Kubernetes Pod 调度约束

Kubernetes Pod 调度约束

• 可以将pod调度到指定的节点Node内
• 默认：根据节点资源利用率等分配Node节点。

• nodeName用于将Pod调度到指定的Node名称上
• nodeSelector用于将Pod调度到匹配Label的Node上

工作流程
K8s通过watch实现组件工作。
1、管理员通过命令创建Pod-->apiserver接收到-->状态写入到etcd-->scheduler通过watch获取etcd中获取新的Pod-->通过算法选出pod应该调度到哪些节点内-->绑定到新的节点并更新到etcd中
2、kubelet通过watch从etcd中获取到绑定到自己节点的pod-->将pod通过docker run启动运行--> 在将状态（运行状态）更新到etcd中，根据kubelet周期上报
3、管理员查看pod状态 --> 查找etcd中pod状态 --> 返回给用户

# 使用方法
apiVersion: v1
kind: Pod
metadata:
  name: pod-example
  labels:
    app: nginx
spec:
  nodeName: 192.168.31.65
  containers:
  - name: nginx
    image: nginx:1.15

apiVersion: v1
kind: Pod
metadata:
  name: pod-example
spec:
  nodeSelector:
    env_role: dev
  containers:
  - name: nginx
    image: nginx:1.15

实践（指定NodeIP）

1、创建测试pod
vim pod5.yaml

apiVersion: v1
kind: Pod
metadata:
  name: pod-example
  labels:
    app: nginx
spec:
  nodeName: 192.168.1.111
  containers:
  - name: nginx
    image: nginx:1.15

2、创建文件

kubectl create -f pod5.yaml

3、查看pod调度节点

NAME READY STATUS RESTARTS AGE IP NODE 
pod-example 1/1 Running 0 42s 172.17.1.4 192.168.1.111 <none>

4、查看详情；直接绕过调度器

...
Events:
  Type    Reason   Age    From                    Message
  ----    ------   ----   ----                    -------
  Normal  Pulling  9m8s   kubelet, 192.168.1.111  pulling image "nginx:1.15"
  Normal  Pulled   8m48s  kubelet, 192.168.1.111  Successfully pulled image "nginx:1.15"
  Normal  Created  8m48s  kubelet, 192.168.1.111  Created container
  Normal  Started  8m47s  kubelet, 192.168.1.111  Started container

实践(指定标签)

1、给指定Node设置标签 ；为 team团队ab队（自定义=自定义）

kubectl label nodes 192.168.1.111 team=a
kubectl label nodes 192.168.1.110 team=b

2、查看标签

NAME STATUS ROLES AGE VERSION LABELS
192.168.1.110 Ready <none> 2d15h v1.12.1 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/hostname=192.168.1.110,team=b
192.168.1.111 Ready <none> 2d15h v1.12.1 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/hostname=192.168.1.111,team=a

3、创建文件通过标签指定Node
vim pod6.yaml

apiVersion: v1
kind: Pod
metadata:
  name: pod-example
spec:
  nodeSelector:
    team: b
  containers:
  - name: nginx
    image: nginx:1.15

4、查看状态

NAME READY STATUS RESTARTS AGE IP NODE
pod-example 1/1 Running 0 29s 172.17.84.2 192.168.1.110 <none>

5、查看详情；走默认调度

...
                 node.kubernetes.io/unreachable:NoExecute for 300s
Events:
  Type    Reason     Age   From                    Message
  ----    ------     ----  ----                    -------
  Normal  Scheduled  83s   default-scheduler       Successfully assigned default/pod-example to 192.168.1.110
  Normal  Pulling    81s   kubelet, 192.168.1.110  pulling image "nginx:1.15"
  Normal  Pulled     67s   kubelet, 192.168.1.110  Successfully pulled image "nginx:1.15"
  Normal  Created    66s   kubelet, 192.168.1.110  Created container
 mal  Started    66s   kubelet, 192.168.1.110  Started container