docker: Error response from daemon: create nvidia_driver_410.73: VolumeDriver.Create: internal error, check logs for details.

nvidia-dockerをインストールして実行してみたら，
こんなエラーが出た．

$ nvidia-docker run --rm nvidia/cuda nvidia-smi
docker: Error response from daemon: create nvidia_driver_410.73: VolumeDriver.Create: internal error, check logs for details.
See 'docker run --help'.

対処法を探して，解決したのでメモします．

ソースはここです．

github.com

まずは，nvidia-dockerの状態を書き換えます．

$ systemctl edit nvidia-docker

これで，下記を追記します．

[Service]
ExecStart=
ExecStart=/usr/bin/nvidia-docker-plugin -s $SOCK_DIR -d /usr/local/nvidia-docker-test

っで，ザザッと設定を行うことで，動くようになりました．

$ sudo mkdir /usr/local/nvidia-docker-test
$ sudo chown nvidia-docker /usr/local/nvidia-docker-test
$ sudo systemctl restart docker nvidia-docker

動作確認するとこんな感じです．

$ nvidia-docker run --rm nvidia/cuda nvidia-smi
Tue Oct 30 03:46:20 2018
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 410.73       Driver Version: 410.73       CUDA Version: 10.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Tesla V100-SXM2...  Off  | 00000000:04:00.0 Off |                    0 |
| N/A   35C    P0    51W / 300W |    417MiB / 16130MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+