在开始 kiali 亲和性调度之前,先演示一个简单的例子介绍 pod 选择调度到指定 node:
使用命令查看当前所有 k8s 节点:
[root@k8s-master ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
k8s-master Ready master 5h11m v1.18.1k8s-node01 Ready <none> 5h8m v1.18.1
现在给 k8s-w-206 这个节点打上一个标签,标签内容为 name: xiao,命令如下:
kubectl label node k8s-node01 name=xiao
编写 pod 资源文件flaskapp-deployment.yaml,文件中使用 nodeSelector 指定该 pod 要调度到 k8s-node01节点之上
apiVersion: apps/v1
kind: Deployment
metadata:
name: flaskapp-1
spec:
selector:
matchLabels:
run: flaskapp-1
replicas: 1
template:
metadata:
labels:
run: flaskapp-1
spec:
containers:
- name: flaskapp-1
image: jcdemo/flaskapp
ports:
- containerPort: 5000
nodeSelector:
name: xiao
部署 flaskapp-deployment.yaml,发现 pod 果然被调度到了 k8s-node01 这个 node,效果如下:
[root@k8s-master ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
flaskapp-1-58b69c66f9-hv498 1/1 Running 0 7m30s 10.244.1.30 k8s-node01 <none> <none>
上面举例 pod 使用 nodeSelector 选择 node,这就是最简单的 k8s 调度方式。
节点亲和性调度策略示例:
apiVersion: v1
kind: Pod
metadata:
name: with-node-affinity
spec:
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: kubernetes.io/e2e-az-name
operator: In
values:
- e2e-az1
- e2e-az2
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 1
preference:
matchExpressions:
- key: another-node-label-key
operator: In
values:
- another-node-label-value
containers:
- name: with-node-affinity
image: k8s.gcr.io/pause:2.0
举一个生活的例子,以前去医院看病,病人(pod)不能挑医生(node),排队叫到谁就是谁,冷冰冰完全没有亲和性而言;如今可以网上挂号了,病人也可以挑选中意的医生,这样就有了亲和性,说明社会进步了。
当然病人在挑选医生的过程中也会有两种情况:一种是硬性(required),比如非要某医生,即使他忙,也愿意一直等下去;还有一种是软性(prefered),比如优先选择某医生,但是如果真不行,其他医生也未尝不可。
下面的理论可以对照上面的例子。
节点亲和性,也就是 NodeAffinity,用来控制 pod 部署或者不能部署在哪台机器上。
节点亲和性调度策略分为硬策略分为软策略和硬策略两种方式。硬策略是如果没有满足条件的节点,就会不断重试直到条件满足了为止;软策略是如果没有满足条件的节点,pod 就会忽略这条规则,继续完成调度过程。
节点亲和性软硬策略的语法分别介绍如下。
requiredDuringSchedulingIgnoredDuringExecution:
pod 必须部署到满足条件的节点上,如果节点不满足条件,就不停重试。
preferredDuringSchedulingIgnoredDuringExecution:
pod 优先部署到满足条件的节点,如果节点不满足条件,就忽略这些条件,调度到其他节点。
apiVersion: apps/v1
kind: Deployment
metadata:
name: redis-cache
spec:
selector:
matchLabels:
app: store
replicas: 3
template:
metadata:
labels:
app: store
spec:
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- store
topologyKey: "kubernetes.io/hostname"
containers:
- name: redis-server
image: redis:3.2-alpine
创建了一个Deployment,副本数为3,指定了反亲和规则如上所示,pod的label为app:store,那么pod调度的时候将不会调度到node上已经运行了label为app:store的pod了,这样就会使得Deployment的三副本分别部署在不同的host的node上.
apiVersion: apps/v1
kind: Deployment
metadata:
name: redis-cache
spec:
selector:
matchLabels:
app: store
replicas: 3
template:
metadata:
labels:
app: store
spec:
affinity:
podAntiAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- web-store
topologyKey: "kubernetes.io/hostname"
podAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- store
topologyKey: "kubernetes.io/hostname"
containers:
- name: web-app
image: nginx:1.12-alpine
在一个例子中基础之上,要求pod的亲和性满足requiredDuringSchedulingIgnoredDuringExecution中topologyKey=”kubernetes.io/hostname”,并且node上需要运行有app=store的label.
举例3
apiVersion: apps/v1
kind: Deployment
metadata:
name: web-server
spec:
selector:
matchLabels:
app: web-server
replicas: 3
template:
metadata:
labels:
app: web-server
spec:
affinity:
podAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- web-store
topologyKey: "kubernetes.io/hostname"
containers:
- name: web-app
image: hub.easystack.io/library/nginx:1.9.0
在一些应用中,pod副本之间需要共享cache,需要将pod运行在一个节点之上
本文参考链接: