下载方式

pip install kubernetes

官方例子

用法参考官方的例子

from kubernetes import client, config

config.load_kube_config()

v1 = client.CoreV1Api()

print("Listing pods with their IPs:")

ret = v1.list_pod_for_all_namespaces(watch=False)

for i in ret.items:
    print("%s\t%s\t%s" %
    (i.status.pod_ip, i.metadata.namespace, i.metadata.name))

可以在文档里找到对应的Class和method，可以看下相信的方法介绍

Deployment资源操作

create

自动化功能更多就是就通过获取元数据来创建需要的资源

示例AppsV1Api下的create_namespaced_deployment方法

• include_uninitialized为是否在response中包含initialized的容器信息
• pretty是否添加tty终端
• dry_run未知

官方示例简化一下就是

from kubernetes import client, config

config.load_kube_config()

DEPLOYMENT_NAME = "nginx-deployment"

def create_deployment_object():
    # Configureate Pod template container
    container = client.V1Container(
        name="nginx",
        image="nginx:1.7.9",
        ports=[client.V1ContainerPort(container_port=80)])
    # Create and configurate a spec section
    template = client.V1PodTemplateSpec(
        metadata=client.V1ObjectMeta(labels={"app": "nginx"}),
        spec=client.V1PodSpec(containers=[container]))
    # Create the specification of deployment
    spec = client.ExtensionsV1beta1DeploymentSpec(
        replicas=3,
        template=template)
    # Instantiate the deployment object
    deployment = client.ExtensionsV1beta1Deployment(
        api_version="extensions/v1beta1",
        kind="Deployment",
        metadata=client.V1ObjectMeta(name=DEPLOYMENT_NAME),
        spec=spec)
    return deployment

extensions_v1beta1 = client.ExtensionsV1beta1Api()

deployment = create_deployment_object()

extensions_v1beta1.create_namespaced_deployment(namespace="default", body=deployment)

执行结果

$ kubectl get deployments
NAME               READY   UP-TO-DATE   AVAILABLE   AGE
nginx-deployment   3/3     3            3           90s

更多的配置还需要去V1Deployment获取，这边也可以直接以yaml或者json的方式

$ kubectl get deployments nginx-deployment  -o json
$ kubectl delete deployments nginx-deployment  
deployment.extensions "nginx-deployment" deleted

然后将一些需要kubernetes自动生成的配置和一些无用的配置删除，直接将Deployment定义为一个json

deployment = {
    "apiVersion": "extensions/v1beta1",
    "kind": "Deployment",
    "metadata": {
        "labels": {
            "app": "nginx"
        },
        "name": "nginx-deployment",
        "namespace": "default",
    },
    "spec": {
        "replicas": 3,
        "selector": {
            "matchLabels": {
                "app": "nginx"
            }
        },
        "template": {
            "metadata": {
                "labels": {
                    "app": "nginx"
                }
            },
            "spec": {
                "containers": [
                    {
                        "image": "nginx:1.7.9",
                        "name": "nginx",
                        "ports": [
                            {
                                "containerPort": 80,
                                "protocol": "TCP"
                            }
                        ],
                    }
                ],
                "restartPolicy": "Always",
            }
        }
    }
}

直接通过字典的方式可以，构建起来非常方便，也可以通过open的方式打开一个yaml作为body

这样也是可以的，create_namespaced_deployment方法的返回结果是一个kubernetes.client.models.extensions_v1beta1_deployment.ExtensionsV1beta1Deployment对象

Delete

官方文档上是说只需要namespace和namespace，但是如果只传这两个参数会报错TypeError: delete_namespaced_deployment() takes exactly 4 arguments (3 given)

当设置body为None的时候也是报错，不过报出相关的代码路径和行数了，这边查看了一下

$ /usr/lib/python2.7/site-packages/kubernetes/client/apis/extensions_v1beta1_api.py +1783
        ...
        if kwargs.get('async_req'):
            return self.delete_namespaced_deployment_with_http_info(name, namespace, body, **kwargs)
        else:
            (data) = self.delete_namespaced_deployment_with_http_info(name, namespace, body, **kwargs)
            return data
        ...

可以看到这里需要body参数的，这边我传了个空字典

extensions_v1beta1.delete_namespaced_deployment(namespace="default",name=DEPLOYMENT_NAME, body={})

Read

c = extensions_v1beta1.read_namespaced_deployment(namespace="default", name=DEPLOYMENT_NAME)
print type(c.api_version)
print type(c.spec)
print type(c.spec.replicas)
print type(c.spec.template)

执行结果

<type 'str'>
<class 'kubernetes.client.models.extensions_v1beta1_deployment_spec.ExtensionsV1beta1DeploymentSpec'>
<type 'int'>
<class 'kubernetes.client.models.v1_pod_template_spec.V1PodTemplateSpec'>

