python3自定义kubernetes的调度器（二）_要求使用python编写一个kubernetes调度器,监听pod的变化,当检测到有pending或

1、集群是kubernetes1.19，关于集群搭建等之前有记录过。
2、使用python3之kubernetes-client-python自定义kubernetes集群的调度器，代码如下：

from kubernetes import client
import kubernetes as k8s
import numpy as np

# k8s自定义调度器
# python : 3.8
# kubernetes : V17.14.0a1
# numpy  : V1.20.1

# 获取可用节点node
def getUseNode(k8sCoreV1api):
    # 获取所有节点
    nodeInstance = k8sCoreV1api.list_node()
    # 存放可用节点的节点名称
    useNodeName = []
    for i in nodeInstance.items:
        if i.status.conditions[-1].status == "True" and i.status.conditions[-1].type == "Ready":
            useNodeName.append(i.metadata.name)
    return useNodeName

# 获取需要调度的pod
def getScheduledPod(k8sCoreV1api,namespace):
    podInstance = k8sCoreV1api.list_namespaced_pod(namespace)
    # 获取需要调度的pod，也就是处于pending状态的pod
    scheduledPodName = []
    for i in podInstance.items:
        if i.status.phase == 'Pending' and i.spec.node_name is None:
            scheduledPodName.append(i.metadata.name)
    return scheduledPodName

# 调度pod，将pod调度到随机抽取的一个node上去（这里没有使用什么算法，只是随机抽取一个可用节点）
def podBinding(k8sCoreV1api,podName, nodeName, namespace):
    target = client.V1ObjectReference()
    target.kind = "Node"
    target.api_version = "v1"
    target.name = nodeName

    meta = client.V1ObjectMeta()
    meta.name = podName
    body = client.V1Binding(target=target)
    body.target = target
    body.metadata = meta
    # 下面会抛出异常，所以必须加try，不然程序就终止
    try:
        k8sCoreV1api.create_namespaced_binding(namespace, body)
        return True
    except Exception as e:
        """
        注意这段话！！
        create_namespaced_binding() throws exception:
        Invalid value for `target`, must not be `None`
        or
        despite the fact this exception is being thrown,
        Pod is bound to a Node and Pod is running
        """
        print('exception' + str(e))
        return False


# 4、将节点绑定上去
def podScheduling(k8sCoreV1api,useNodeName,scheduledPodName,namespace):
    nodeName = np.random.choice(useNodeName)
    print("选中节点：",nodeName)
    for podNAame in scheduledPodName:
        print("开始调度")
        re = podBinding(k8sCoreV1api,podNAame ,nodeName,namespace)
        print("完成一个pod的调度")

# 主方法，是入口
if __name__ == '__main__':
    namespace = "default"
    # 1、加载配置文件 这里的配置文件是集群中的.kube/config 文件，直接去集群中粘贴过来即可
    k8s.config.load_kube_config(config_file="kubeconfig.yaml")
    k8sCoreV1api = client.CoreV1Api()
    # 2、获取可用节点
    useNodeName = getUseNode(k8sCoreV1api)
    # 3、获取指定命名空间中处于待调度的pod
    scheduledPodName = getScheduledPod(k8sCoreV1api,namespace)
    # 4、将待调度的pod绑定到可用节点上，随机选取一个可用节点绑定上去
    podScheduling(k8sCoreV1api,useNodeName,scheduledPodName,namespace)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80

已经测试过了，可行。