options through k8s OCI

Issue - State: closed - Opened by perrywang over 3 years ago - 1 comment

#1508 - Nvidia CUDA driver does not detect my graphical device in docker

Issue - State: closed - Opened by Telluro over 3 years ago - 1 comment

#1504 - buildkit loose the binding of the nvidia driver libraries available on the host

Issue - State: closed - Opened by rafraph over 3 years ago - 14 comments

#1499 - OCI hooks for podman in Debian??

Issue - State: closed - Opened by jmpolom almost 4 years ago - 4 comments

#1498 - install uses apt key and no support for Ubuntu 21

Issue - State: closed - Opened by caseykneale almost 4 years ago - 8 comments

#1495 - oracle

Issue - State: closed - Opened by inoray almost 4 years ago

#1494 - grid k2

Issue - State: closed - Opened by aiaiai-czh almost 4 years ago - 1 comment

#1493 - Cannot download nvidia-container-toolkit

Issue - State: closed - Opened by siyangxie almost 4 years ago - 8 comments

#1492 - docker: Error response from daemon: OCI runtime create failed: container_linux.go:367: starting container process caused: process_linux.go:495: container init caused: Running hook #0:: error running hook: exit status 1, stdout: , stderr: nvidia-container-cli: detection error: nvml error: function not found: unknown.

Issue - State: open - Opened by aiaiai-czh almost 4 years ago - 29 comments

#1491 - Looking for centos7 15.2 package

Issue - State: closed - Opened by Chwei-L almost 4 years ago

#1490 - Looking for openSUSE Leap 15.2 package

Issue - State: closed - Opened by Doctor-Who almost 4 years ago - 2 comments

#1489 - Docker build failed to register layer; read-only file system with NVIDIA base image

Issue - State: closed - Opened by Sicily-F almost 4 years ago

#1484 - Nvidia-container-cli: initialization error: driver error: timed out while executing "nvidia-docker run -it nvcr.io/nvidia/pytorch:20.03-py3"

Issue - State: closed - Opened by anirudhb11 almost 4 years ago - 1 comment

#1482 - nvidia-smi not found when using nvidia/cuda:* images in WSL2/Ubuntu 20.04

Issue - State: closed - Opened by f0r4y312 almost 4 years ago - 1 comment

#1478 - Does nvidia-dcoker support win10? If so, how to configure it?

Issue - State: closed - Opened by huang229 almost 4 years ago

#1477 - Performing AI calculations in docker containers for a long time, causing the server to crash

Issue - State: closed - Opened by Seealso almost 4 years ago

#1475 - Client Error: Not Found ("No such container: xxxxx

Issue - State: closed - Opened by sguerrer almost 4 years ago

#1470 - Docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]]

Issue - State: closed - Opened by btalb almost 4 years ago - 4 comments

#1469 - GPU becomes unavailable after some time in Docker container

Issue - State: closed - Opened by tobigue almost 4 years ago - 9 comments

#1463 - Jetson nano cannot upgrade NVIDIA drive

Issue - State: closed - Opened by zc2638 almost 4 years ago

#1461 - Doc: Typo in warning present on Getting Started page

Issue - State: closed - Opened by 1nF0rmed almost 4 years ago - 2 comments

#1459 - Possible to use docker image for builds without a GPU but run on a GPU machine?

Issue - State: closed - Opened by msmcconnell about 4 years ago

#1456 - docker: Error response from daemon: OCI runtime create failed (ldconfig.real terminated with signal 31)

Issue - State: closed - Opened by alextoind about 4 years ago - 15 comments

#1452 - Sometimes NVIDIA docker hung when using torch.cuda() Call Trace

Issue - State: closed - Opened by denghi about 4 years ago

#1450 - Multi GPU from 1 GPU on metal?

Issue - State: closed - Opened by tyoc213 about 4 years ago

#1447 - cgroup issue with nvidia container runtime on Debian testing

Issue - State: closed - Opened by super-cooper about 4 years ago - 62 comments

#1445 - Pytorch/CUDA issue (driver mismatch) on Docker ?

Issue - State: closed - Opened by apiszcz about 4 years ago - 7 comments

#1438 - ROS in docker: rviz::RenderSystem: error creating render window: OGRE EXCEPTION(3:RenderingAPIException)

Issue - State: closed - Opened by Durant35 about 4 years ago - 5 comments

#1436 - 2 3090s + nvidia-docker containers = READ-ONLY file system crash

Issue - State: closed - Opened by lonce about 4 years ago - 4 comments

#1434 - Nvidia-Docker on windows with WSL1

Issue - State: closed - Opened by Melliti about 4 years ago

#1424 - Docker error

Issue - State: closed - Opened by stephcar75020 about 4 years ago - 9 comments

#1420 - The "runtimes" settings in docker daemon config file for nvidia-docker on Ubuntu 20.04 will defeat the docker service.

Issue - State: closed - Opened by hongyi-zhao over 4 years ago - 5 comments

#1414 - Cuda unable to detect GPU in Ubuntu20.04 container

Issue - State: closed - Opened by Mithrandir2k18 over 4 years ago - 2 comments

#1413 - https://nvidia.github.io/nvidia-container-runtime/ubuntu18.04/amd64/

Issue - State: closed - Opened by siyiwolf over 4 years ago - 2 comments

#1412 - https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/amd64/

Issue - State: closed - Opened by siyiwolf over 4 years ago - 2 comments

#1411 - The https://nvidia.github.io/nvidia-docker/ can not connect

Issue - State: closed - Opened by siyiwolf over 4 years ago - 1 comment

#1410 - nvidia docker does not run with winver 20251.1

Issue - State: closed - Opened by tadam98 over 4 years ago - 12 comments

#1409 - nvidia-container-cli reports incorrect CUDA driver version on WSL2

Issue - State: open - Opened by danfairs over 4 years ago - 15 comments

#1403 - Centos 8 repomd.xml GPG signature verification error and Update not working

Issue - State: closed - Opened by m-reuter over 4 years ago - 1 comment

#1400 - nvidia-container-cli fails to initialize dxcore

Issue - State: closed - Opened by suntaochun over 4 years ago - 5 comments

#1397 - cuda:10.0-cudnn7-devel-ubuntu16.04 - Facing Issues

Issue - State: closed - Opened by suresh-s over 4 years ago - 1 comment

#1396 - Cuda causes segmentation fault using MPS

Issue - State: closed - Opened by steeler18 over 4 years ago - 5 comments

#1393 - NVIDIA Docker - initialization error: nvml error: driver not loaded

Issue - State: closed - Opened by RylanSchaeffer over 4 years ago - 15 comments

#1391 - docker: Error response from daemon: OCI runtime create failed: container_linux.go (on jetson nano with nvidia-docker2)

Issue - State: closed - Opened by kaisark over 4 years ago

#1388 - Error: nvidia-docker2 : Depends: nvidia-container-runtime (>= 3.4.0)

Issue - State: closed - Opened by JingL1014 over 4 years ago - 55 comments

#1385 - nvidia-container-cli: mount error: stat failed: /Tesla V100-SXM2-16GB: no such file or directory

Issue - State: closed - Opened by xhyandwyy over 4 years ago

#1383 - unmet dependencies when trying to install nvidia-docker2

Issue - State: closed - Opened by jhaggle over 4 years ago

#1382 - centos container_linux.go:345: starting container process caused "exec: \"nvidia-smi\": executable file not found in $PATH": unknown.

Issue - State: open - Opened by SubMarineas over 4 years ago - 12 comments

GitHub / NVIDIA/nvidia-docker issues and pull requests