k8s入门（一）之pod创建、label使用、污点、亲和性、RS

一、创建一个pod

---

[root@master01 ~]# kubectl create ns prod
[root@master01 ~]# cat pod.yaml
apiVersion: v1
kind: Pod
metadata:
  name: pod-demo
  namespace: prod
  labels:
     app: myapp
spec:
  containers:
  - name: test1
    image: busybox:latest
    command:
    - "/bin/sh"
    - "-c"
    - "sleep 100000000000"
  - name: test2
    image: busybox:latest
    args: ["sleep","100000000000000"]
    
[root@master01 ~]# kubectl create -f pod.yaml        ###创建pod
[root@master01 ~]# kubectl describe pod pod-demo -n prod   ###查看日志
[root@master01 ~]# kubectl get pod pod-demo -n prod -o wide
NAME       READY   STATUS    RESTARTS   AGE     IP          NODE       NOMINATED NODE   READINESS GATES
pod-demo   2/2     Running   0          4m41s   10.0.0.70   master03   <none>           <none>


[root@master01 ~]# kubectl -n prod exec -it pod/pod-demo test1 -- sh      ###进入容器
Defaulted container "test1" out of: test1, test2
/ #

[root@master01 ~]# kubectl delete -f pod.yaml      ###删除pod

二、label使用

---

优先级：nodeName -> nodeSelector -> 污点（Taints） ->亲和性（Affinity）

1、创建httpd

[root@master01 ~]# cat web-pod.yaml
apiVersion: v1
kind: Pod
metadata:
  name: web
  labels:
     app: as
     rel: stable
spec:
  containers:
  - name: web
    image: httpd:latest
    ports:
    - containerPort: 80

[root@master01 ~]# kubectl apply -f web-pod.yaml

2、label常用操作

[root@master01 ~]# kubectl get pod
NAME   READY   STATUS    RESTARTS   AGE
web    1/1     Running   0          75s

[root@master01 ~]# kubectl get pod --show-labels     ###查看所有label
NAME   READY   STATUS    RESTARTS   AGE   LABELS
web    1/1     Running   0          87s   app=as,rel=stable

[root@master01 ~]# kubectl get pod -l app=as       ###匹配app=as的label
NAME   READY   STATUS    RESTARTS   AGE
web    1/1     Running   0          2m

[root@master01 ~]# kubectl get pod -l app!=test   ###匹配app不等于test的label
NAME   READY   STATUS    RESTARTS   AGE
web    1/1     Running   0          2m23s

[root@master01 ~]# kubectl get pod -l 'app in (test,as)'       ###匹配app为test或者as的label
NAME   READY   STATUS    RESTARTS   AGE
web    1/1     Running   0          3m7s
[root@master01 ~]# kubectl get pod -l 'app notin (test,as)'   ###匹配app不为test或者as的label
No resources found in default namespace.

[root@master01 ~]# kubectl get pod -L app               ### 匹配所有app
NAME   READY   STATUS    RESTARTS   AGE     APP
web    1/1     Running   0          3m34s   as

[root@master01 ~]# kubectl get pod --show-labels
NAME   READY   STATUS    RESTARTS   AGE     LABELS
web    1/1     Running   0          4m21s   app=as,rel=stable

[root@master01 ~]# kubectl label pod web rel=canary --overwrite        ###修改label
[root@master01 ~]# kubectl get pod --show-labels
NAME   READY   STATUS    RESTARTS   AGE     LABELS
web    1/1     Running   0          4m49s   app=as,rel=canary

[root@master01 ~]# kubectl label nodes master02 disktype=ssd     ###给节点打label
[root@master01 ~]# kubectl label nodes master03 env=prod        
[root@master01 ~]# kubectl get nodes --show-labels

3、nodeSelector使用

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 2
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80
      nodeSelector:
        disktype: ssd

