阿里云轻量级 GPU 实例安装 NVIDIA 驱动

实例规格:轻量级 GPU 实例 vgn6i-vws / ecs.vgn6i-m4-vws.xlarge(4vCPU 23GiB)
操作系统:Ubuntu 22.04

第一部分:尝试失败的安装方法

查询 NVIDIA 产品型号

lspci | grep -i nvidia

输出

00:07.0 VGA compatible controller: NVIDIA Corporation TU104GL [Tesla T4] (rev a1)

根据产品型号去 NVIDIA 官网下载驱动

wget -c https://us.download.nvidia.cn/tesla/535.154.05/nvidia-driver-local-repo-ubuntu2204-535.154.05_1.0-1_amd64.deb

安装驱动

cp /var/nvidia-driver-local-repo-ubuntu2204-535.154.05/nvidia-driver-local-91B8C5A2-keyring.gpg /usr/share/keyrings/
dpkg -i nvidia-driver-local-repo-ubuntu2204-535.154.05_1.0-1_amd64.deb
apt update
apt install nvidia-driver-535 nvidia-dkms-535
reboot

重启后运行 nvidia-smi 命令却出现下面的错误,驱动没有安装成功

NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.

接着用 ubuntu-drivers devices 命令查看 nvidia 驱动版本

modalias : pci:v000010DEd00001EB8sv000010DEsd0000130Ebc03sc00i00
vendor   : NVIDIA Corporation
model    : TU104GL [Tesla T4]
manual_install: True
driver   : nvidia-driver-470-server - distro non-free
driver   : nvidia-driver-470 - distro non-free
driver   : nvidia-driver-525-server - distro non-free
driver   : nvidia-driver-418-server - distro non-free
driver   : nvidia-driver-535-server - distro non-free
driver   : nvidia-driver-545 - distro non-free
driver   : nvidia-driver-525 - distro non-free recommended
driver   : nvidia-driver-450-server - distro non-free
driver   : xserver-xorg-video-nouveau - distro free builtin

然后用下面的命令安装

apt install nvidia-driver-525-server

重启后问题依旧

第二部分:正确的安装方法

在阿里云官网找到这篇帮助文档——在GPU虚拟化型实例中安装GRID驱动(Linux),通过下面的命令成功完成了安装

if acs-plugin-manager --list --local | grep grid_driver_install > /dev/null 2>&1
then
    acs-plugin-manager --remove --plugin grid_driver_install
fi

acs-plugin-manager --exec --plugin grid_driver_install

nvidia-smi 命令输出结果:

相关博问:Ubuntu 安装 nvidia-container-toolkit 遇到问题 "load library failed: libnvidia-ml.so.1"

posted @   dudu  阅读(717)  评论(0编辑  收藏  举报
相关博文:
阅读排行:
· Deepseek官网太卡,教你白嫖阿里云的Deepseek-R1满血版
· 2分钟学会 DeepSeek API,竟然比官方更好用!
· .NET 使用 DeepSeek R1 开发智能 AI 客户端
· DeepSeek本地性能调优
· 一文掌握DeepSeek本地部署+Page Assist浏览器插件+C#接口调用+局域网访问!全攻略
历史上的今天:
2015-02-06 Mac OS X上尝试编译CoreCLR源代码
2010-02-06 OutputCache造成页面响应内容类型为text/vnd.wap.wml的问题
2009-02-06 心态
2008-02-06 聊聊今年的春节联欢晚会
2008-02-06 祝大家新年快乐
2007-02-06 VS 2005 Add-in开发随笔
2006-02-06 博客园准备购买新服务器
点击右上角即可分享
微信分享提示