之前安装autoware docker中出现的错误有来自于nvidia，或者是CUDA，或者是nvidia docker，这此autoware最后是安装成功了，但是还有一些问题。
dong@cvad:~$ docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi
Unable to find image 'nvidia/cuda:latest' locally
latest: Pulling from nvidia/cuda
898c46f3b1a1: Pull complete
63366dfa0a50: Pull complete
041d4cd74a92: Pull complete
6e1bee0f8701: Pull complete
131dbe7c254d: Pull complete
5bca6b05dcd6: Pull complete
0d286a7b6e12: Pull complete
5776d2c6371d: Pull complete
768e84e7fc24: Pull complete
Digest: sha256:eba1dc5810e40f60625ee797d618a6bd11be24cb67bc6647a4d36392202bb013
Status: Downloaded newer image for nvidia/cuda:latest
docker: Error response from daemon: OCI runtime create failed: container_linux.go:345:
starting container process caused "process_linux.go:424: container init caused \
"process_linux.go:407: running prestart hook 1 caused \\\"error running hook:
exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --
ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --require=cuda>=10.1 brand=
tesla,driver>=384,driver<385 brand=tesla,driver>=410,driver<411 --pid=9333/
var/lib/docker/overlay2/b64e08438f267f9c15fe7e68340df81ab085636359d502d92b9586324c5d8b41/
merged]\\\\nnvidia-container-cli: requirement error: unsatisfied condition: brand = tesla\\\\n\\\"\"": unknown.
这条命令
docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi
dong@cvad:~$ sudo docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi
docker: Error response from daemon: OCI runtime create failed: container_linux.go:345: starting container process caused "process_linux.go:424: container init caused "process_linux.go:407: running prestart hook 1 caused \"error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --require=cuda>=10.1 brand=tesla,driver>=384,driver<385 brand=tesla,driver>=410,driver<411 --pid=14008 /var/lib/docker/overlay2/48fa7509300f737df294dd36ae954b441b8343d20b039020eabecf5eeb90fe91/merged]\\nnvidia-container-cli: requirement error: unsatisfied condition: brand = tesla\\n\""": unknown.
是用来验证nvidia docker是否安装成功的，如果成功的话，应该显示显卡的信息才对。
现在怎么来解决问题呢。
先看看相关nvidia docker安装的第三方文档，是否能找到一点信息。搜索：nvidia docker安装
在Ubuntu 16.04上制作 NVIDIA CUDA Docker image
这篇有安装docker ce，卸载docker ce，获取sudo docker权限，runtime版本安装nvidia docker image，看晕了。
搭建nvidia-docker运行环境-Ubutu16.04
这一篇安装成功了，是用deb文件安装的。
直接搜索错误提示。
github | nvidia | k8s-device-plugin
这一篇提示的错误相同。
github | Docker image tensorflow/tensorflow:latest-gpu failed to use nvidia/cuda image #25618
从这一篇，我想到可能是匹配的问题，这里说题主的问题是驱动太老了，那我这里安装的是CUDA8.0和NVIDIA DOCKER 2.0，CUDA 8.0 安装的是老版本的，但NVIDIA DOCKER 2.0 是新版本的，现在要么把CUDA更新一下，要么把NVIDIA DOCKER安装回1.0。
搜索：docker daemon 是什么
了解docker daemon | Docker 实践（一）：了解架构
也是关于daemon | Docker系列之一：入门介绍
现在我想问题应该是cuda驱动和nvidia docker版本匹配的问题，现在就是要确认一下，1、autoware docker安装中相关教程版本的选择，2、直接检索出错的那一行验证代码，而不是错误结果，而且，我还记得，nvidia docker github上就有类似345的错误提示。
1、
| isl_qdu | 用 Docker 方式安装 Autoware
Installation (version 1.0)
参考的这一篇用的是nvidia docker 1.0
2、
官方autoware docker安装中也是基于nvidia docker 1的，
autoware github | autoware docker install | Generic x86 Docker
$ wget https://github.com/NVIDIA/nvidia-docker/releases/download/v1.0.1/nvidia-docker_1.0.1-1_amd64.deb
3、
基本的想到的解决方法就是安装nvidia docker1，这里还是要补充一些docker和nvidia docker的基本知识。
docer一般是服务于基于cpu的应用，
如果是要基于gpu的话，比如我们现在要做的事情，就要用到nvidia-docker，
nvidia-docker是在docker上做了一层封装，通过nvidia-docker-plugin封装之后调用docker，
更多内容
4、
现在重新安装nvidia docker 1
dong@cvad:~$ sudo docker --version
Docker version 18.09.5, build e8ff056
dong@cvad:~$ nvidia-smi
Tue Apr 30 10:36:35 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 384.130 Driver Version: 384.130 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GTX 106... Off | 00000000:01:00.0 On | N/A |
