服务器突然找不到GPU了。nvidia-smiNVIDIA-SMI has failed because it couldnt communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.这个报错说明NVIDIA 内核驱动没有正常加载意思是系统里没有正在运行的 NVIDIA kernel module。于是我去查了Linux内核版本发现确实升级了刚跟之前的不匹配。那么我们就需要降低内核版本。 第一步查看当前内核uname -r # 6.14.0-36-generic说明我现在确实在用6.14.0-36之前用的是6.14.0-29 第二步查看系统里已有的旧内核dpkg --list | grep linux-image如下所示是系统里面已经有的内核ii linux-image-6.14.0-29-generic 6.14.0-29.29~24.04.1 amd64 Signed kernel image generic rc linux-image-6.14.0-35-generic 6.14.0-35.35~24.04.1 amd64 Signed kernel image generic ii linux-image-6.14.0-36-generic 6.14.0-36.36~24.04.1 amd64 Signed kernel image generic rc linux-image-6.8.0-31-generic 6.8.0-31.31 amd64 Signed kernel image generic rc linux-image-6.8.0-44-generic 6.8.0-44.44 amd64 Signed kernel image generic rc linux-image-6.8.0-45-generic 6.8.0-45.45 amd64 Signed kernel image generic rc linux-image-6.8.0-48-generic 6.8.0-48.48 amd64 Signed kernel image generic rc linux-image-6.8.0-51-generic 6.8.0-51.52 amd64 Signed kernel image generic rc linux-image-6.8.0-52-generic 6.8.0-52.53 amd64 Signed kernel image generic ii linux-image-generic-hwe-24.04 6.14.0-36.36~24.04.1 amd64 Generic Linux kernel image系统里有6.14.0-29这里可以直接设置默认启动跳到第四步。如果系统里没有的话则要安装目标内核。 第三步安装目标内核6.14.0-29sudo apt update sudo apt install linux-image-6.14.0-29-generic linux-headers-6.14.0-29-generic 第四步设置默认启动为 6.14.0-29编辑grub文件sudo vim /etc/default/grub找到GRUB_DEFAULT0改成GRUB_DEFAULTAdvanced options for UbuntuUbuntu, with Linux 6.14.0-29-generic保存后sudo update-grub如下所示 第五步重启sudo reboot✅ 验证uname -r显示6.14.0-29-generic