[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
deployment.apps/nginx-deployment created
[root@master01 ~]# kubectl get pod -owide
NAME                                READY   STATUS    RESTARTS   AGE   IP           NODE       NOMINATED NODE   READINESS GATES
nginx-deployment-697c567f87-gddsh   1/1     Running   0          9s    10.0.1.77    master02   <none>           <none>
nginx-deployment-697c567f87-lltln   1/1     Running   0          9s    10.0.1.144   master02   <none>           <none>
web                                 1/1     Running   0          25m   10.0.0.38    master03   <none>           <none>

4、nodeName使用

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 2
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80
      nodeName: master02

[root@master01 ~]# kubectl apply -f nginx-deploy.yaml

三、污点和容忍度

---

1、查看Taints

[root@master01 ~]# kubectl describe node master01|grep Taints
Taints:             <none>

[root@master01 ~]# kubectl taint node master01 node-role.kubernets.io/master="":NoSchedule
[root@master01 ~]# kubectl describe node master01|grep Taints
Taints:             node-role.kubernets.io/master:NoSchedule

[root@master01 ~]# kubectl taint node master01 node-role.kubernets.io/master-     ###删除
[root@master01 ~]# kubectl get pod -A -owide
[root@master01 ~]# kubectl describe pod cilium-operator-58bf55d99b-zxmxp -n kube-system|grep Tolerations
Tolerations:                 op=Exists

2、添加污点

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl taint node master02 node-type=production:NoSchedule
[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml

# 如果之前已经部署在master02，不会把之前的驱离
[root@master01 ~]# kubectl get pod -A -owide|grep nginx
default       nginx-deployment-66b9f7ff85-6hxqf   1/1     Running             0              22s    10.0.4.110    master01   <none>           <none>
default       nginx-deployment-66b9f7ff85-d8jsz   1/1     Running             0              22s    10.0.2.168    node01     <none>           <none>
default       nginx-deployment-66b9f7ff85-dmptx   1/1     Running             0              22s    10.0.0.223    master03   <none>           <none>
default       nginx-deployment-66b9f7ff85-h5wts   1/1     Running             0              22s    10.0.4.201    master01   <none>           <none>
default       nginx-deployment-66b9f7ff85-jxjw4   0/1     ContainerCreating   0              22s    <none>        node02     <none>           <none>

3、容忍污点

（1）容忍NoSchedule

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      tolerations:
      - key: node-type
        operator: Equal
        value: production
        effect: NoSchedule
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -A -owide|grep nginx
default       nginx-deployment-7cffd544c8-7d677   1/1     Running   0              11s    10.0.2.15     node01     <none>           <none>
default       nginx-deployment-7cffd544c8-nszsr   1/1     Running   0              11s    10.0.0.229    master03   <none>           <none>
default       nginx-deployment-7cffd544c8-v4cw2   1/1     Running   0              11s    10.0.4.192    master01   <none>           <none>
default       nginx-deployment-7cffd544c8-v75dd   1/1     Running   0              11s    10.0.1.52     master02   <none>           <none>
default       nginx-deployment-7cffd544c8-x9mdm   1/1     Running   0              11s    10.0.3.93     node02     <none>           <none>

