python docker api,,开启Remote A


开启Remote API

docker默认是没有开启Remote API的,需要我们手动开启。编辑/lib/systemd/system/docker.service文件,

在文件里的ExecStart参数后面添加-H tcp://0.0.0.0:2375 -H unix:///var/run/docker.sock

然后重启sudo systemctl daemon-reload sudo service docker restart 在浏览器查看:

技术图片

备注:我这里用了JSONView-for-Chrome插件 ,用git下载下来,chrome设置 - 扩展程序 (地址栏输入chrome://extensions/)-> 选中 开发模式-> 点击"加载已解压的扩展程序" -> 选择插件目录(xxx\JSONView-for-Chrome-master\WebContent)

Python API

需要安装pip3 install docker-py,理论上docker-py可以干所有事情,但实际现在还不不怎么好用

import dockerclient = docker.DockerClient("http://192.168.100.5:2375")#打印现有的images名称#images =client.images.list()#for img in images:    #print(img.attrs["RepoTags"][0])#获取containerfor container in client.containers.list():    #print(dir(container))    print("container:"+container.name+" image:"+container.image.attrs["RepoTags"][0])#拉取镜像 相当于sudo docker pull alpineimage = client.images.pull("alpine")#运行镜像 如果detach=True,会立即返回一个container对象container = client.containers.run("alpine", ["touch", "/helloworld"], detach=True)container.wait()#提交新镜像image = container.commit("helloworld")
import dockerclient = docker.DockerClient("http://192.168.100.5:2375")#打印现有的images名称#images =client.images.list()#for img in images:    #print(img.attrs["RepoTags"][0])#获取containerfor container in client.containers.list():    #print(dir(container))    print("container:"+container.name+" image:"+container.image.attrs["RepoTags"][0])#拉取镜像 相当于sudo docker pull alpineimage = client.images.pull("alpine")#运行镜像 如果detach=True,会立即返回一个container对象container = client.containers.run("alpine", ["touch", "/helloworld"], detach=True)container.wait()#提交新镜像image = container.commit("helloworld")

参考:

https://docs.docker.com/develop/sdk/examples/

https://github.com/docker/docker-py

https://letong.gitbooks.io/docker/content/API/python_api.html

python docker api

评论关闭