| 0% 29C P8 7W / 120W | 338MiB / 6069MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 1470 G /usr/lib/xorg/Xorg 181MiB |
| 0 2954 G compiz 149MiB |
| 0 3248 G fcitx-qimpanel 5MiB |
+-----------------------------------------------------------------------------+
dong@cvad:~$
nvidia docker wili | Installation (version 1.0)
- GNU/Linux x86_64 with kernel version > 3.10
- Docker >= 1.9 (official
docker-engine,docker-ceordocker-eeonly)- NVIDIA GPU with Architecture > Fermi (2.1)
- NVIDIA drivers >= 340.29 with binary
nvidia-modprobe
nvidia docker wili | Installation (version 2.0)
- GNU/Linux x86_64 with kernel version > 3.10
- Docker >= 1.12
- NVIDIA GPU with Architecture > Fermi (2.1)
- NVIDIA drivers ~= 361.93 (untested on older versions)
按理来说，安装nvidia docker也未必能解决问题。
Operation
dong@cvad:~$ curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | \
> sudo apt-key add -
OK
dong@cvad:~$ distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
dong@cvad:~$ curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | \
> sudo tee /etc/apt/sources.list.d/nvidia-docker.list
deb https://nvidia.github.io/libnvidia-container/ubuntu16.04/$(ARCH) /
deb https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/$(ARCH) /
deb https://nvidia.github.io/nvidia-docker/ubuntu16.04/$(ARCH) /
dong@cvad:~$ sudo apt-get update
获取:1 file:/var/cuda-repo-8-0-local InRelease
忽略:1 file:/var/cuda-repo-8-0-local InRelease
获取:2 file:/var/cuda-repo-8-0-local Release [574 B]
获取:2 file:/var/cuda-repo-8-0-local Release [574 B]
命中:4 https://nvidia.github.io/libnvidia-container/ubuntu16.04/amd64 InRelease
命中:5 https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/amd64 InRelease
命中:6 https://download.docker.com/linux/ubuntu xenial InRelease
命中:7 https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64 InRelease
命中:8 http://packages.ros.org/ros/ubuntu xenial InRelease
获取:9 http://security.ubuntu.com/ubuntu xenial-security InRelease [109 kB]
命中:10 http://cn.archive.ubuntu.com/ubuntu xenial InRelease
获取:11 http://cn.archive.ubuntu.com/ubuntu xenial-updates InRelease [109 kB]
获取:12 http://security.ubuntu.com/ubuntu xenial-security/main amd64 DEP-11 Metadata [68.0 kB]
命中:13 http://ppa.launchpad.net/apt-fast/stable/ubuntu xenial InRelease
获取:14 http://cn.archive.ubuntu.com/ubuntu xenial-backports InRelease [107 kB]
获取:15 http://security.ubuntu.com/ubuntu xenial-security/main DEP-11 64x64 Icons [67.1 kB]
忽略:16 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial InRelease
获取:17 http://security.ubuntu.com/ubuntu xenial-security/universe amd64 DEP-11 Metadata [116 kB]
获取:18 http://cn.archive.ubuntu.com/ubuntu xenial-updates/main amd64 Packages [947 kB]
获取:19 http://security.ubuntu.com/ubuntu xenial-security/universe DEP-11 64x64 Icons [173 kB]
忽略:20 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial Release
获取:21 http://security.ubuntu.com/ubuntu xenial-security/multiverse amd64 DEP-11 Metadata [2,464 B]
获取:22 http://cn.archive.ubuntu.com/ubuntu xenial-updates/main i386 Packages [819 kB]
获取:23 http://cn.archive.ubuntu.com/ubuntu xenial-updates/main amd64 DEP-11 Metadata [318 kB]
获取:24 http://cn.archive.ubuntu.com/ubuntu xenial-updates/main DEP-11 64x64 Icons [233 kB]
获取:25 http://cn.archive.ubuntu.com/ubuntu xenial-updates/universe amd64 Packages [747 kB]
忽略:26 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main amd64 Packages
获取:27 http://cn.archive.ubuntu.com/ubuntu xenial-updates/universe i386 Packages [684 kB]
获取:28 http://cn.archive.ubuntu.com/ubuntu xenial-updates/universe amd64 DEP-11 Metadata [252 kB]
