Autodl
自建了私有云,但是偶尔会报错:
1 2 |
Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running prestart hook #0: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: mount error: failed to add device rules: unable to find any existing device filters attached to the cgroup: bpf_prog_query(BPF_CGROUP_DEVICE) failed: operation not permitted: unknown |
这是NVIDIA Container Toolkit
正在尝试用 cgroup/eBPF
做设备过滤,但当前的运行环境( LXD
容器里跑 Docker
)不允许这么做,于是容器启动在 prestart hook
阶段失败。
因此,我们只需要把 NVIDIA
的 cgroup/eBPF
设备过滤关掉即可。
vi /etc/nvidia-container-runtime/config.toml
修改以下字段:
1 2 |
[nvidia-container-cli] no-cgroups = true |
然后:sudo systemctl restart docker
然后就可以了~