nvidia-container-cli: mount error: stat failed: /GeForce: no such file or directory\\n\""": unknown错误

docker: Error response from daemon: OCI runtime create failed: container_linux.go:348: starting container process caused “process_linux.go:402: container init caused “process_linux.go:385: running prestart hook 1 caused \“error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --require=cuda>=9.0 --pid=7834 \\nnvidia-container-cli: mount error: stat failed: /GeForce: no such file or directory\\n\”””: unknown. 更新显卡驱动之后出现nvidia-docker启动错误

解决方案

dpkg -l | grep libnvidia 查看libnvidia-container-tools的版本
ii libnvidia-container-tools 1.0.0~rc.2-1 amd64 NVIDIA container runtime library (command-line tools)
ii libnvidia-container1:amd64 1.0.2-1 amd64 NVIDIA container runtime library

发现 libnvidia-container-tools的版本不够
运行以下代码升级libnvidia-container-tools 版本：

sudo apt-get install libnvidia-container-tools=1.0.2-1

大功告成！

本机环境：1080Ti+cuda-10.0+Ubuntu

参考链接

https://github.com/NVIDIA/nvidia-docker/issues/957