（2）容忍NoExecute

会立刻把master03的机器迁移走

[root@master01 ~]# kubectl get pod -A -owide|grep nginx
default       nginx-deployment-7cffd544c8-7d677   1/1     Running   0              4m10s   10.0.2.15     node01     <none>           <none>
default       nginx-deployment-7cffd544c8-nszsr   1/1     Running   0              4m10s   10.0.0.229    master03   <none>           <none>
default       nginx-deployment-7cffd544c8-v4cw2   1/1     Running   0              4m10s   10.0.4.192    master01   <none>           <none>
default       nginx-deployment-7cffd544c8-v75dd   1/1     Running   0              4m10s   10.0.1.52     master02   <none>           <none>
default       nginx-deployment-7cffd544c8-x9mdm   1/1     Running   0              4m10s   10.0.3.93     node02     <none>           <none>
[root@master01 ~]# kubectl taint node master03 node-type=test:NoExecute
[root@master01 ~]# kubectl get pod -A -owide|grep nginx
default       nginx-deployment-7cffd544c8-7d677   1/1     Running   0              4m41s   10.0.2.15     node01     <none>           <none>
default       nginx-deployment-7cffd544c8-fpn9l   1/1     Running   0              6s      10.0.4.252    master01   <none>           <none>
default       nginx-deployment-7cffd544c8-v4cw2   1/1     Running   0              4m41s   10.0.4.192    master01   <none>           <none>
default       nginx-deployment-7cffd544c8-v75dd   1/1     Running   0              4m41s   10.0.1.52     master02   <none>           <none>
default       nginx-deployment-7cffd544c8-x9mdm   1/1     Running   0              4m41s   10.0.3.93     node02     <none>           <none>

（3）设置tolerationSeconds

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      tolerations:
      - key: node-type
        operator: Equal
        value: production
        effect: NoSchedule
      - effect: NoExecute
        key: node.kubernetes.io/not-ready
        operator: Exists
        tolerationSeconds: 300
      - effect: NoExecute
        key: node.kubernetes.io/unreachable
        operator: Exists
        tolerationSeconds: 30
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -A -owide|grep nginx
default       nginx-deployment-5c5996fdc-7fjl9   1/1     Running   0              9m47s   10.0.1.124    master02   <none>           <none>
default       nginx-deployment-5c5996fdc-8qh2r   1/1     Running   0              9m47s   10.0.4.109    master01   <none>           <none>
default       nginx-deployment-5c5996fdc-b2nbz   1/1     Running   0              9m47s   10.0.3.200    node02     <none>           <none>
default       nginx-deployment-5c5996fdc-hdnsl   1/1     Running   0              9m47s   10.0.2.251    node01     <none>           <none>
default       nginx-deployment-5c5996fdc-j78nd   1/1     Running   0              9m47s   10.0.4.135    master01   <none>           <none>

[root@master01 ~]# kubectl describe pod nginx-deployment-5c5996fdc-7fjl9 |grep -A 3 Tolerations
Tolerations:                 node-type=production:NoSchedule
                             node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                             node.kubernetes.io/unreachable:NoExecute op=Exists for 30s

（4）删除之前污点

[root@master01 ~]# kubectl taint node master02 node-type-
[root@master01 ~]# kubectl taint node master03 node-type-

四、亲和性

---

1、node亲和性

如果master02不够话，会调度到其他节点上

[root@master01 ~]# kubectl label nodes master02 gpu=true
node/master02 labeled
[root@master01 ~]# kubectl get nodes --show-labels
NAME       STATUS   ROLES    AGE   VERSION   LABELS
master01   Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=master01,kubernetes.io/os=linux,node.kubernetes.io/node=
master02   Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disktype=ssd,gpu=true,kubernetes.io/arch=amd64,kubernetes.io/hostname=master02,kubernetes.io/os=linux,node.kubernetes.io/node=
master03   Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,env=prod,kubernetes.io/arch=amd64,kubernetes.io/hostname=master03,kubernetes.io/os=linux,node.kubernetes.io/node=
node01     Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node01,kubernetes.io/os=linux,node.kubernetes.io/node=
node02     Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node02,kubernetes.io/os=linux,node.kubernetes.io/node=

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: gpu
                operator: In
                values:
                - "true"
      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -owide|grep nginx
nginx-deployment-596d5dfbb-6v9wc   1/1     Running   0          19s   10.0.1.49    master02   <none>           <none>
nginx-deployment-596d5dfbb-gzsx7   1/1     Running   0          19s   10.0.1.213   master02   <none>           <none>
nginx-deployment-596d5dfbb-ktwhw   1/1     Running   0          19s   10.0.1.179   master02   <none>           <none>
nginx-deployment-596d5dfbb-r96cz   1/1     Running   0          19s   10.0.1.183   master02   <none>           <none>
nginx-deployment-596d5dfbb-tznqh   1/1     Running   0          19s   10.0.1.210   master02   <none>           <none>