获取:29 http://cn.archive.ubuntu.com/ubuntu xenial-updates/universe DEP-11 64x64 Icons [353 kB]
获取:30 http://cn.archive.ubuntu.com/ubuntu xenial-updates/multiverse amd64 DEP-11 Metadata [5,960 B]
获取:31 http://cn.archive.ubuntu.com/ubuntu xenial-updates/multiverse DEP-11 64x64 Icons [14.3 kB]
获取:32 http://cn.archive.ubuntu.com/ubuntu xenial-backports/main amd64 DEP-11 Metadata [3,324 B]
获取:33 http://cn.archive.ubuntu.com/ubuntu xenial-backports/universe amd64 DEP-11 Metadata [5,104 B]
忽略:34 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main i386 Packages
忽略:35 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main all Packages
忽略:36 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-zh_CN
忽略:37 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-zh
忽略:38 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-en
忽略:39 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main amd64 DEP-11 Metadata
忽略:40 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main DEP-11 64x64 Icons
忽略:26 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main amd64 Packages
-----------------省略
忽略:40 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main DEP-11 64x64 Icons
错误:26 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main amd64 Packages
404 Not Found
忽略:34 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main i386 Packages
忽略:35 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main all Packages
忽略:36 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-zh_CN
忽略:37 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-zh
忽略:38 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main Translation-en
忽略:39 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main amd64 DEP-11 Metadata
忽略:40 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial/main DEP-11 64x64 Icons
已下载 5,132 kB，耗时 1分 13秒 (69.6 kB/s)
正在读取软件包列表... 完成
W: 软件包仓库 Release 文件 /var/lib/apt/lists/_var_cuda-repo-8-0-local_Release 内 Date 条目无效
W: 仓库 “http://ppa.launchpad.net/fcitx-team/nightly/ubuntu xenial Release” 没有 Release 文件。
N: 无法认证来自该源的数据，所以使用它会带来潜在风险。
N: 参见 apt-secure(8) 手册以了解仓库创建和用户配置方面的细节。
E: 无法下载 http://ppa.launchpad.net/fcitx-team/nightly/ubuntu/dists/xenial/main/binary-amd64/Packages 404 Not Found
E: 部分索引文件下载失败。如果忽略它们，那将转而使用旧的索引文件。
dong@cvad:~$ sudo apt-fast install nvidia-docker
nvidia-docker 1.0.1-1 2.2MiB
Download size: 2.2MiB
Do you want to download the packages? [Y/n]
[#223a1a 1.1MiB/2.1MiB(53%) CN:2 DL:167KiB ETA:6s]
[#223a1a 2.1MiB/2.1MiB(98%) CN:1 DL:80KiB]
04/30 15:09:58 [NOTICE] Verification finished successfully. file=/var/cache/apt/apt-fast/nvidia-docker_1.0.1-1_amd64.deb
04/30 15:09:58 [NOTICE] Download complete: /var/cache/apt/apt-fast/nvidia-docker_1.0.1-1_amd64.deb
下载结果：
gid |stat|avg speed |path/URI
======+====+===========+=======================================================
223a1a|OK | 110KiB/s|/var/cache/apt/apt-fast/nvidia-docker_1.0.1-1_amd64.deb
状态标识：
(OK)：完成下载。
正在读取软件包列表... 完成
正在分析软件包的依赖关系树
正在读取状态信息... 完成
下列软件包是自动安装的并且现在不需要了：
libnvidia-container-tools libnvidia-container1 nvidia-container-runtime
nvidia-container-runtime-hook
使用'sudo apt autoremove'来卸载它(它们)。
下列【新】软件包将被安装：
nvidia-docker
升级了 0 个软件包，新安装了 1 个软件包，要卸载 0 个软件包，有 164 个软件包未被升级。
需要下载 0 B/2,266 kB 的归档。
解压缩后会消耗 14.1 MB 的额外空间。
正在选中未选择的软件包 nvidia-docker。
(正在读取数据库 ... 系统当前共安装有 302907 个文件和目录。)
正准备解包 .../nvidia-docker_1.0.1-1_amd64.deb ...
正在解包 nvidia-docker (1.0.1-1) ...
正在处理用于 ureadahead (0.100.0-19) 的触发器 ...
ureadahead will be reprofiled on next reboot
正在设置 nvidia-docker (1.0.1-1) ...
Configuring user
Setting up permissions
正在处理用于 ureadahead (0.100.0-19) 的触发器 ...
