nvidia-docker无法对话: http://localhost:3476/docker/cli/json

Question

nvidia-docker无法与http://localhost:3476/docker/cli/json对话

Traceback (most recent call last):
  File "/usr/local/bin/nvidia-docker-compose", line 43, in <module>
    resp = request.urlopen('http://{0}/docker/cli/json'.format(args.nvidia_docker_host)).read().decode()
  File "/usr/lib/python2.7/urllib2.py", line 154, in urlopen
    return opener.open(url, data, timeout)
  File "/usr/lib/python2.7/urllib2.py", line 429, in open
    response = self._open(req, data)
  File "/usr/lib/python2.7/urllib2.py", line 447, in _open
    '_open', req)
  File "/usr/lib/python2.7/urllib2.py", line 407, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.7/urllib2.py", line 1228, in http_open
    return self.do_open(httplib.HTTPConnection, req)
  File "/usr/lib/python2.7/urllib2.py", line 1198, in do_open
    raise URLError(err)
urllib2.URLError: <urlopen error [Errno 111] Connection refused>

Answer 1

全新安装的nvidia-docker-compose修复了此问题：

wget -P /tmp https://github.com/NVIDIA/nvidia-docker/releases/download/v1.0.1/nvidia-docker_1.0.1-1_amd64.deb
sudo dpkg -i /tmp/nvidia-docker*.deb && rm /tmp/nvidia-docker*.deb

然后测试它：

测试nvidia-smi

nvidia-docker run --rm nvidia/cuda nvidia-smi

Answer 2

也遇到了这个问题，一个客户没有设法运行nvidia-docker-compose。事实证明，即使在重新安装docker和nvidia-docker之后，nvidia-docker向localhost:3476上的docker发出的查询也没有得到任何响应(参见nvidia-docker-compose code here)

我设法解决了这个问题，生成了一个手工的docker-compose文件，因为它们被证明是相当简单的，遵循这个例子，用你的nvidia驱动程序版本替换375.66，并放入和你的显卡一样多的/dev/nvidian行(我没有试图把服务放在单独的GPU上，但是去做吧!)：

services:
  exampleservice0:
    devices:
    - /dev/nvidia0
    - /dev/nvidia1
    - /dev/nvidiactl
    - /dev/nvidia-uvm
    - /dev/nvidia-uvm-tools
    environment:
    - EXAMPLE_ENV_VARIABLE=example
    image: company/image
    volumes:
    - ./disk:/disk
    - nvidia_driver_375.66:/usr/local/nvidia:ro
version: '2'
volumes:
  media: null
  nvidia_driver_375.66:
    external: true

然后用一个经典的docker-compose命令运行这个手工创建的docker-compose文件。