2、preferredDuringSchedulingIgnoredDuringExecution

[root@master01 ~]# kubectl label node master02 available-zone=zone1
[root@master01 ~]# kubectl label node master03 available-zone=zone2
[root@master01 ~]# kubectl label nodes master02 share-type=dedicated
[root@master01 ~]# kubectl label nodes master03 share-type=shared
[root@master01 ~]# kubectl get nodes --show-labels
NAME       STATUS   ROLES    AGE   VERSION   LABELS
master01   Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=master01,kubernetes.io/os=linux,node.kubernetes.io/node=
master02   Ready    <none>   24h   v1.26.5   available-zone=zone1,beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disktype=ssd,gpu=true,kubernetes.io/arch=amd64,kubernetes.io/hostname=master02,kubernetes.io/os=linux,node.kubernetes.io/node=,share-type=dedicated
master03   Ready    <none>   24h   v1.26.5   available-zone=zone2,beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,env=prod,kubernetes.io/arch=amd64,kubernetes.io/hostname=master03,kubernetes.io/os=linux,node.kubernetes.io/node=,share-type=shared
node01     Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node01,kubernetes.io/os=linux,node.kubernetes.io/node=
node02     Ready    <none>   24h   v1.26.5   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node02,kubernetes.io/os=linux,node.kubernetes.io/node=

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 10
  template:
    metadata:
      labels:
        app: nginx
    spec:
      affinity:
        nodeAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 80
            preference:
              matchExpressions:
              - key: available-zone
                operator: In
                values:
                - zone1
          - weight: 20
            preference:
              matchExpressions:
              - key: share-type
                operator: In
                values:
                - dedicated

      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -A -o wide|grep nginx
default       nginx-deployment-5794574555-22xqh   1/1     Running   0               13s     10.0.1.28     master02   <none>           <none>
default       nginx-deployment-5794574555-5b8kq   1/1     Running   0               13s     10.0.1.168    master02   <none>           <none>
default       nginx-deployment-5794574555-6fm9h   1/1     Running   0               13s     10.0.1.152    master02   <none>           <none>
default       nginx-deployment-5794574555-7gqm7   1/1     Running   0               13s     10.0.3.163    node02     <none>           <none>
default       nginx-deployment-5794574555-7mp2p   1/1     Running   0               13s     10.0.1.19     master02   <none>           <none>
default       nginx-deployment-5794574555-8rrmw   1/1     Running   0               13s     10.0.4.107    master01   <none>           <none>
default       nginx-deployment-5794574555-c6s7z   1/1     Running   0               13s     10.0.1.52     master02   <none>           <none>
default       nginx-deployment-5794574555-f94p6   1/1     Running   0               13s     10.0.0.125    master03   <none>           <none>
default       nginx-deployment-5794574555-hznds   1/1     Running   0               13s     10.0.1.163    master02   <none>           <none>
default       nginx-deployment-5794574555-vx85n   1/1     Running   0               13s     10.0.2.179    node01     <none>           <none>

3、pod亲和性

（1）创建一个busybox

发现生成在master03

[root@master01 ~]# kubectl run backend -l app=backend --image busybox -- sleep 99999999
[root@master01 ~]# kubectl get pod -owide
NAME                                READY   STATUS    RESTARTS   AGE     IP           NODE       NOMINATED NODE   READINESS GATES
backend                             1/1     Running   0          7m35s   10.0.0.242   master03   <none>           <none>


[root@master01 ~]# kubectl get pod --show-labels
NAME      READY   STATUS    RESTARTS   AGE   LABELS
backend   1/1     Running   0          45s   app=backend

（2）nginx亲和busybox（方式一）

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      affinity:
        podAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
          - topologyKey: "kubernetes.io/hostname"
            labelSelector:
              matchExpressions:
              - key: app
                operator: In
                values:
                - backend

      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl delete -f nginx-deploy.yaml

（3）nginx亲和busybox（方式二）

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      affinity:
        podAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
          - topologyKey: "kubernetes.io/hostname"
            labelSelector:
              matchLabels:
                app: backend

      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml

（4）优先级亲和pod

优先部署在满足的节点上

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 5
  template:
    metadata:
      labels:
        app: nginx
    spec:
      affinity:
        podAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 80
            podAffinityTerm:
              topologyKey: "kubernetes.io/hostname"
              labelSelector:
                matchLabels:
                  app: backend

      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80


[root@master01 ~]# kubectl delete -f nginx-deploy.yaml
[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -owide
NAME                                READY   STATUS    RESTARTS   AGE   IP           NODE       NOMINATED NODE   READINESS GATES
backend                             1/1     Running   0          19m   10.0.0.242   master03   <none>           <none>
nginx-deployment-6b6b6646cf-2g2nt   1/1     Running   0          30s   10.0.0.158   master03   <none>           <none>
nginx-deployment-6b6b6646cf-62xlh   1/1     Running   0          30s   10.0.0.4     master03   <none>           <none>
nginx-deployment-6b6b6646cf-6nhl4   1/1     Running   0          30s   10.0.0.123   master03   <none>           <none>
nginx-deployment-6b6b6646cf-gfbxt   1/1     Running   0          30s   10.0.0.125   master03   <none>           <none>
nginx-deployment-6b6b6646cf-sg6bx   1/1     Running   0          30s   10.0.4.206   master01   <none>           <none>

（5）pod非亲和

因为5个节点都安装了，因为第六个无法进行安装

[root@master01 ~]# cat nginx-deploy.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: frontend
spec:
  selector:
    matchLabels:
      app: frontend
  replicas: 6
  template:
    metadata:
      labels:
        app: frontend
    spec:
      affinity:
        podAntiAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
          - topologyKey: "kubernetes.io/hostname"
            labelSelector:
              matchLabels:
                app: frontend

      containers:
      - name: nginx
        image: nginx:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl apply -f nginx-deploy.yaml
[root@master01 ~]# kubectl get pod -owide
NAME                       READY   STATUS    RESTARTS   AGE   IP           NODE       NOMINATED NODE   READINESS GATES
backend                    1/1     Running   0          32m   10.0.0.242   master03   <none>           <none>
frontend-84bdd684b-825hs   1/1     Running   0          74s   10.0.1.145   master02   <none>           <none>
frontend-84bdd684b-b8vd9   0/1     Pending   0          74s   <none>       <none>     <none>           <none>
frontend-84bdd684b-m7d79   1/1     Running   0          74s   10.0.2.20    node01     <none>           <none>
frontend-84bdd684b-pj9vv   1/1     Running   0          74s   10.0.4.42    master01   <none>           <none>
frontend-84bdd684b-qn6lb   1/1     Running   0          74s   10.0.3.91    node02     <none>           <none>
frontend-84bdd684b-tl84p   1/1     Running   0          74s   10.0.0.214   master03   <none>           <none>

[root@master01 ~]# kubectl describe pod frontend-84bdd684b-b8vd9
Events:
  Type     Reason            Age                  From               Message
  ----     ------            ----                 ----               -------
  Warning  FailedScheduling  102s (x2 over 103s)  default-scheduler  0/5 nodes are available: 5 node(s) didn't match pod anti-affinity rules. preemption: 0/5 nodes are available: 5 No preemption victims found for incoming pod..

五、RC与RS

---

适应于不需要更新的程序，最大的问题不支持滚动更新
RS比RC主要多标签选择器

1、RC体验

[root@master01 ~]# cat rc.yaml
apiVersion: v1
kind: ReplicationController
metadata:
  name: rc-test
spec:
  replicas: 3
  selector:
    app: rc-pod
  template:
    metadata:
      labels:
        app: rc-pod
    spec:
      containers:
      - name: rc-test
        image: httpd:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl apply -f rc.yaml
[root@master01 ~]# kubectl get pod -owide
NAME            READY   STATUS    RESTARTS   AGE   IP           NODE       NOMINATED NODE   READINESS GATES
rc-test-bjrqn   1/1     Running   0          35s   10.0.0.244   master03   <none>           <none>
rc-test-srbfn   1/1     Running   0          35s   10.0.4.207   master01   <none>           <none>
rc-test-tjrmf   1/1     Running   0          35s   10.0.0.67    master03   <none>           <none>

[root@master01 ~]# kubectl get rc
NAME      DESIRED   CURRENT   READY   AGE
rc-test   3         3         3       94s


[root@master01 ~]# kubectl describe rc rc-test|grep -i Replicas
Replicas:     3 current / 3 desired

2、RS初体验

[root@master01 ~]# cat rs.yaml
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: rs-test
  labels:
    app: guestbool
    tier: frontend
spec:
  replicas: 3
  selector:
    matchLabels:
      tier: frontend
  template:
    metadata:
      labels:
        tier: frontend
    spec:
      containers:
      - name: rs-test
        image: httpd:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80

[root@master01 ~]# kubectl apply -f rs.yaml
[root@master01 ~]# kubectl get pod -o wide
NAME            READY   STATUS    RESTARTS   AGE     IP           NODE       NOMINATED NODE   READINESS GATES
rs-test-7dbz8   1/1     Running   0          2m18s   10.0.4.245   master01   <none>           <none>
rs-test-m8tqv   1/1     Running   0          2m18s   10.0.1.202   master02   <none>           <none>
rs-test-pp6b6   1/1     Running   0          2m18s   10.0.0.93    master03   <none>           <none>

[root@master01 ~]# kubectl get pod rs-test-7dbz8 -oyaml      # 通过uuid进行的管理

3、RS接管pod

注意标签相同，必须所有标签都相同，执行运行pod启动不了的

[root@master01 ~]# cat pod-rs.yaml
apiVersion: v1
kind: Pod
metadata:
  name: pod1
  labels:
    tier: frontend
spec:
  containers:
  - name: test1
    image: httpd:latest
    imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Pod
metadata:
  name: pod2
  labels:
    tier: frontend
spec:
  containers:
  - name: test2
    image: httpd:latest
    imagePullPolicy: IfNotPresent

[root@master01 ~]# kubectl delete -f rs.yaml
[root@master01 ~]# kubectl apply -f pod-rs.yaml
[root@master01 ~]# kubectl apply -f rs.yaml

[root@master01 ~]# kubectl get pod
NAME            READY   STATUS    RESTARTS   AGE
pod1            1/1     Running   0          102s
pod2            1/1     Running   0          102s
rs-test-zw2sj   1/1     Running   0          3s

[root@master01 ~]# kubectl delete -f rs.yaml

4、修改label测试

[root@master01 ~]# kubectl apply -f rs.yaml
[root@master01 ~]# kubectl get pod --show-labels
NAME            READY   STATUS    RESTARTS   AGE   LABELS
rs-test-cn6cc   1/1     Running   0          11s   tier=frontend
rs-test-jxwfb   1/1     Running   0          11s   tier=frontend
rs-test-sztrt   1/1     Running   0          11s   tier=frontend
[root@master01 ~]# kubectl label pod rs-test-cn6cc tier=canary --overwrite
[root@master01 ~]# kubectl get pod --show-labels
NAME            READY   STATUS    RESTARTS   AGE   LABELS
rs-test-cn6cc   1/1     Running   0          60s   tier=canary
rs-test-jxwfb   1/1     Running   0          60s   tier=frontend
rs-test-qw5fj   1/1     Running   0          14s   tier=frontend
rs-test-sztrt   1/1     Running   0          60s   tier=frontend