dong@cvad:~$ sudo pkill -SIGHUP dockerd
#test------------------------------test-------------------------------test
dong@cvad:~$ sudo docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi
Unable to find image 'nvidia/cuda:latest' locally
latest: Pulling from nvidia/cuda
898c46f3b1a1: Pull complete
63366dfa0a50: Pull complete
041d4cd74a92: Pull complete
6e1bee0f8701: Pull complete
131dbe7c254d: Pull complete
5bca6b05dcd6: Pull complete
0d286a7b6e12: Pull complete
5776d2c6371d: Pull complete
768e84e7fc24: Pull complete
Digest: sha256:eba1dc5810e40f60625ee797d618a6bd11be24cb67bc6647a4d36392202bb013
Status: Downloaded newer image for nvidia/cuda:latest
docker: Error response from daemon: OCI runtime create failed: container_linux.go:345: starting container process caused "process_linux.go:424: container init caused \"process_linux.go:407: running prestart hook 1 caused \\\"error running hook: exit status 1, stdout: , stderr: exec command: [/usr/bin/nvidia-container-cli --load-kmods configure --ldconfig=@/sbin/ldconfig.real --device=all --compute --utility --require=cuda>=10.1 brand=tesla,driver>=384,driver<385 brand=tesla,driver>=410,driver<411 --pid=18238 /var/lib/docker/overlay2/9125ce2210561fc0820c654d4d95828a66b48367be02a67ac548ed635cc3d87b/merged]\\\\nnvidia-container-cli: requirement error: unsatisfied condition: brand = tesla\\\\n\\\"\"": unknown.
dong@cvad:~$ sudo docker version
Client:
Version: 18.09.5
API version: 1.39
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:44:24 2019
OS/Arch: linux/amd64
Experimental: false
Server: Docker Engine - Community
Engine:
Version: 18.09.5
API version: 1.39 (minimum version 1.12)
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:10:53 2019
OS/Arch: linux/amd64
Experimental: false
dong@cvad:~$ sudo nvidia-docker version
NVIDIA Docker: 1.0.1
Client:
Version: 18.09.5
API version: 1.39
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:44:24 2019
OS/Arch: linux/amd64
Experimental: false
Server: Docker Engine - Community
Engine:
Version: 18.09.5
API version: 1.39 (minimum version 1.12)
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:10:53 2019
OS/Arch: linux/amd64
Experimental: false
dong@cvad:~$ sudo docker image ls
REPOSITORY TAG IMAGE ID CREATED SIZE
nvidia/cuda latest 3517732c5437 3 weeks ago 2.74GB
hello-world latest fce289e99eb9 3 months ago 1.84kB
autoware/autoware 1.7.0-kinetic 4dd12ac386c0 11 months ago 9.28GB
dong@cvad:~$ sudo docker image rm 351
Untagged: nvidia/cuda:latest
Untagged: nvidia/cuda@sha256:eba1dc5810e40f60625ee797d618a6bd11be24cb67bc6647a4d36392202bb013
Deleted: sha256:3517732c5437ffd87d7c144e1d62cadef128c07a6ccf29f5eb1877d491b0304d
Deleted: sha256:966a04f6408c5c189cd54530de4d3874f202def3de20fb6f2c7d30a39afe2a53
Deleted: sha256:0d53da6826e47b0378024562532b5046946a3bb361ebd17efabcd67b257b7fd2
Deleted: sha256:6c6dbf9c188adebee58cf5d0d3f246c9c45d38adc5aaebf3f433c848be3dc6e9
Deleted: sha256:efde8f73f882aa3a2babd65103b465c7d48410d6824cd2909552aa21d5067079
Deleted: sha256:678a66a6df0cbb0c989b81d83cc2d30ddf1af0414479be003ee7a40e1f2e9953
Deleted: sha256:e783d8ee44ce099d51cbe699f699a04e43c9af445d85d8576f0172ba92e4e16c
Deleted: sha256:cc7fae10c2d465c5e4b95167987eaa53ae01a13df6894493efc5b28b95c1bba2
Deleted: sha256:99fc3504db138523ca958c0c1887dd5e8b59f8104fbd6fd4eed485c3e25d2446
Deleted: sha256:762d8e1a60542b83df67c13ec0d75517e5104dee84d8aa7fe5401113f89854d9
dong@cvad:~$ sudo docker image ls
REPOSITORY TAG IMAGE ID CREATED SIZE
hello-world latest fce289e99eb9 3 months ago 1.84kB
autoware/autoware 1.7.0-kinetic 4dd12ac386c0 11 months ago 9.28GB
dong@cvad:~$ sudo nvidia-docker version
NVIDIA Docker: 1.0.1
Client:
Version: 18.09.5
API version: 1.39
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:44:24 2019
OS/Arch: linux/amd64
Experimental: false
Server: Docker Engine - Community
Engine:
Version: 18.09.5
API version: 1.39 (minimum version 1.12)
Go version: go1.10.8
Git commit: e8ff056
Built: Thu Apr 11 04:10:53 2019
OS/Arch: linux/amd64
Experimental: false
dong@cvad:~$
调整版本后，发现还是不行，所以我现在想的是，nvidia-docker检查的时候，是直接pull cuda lasted，
我现在手动安装cuda 8.0试试看。
5、
安装nvidia/cuda:8.0，
[end]