可以到返回的有些是对象，有些则是值了，更多参考V1Deployment对象的定义

Patch

先获取原来的deployment对象，修改对应的值

old_deployment_obj = extensions_v1beta1.read_namespaced_deployment(namespace="default", name=DEPLOYMENT_NAME)
print old_deployment_obj.spec.template.spec.containers[0].image
old_deployment_obj.spec.template.spec.containers[0].image = "nginx:1.9.1"
new_deployment_obj = extensions_v1beta1.patch_namespaced_deployment(namespace="default",name=DEPLOYMENT_NAME, body=old_deployment_obj)
print new_deployment_obj.spec.template.spec.containers[0].image

Replace

Replace和Patch在参数上看起来是一样的

Rollback

回滚的deployment可以使用create_namespaced_deployment_rollback

通过api创建的默认是使用了--record参数的

extensions_v1beta1.create_namespaced_deployment_rollback(namespace="default",name=DEPLOYMENT_NAME, body={ "name" : DEPLOYMENT_NAME, "rollback_to" : { "revision" : 3} })

详细的信息可以通过kubectl rollout history进行回滚

$ kubectl rollout history deployment/nginx-deployment --revision 2

部署版本等信息最好还是自行保存或者调控一下会比较好

创建自定义对象

参考官方示例

from __future__ import print_function

from pprint import pprint

import kubernetes
from kubernetes import config
from kubernetes.rest import ApiException

config.load_kube_config()
api_instance = kubernetes.ThirdPartyResources()

namespace = 'default'
resource = 'repos'
fqdn = 'git.k8s.com'

body = {}
body['apiVersion'] = "git.k8s.com/v1"
body['kind'] = "RePo"
body['metadata'] = {}
body['metadata']['name'] = "blog-repo"
body['repo'] = "github.com/user/my-blog"
body['username'] = "username"
body['password'] = "password"
body['branch'] = "branch"



try: 
    # Create a Resource
    api_response = api_instance.apis_fqdn_v1_namespaces_namespace_resource_post(
        namespace, fqdn, resource, body)
    pprint(api_response)
except ApiException as e:
    print(
        "Exception when calling DefaultApi->apis_fqdn_v1_namespaces_namespace_resource_post: %s\n" % 
        e)

watch

https://github.com/kubernetes-client/python/blob/master/kubernetes/docs/V1WatchEvent.md

from kubernetes import client,config,watch

config.load_kube_config()

v1=client.CoreV1Api()

def main():
    w = watch.Watch()
    for event in w.stream(v1.list_namespaced_pod,"default"):
        try:
            print("Event: %s, Status: %s, PodName: %s" % (event['type'], event['object'].status.phase, event['object'].metadata.name))
        except:
            print("Event: %s, Status: %s, PodName: %s" % (event['type'], None, event['object'].metadata.name))

if __name__ == '__main__':
    main()

获取Pod

$ kubectl get pod
NAME                                READY   STATUS    RESTARTS   AGE
nginx-deployment-76bf4969df-7gmcp   1/1     Running   0          43m
nginx-deployment-76bf4969df-cmv66   1/1     Running   0          4m37s
nginx-deployment-76bf4969df-djnwj   1/1     Running   0          8m40s

执行脚本

$ python watchk8s.py
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-cmv66
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-7gmcp

删除掉一个Pod

$ kubectl delete pod nginx-deployment-76bf4969df-djnwj
pod "nginx-deployment-76bf4969df-djnwj" deleted

$ python watchk8s.py
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-cmv66
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj
Event: ADDED, Status: Running, PodName: nginx-deployment-76bf4969df-7gmcp


Event: MODIFIED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj
Event: ADDED, Status: Pending, PodName: nginx-deployment-76bf4969df-hw4v5
Event: MODIFIED, Status: Pending, PodName: nginx-deployment-76bf4969df-hw4v5
Event: MODIFIED, Status: Pending, PodName: nginx-deployment-76bf4969df-hw4v5
Event: MODIFIED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj
Event: MODIFIED, Status: Running, PodName: nginx-deployment-76bf4969df-hw4v5
Event: MODIFIED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj
Event: DELETED, Status: Running, PodName: nginx-deployment-76bf4969df-djnwj

可以看到有四个新创建的nginx-deployment-76bf4969df-hw4v5的Event，分别对应的就是描述的四个Reason

$ kubectl describe pod nginx-deployment-76bf4969df-hw4v5
...
Events:
  Type    Reason     Age   From               Message
  ----    ------     ----  ----               -------
  Normal  Scheduled  40s   default-scheduler  Successfully assigned default/nginx-deployment-76bf4969df-hw4v5 to node-01
  Normal  Pulled     39s   kubelet, node-01   Container image "nginx:1.7.9" already present on machine
  Normal  Created    39s   kubelet, node-01   Created container
  Normal  Started    39s   kubelet, node-01   Started container

可以在Pod，Deployment或者Service创建或者删除的时候进行一些自定义的操作