排查故障
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
这个问题在机器重启之后出现
Ubuntu22.04回退系统内核_ubuntu回退内核版本-CSDN博客
Ubuntu断电/重启显卡驱动消失：“NVIDIA-SMI has failed because it couldn‘t communicate with the NVIDIA driver....“ - 延木 - 博客园
运维系列&AI系列&Lenovo-G双系统系列（前传-装之前搜索的）：NVIDIA驱动失效简单解决方案：NVIDIA-SMI has failed because it couldn‘t commun - 坦笑&&life - 博客园
运维系列&AI系列&Lenovo-G双系统系列（前传-装之前搜索的）：nvidia-smi报错：NVIDIA-SMI has failed because it couldn‘t communicat_麒麟系统 nvidia-smi has failed because it couldn't com-CSDN博客
$ sudo apt update
$ sudo apt upgrade
# 解决(base) y@y-System-Product-Name:~/yws_code$ uname -r
6.8.0-65-generic
(base) y@y-System-Product-Name:~/yws_code$ sudo dpkg --get-selections | grep linux-image
[sudo] y 的密码：
linux-image-6.8.0-40-generic install
linux-image-6.8.0-58-generic install
linux-image-6.8.0-59-generic install
linux-image-6.8.0-60-generic install
linux-image-6.8.0-64-generic install
linux-image-6.8.0-65-generic deinstall
linux-image-generic-hwe-22.04 install
(base) y@y-System-Product-Name:~/yws_code$ sudo apt-mark hold linux-image-6.8.0-65-generic
linux-image-6.8.0-65-generic 设置为保留。
(base) y@y-System-Product-Name:~/yws_code$ sudo apt-mark hold linux-image-generic-hwe-22.04
linux-image-generic-hwe-22.04 设置为保留。
(base) y@y-System-Product-Name:~/yws_code$ sudo apt-mark hold linux-image-generic linux-headers-generic linux-generic
linux-image-generic 设置为保留。
linux-headers-generic 设置为保留。
linux-generic 设置为保留。