[root@master01 ~]# kubectl get pod  -l tier=frontend
NAME            READY   STATUS    RESTARTS   AGE
rs-test-jxwfb   1/1     Running   0          3m30s
rs-test-qw5fj   1/1     Running   0          2m44s
rs-test-sztrt   1/1     Running   0          3m30s

[root@master01 ~]# kubectl get pod  -l 'tier in (canary,frontend)'
NAME            READY   STATUS    RESTARTS   AGE
rs-test-cn6cc   1/1     Running   0          3m48s
rs-test-jxwfb   1/1     Running   0          3m48s
rs-test-qw5fj   1/1     Running   0          3m2s
rs-test-sztrt   1/1     Running   0          3m48s

[root@master01 ~]# kubectl get pod  -l 'tier notin (canary)'
NAME            READY   STATUS    RESTARTS   AGE
rs-test-jxwfb   1/1     Running   0          4m2s
rs-test-qw5fj   1/1     Running   0          3m16s
rs-test-sztrt   1/1     Running   0          4m2s

5、多标签

（1）测试rs

[root@master01 ~]# cat rs.yaml
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: rs-test
  labels:
    app: guestbool
    tier: frontend
spec:
  replicas: 3
  selector:
    matchLabels:
      tier: frontend
    matchExpressions:
      - key: tier
        operator: In
        values:
          - frontend
          - canary
      - {key: app, operator: In, values: [guestbool, test]}
  template:
    metadata:
      labels:
        tier: frontend
        app: guestbool
    spec:
      containers:
      - name: rs-test
        image: httpd:latest
        imagePullPolicy: IfNotPresent
        ports:
        - containerPort: 80


[root@master01 ~]# kubectl delete -f rs.yaml
[root@master01 ~]# kubectl apply -f rs.yaml
replicaset.apps/rs-test created
[root@master01 ~]# kubectl get pod --show-labels
NAME            READY   STATUS              RESTARTS   AGE   LABELS
rs-test-2srdx   1/1     Running             0          3s    app=guestbool,tier=frontend
rs-test-ftzsk   1/1     Running             0          3s    app=guestbool,tier=frontend
rs-test-v2nlj   0/1     ContainerCreating   0          3s    app=guestbool,tier=frontend

（2）测试pod-rs

[root@master01 ~]# kubectl label pod rs-test-2srdx tier=canary --overwrite
[root@master01 ~]# kubectl get pod --show-labels
NAME            READY   STATUS    RESTARTS   AGE     LABELS
rs-test-2srdx   1/1     Running   0          2m46s   app=guestbool,tier=canary
rs-test-4t548   1/1     Running   0          4s      app=guestbool,tier=frontend
rs-test-ftzsk   1/1     Running   0          2m46s   app=guestbool,tier=frontend
rs-test-v2nlj   1/1     Running   0          2m46s   app=guestbool,tier=frontend

[root@master01 ~]# cat pod-rs.yaml
apiVersion: v1
kind: Pod
metadata:
  name: pod1
  labels:
    tier: canary
spec:
  containers:
  - name: test1
    image: httpd:latest
    imagePullPolicy: IfNotPresent
---
apiVersion: v1
kind: Pod
metadata:
  name: pod2
  labels:
    tier: frontend
spec:
  containers:
  - name: test2
    image: httpd:latest
    imagePullPolicy: IfNotPresent

[root@master01 ~]# kubectl apply -f pod-rs.yaml
[root@master01 ~]# kubectl get pod --show-labels
NAME            READY   STATUS    RESTARTS   AGE     LABELS
pod1            1/1     Running   0          9s      tier=canary
pod2            1/1     Running   0          9s      tier=frontend
rs-test-2srdx   1/1     Running   0          3m49s   app=guestbool,tier=canary
rs-test-4t548   1/1     Running   0          67s     app=guestbool,tier=frontend
rs-test-ftzsk   1/1     Running   0          3m49s   app=guestbool,tier=frontend
rs-test-v2nlj   1/1     Running   0          3m49s   app=guestbool,tier=frontend