禁止Ubuntu自动更新
在运行 nvidia-docker 的时候出现这样的错误:
docker: Error response from daemon: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #1: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: initialization error: nvml error: driver not loaded: unknown.
看上去是显卡驱动出问题了,在运行 nvidia-smi,出现以下错误:
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
可以确认显卡驱动出问题了。
这个情况出现过好多次了,都是莫名其妙的出现这个问题,没人动过服务器。现在分析,那大概率是 Ubuntu 自动安装了更新。
禁用系统自动更新。编辑 /etc/apt/apt.conf.d/20auto-upgrades
。默认是这样的:
APT::Periodic::Update-Package-Lists "1"; APT::Periodic::Unattended-Upgrade "1";
将 APT::Periodic::Unattended-Upgrade
改成 0。并保存,即可。
APT::Periodic::Update-Package-Lists
是检测是否有新的包,我觉得是有意义的,系统还是需要更新的,但应该是受控的更新。因此,我们需要了解是否有宝需要更新了,以便安排更新时间。