kubernetes-daemonset

保证pod高可用，监控pod运行状态，保证无时无刻都有期望的pod副本运行，要求同一个daemonset部署出来的pod一定要运行不同节点，保证节点 (有且只有一个pod),一般做agent代理程序或监控，没副本数，等于节点数量。
可以直接拷贝deployment的yaml文件直接更改

[root@master01 ~]# cat deploy.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  creationTimestamp: null
  labels:
    app: my-dep
apiVersion: apps/v1
kind: DaemonSet
metadata:
  creationTimestamp: null
  labels:
    app: my-dae
  name: my-dae
spec:
  selector:
    matchLabels:
      app: my-daemon
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: my-daemon
    spec:
      containers:
      - image: nginx
        name: nginx
        imagePullPolicy: IfNotPresent
        resources: {}
                                                                                                                                               

创建daemonset ,默认不会调度到master 上，因为master有污点。后面再写污点和容忍度。
只调度至work工作节点

[root@master01 ~]# kubectl create  -f daemonset.yaml 
daemonset.apps/my-dae created
[root@master01 ~]# kubectl get daemonsets.apps  my-dae 
NAME     DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR   AGE
my-dae   2         2         2       2            2           <none>          9s
[root@master01 ~]# kubectl get pods -o wide
NAME           READY   STATUS    RESTARTS   AGE   IP               NODE     NOMINATED NODE   READINESS GATES
my-dae-g5gww   1/1     Running   0          15s   10.244.196.185   node01   <none>           <none>
my-dae-r2kmg   1/1     Running   0          15s   10.244.140.118   node02   <none>           <none>

删除一个pod ，daemonset 也会从新拉起一个新的pod

[root@master01 ~]# kubectl get pods -o wide
NAME           READY   STATUS    RESTARTS   AGE   IP               NODE     NOMINATED NODE   READINESS GATES
my-dae-g5gww   1/1     Running   0          15s   10.244.196.185   node01   <none>           <none>
my-dae-r2kmg   1/1     Running   0          15s   10.244.140.118   node02   <none>           <none>
[root@master01 ~]# 
[root@master01 ~]# kubectl delete pods my-dae-r2kmg 
pod "my-dae-r2kmg" deleted
[root@master01 ~]# kubectl get pods -o wide
NAME           READY   STATUS    RESTARTS   AGE    IP               NODE     NOMINATED NODE   READINESS GATES
my-dae-g5gww   1/1     Running   0          3m3s   10.244.196.185   node01   <none>           <none>
my-dae-r82dp   1/1     Running   0          2s     10.244.140.119   node02   <none>           <none>

创建一个容忍master污点的daemonset

[root@master01 ~]# kubectl create -f daemon.yaml 
daemonset.apps/my-dae created
[root@master01 ~]# kubectl get daemonsets.apps my-dae 
NAME     DESIRED   CURRENT   READY   UP-TO-DATE   AVAILABLE   NODE SELECTOR            AGE
my-dae   3         3         3       3            3           kubernetes.io/os=linux   11s
[root@master01 ~]# kubectl get pods -o wide
NAME           READY   STATUS    RESTARTS   AGE   IP               NODE       NOMINATED NODE   READINESS GATES
my-dae-km5jw   1/1     Running   0          16s   10.244.140.120   node02     <none>           <none>
my-dae-lkhs7   1/1     Running   0          16s   10.244.196.186   node01     <none>           <none>
my-dae-xnp6q   1/1     Running   0          16s   10.244.241.71    master01   <none>           <none>  #master
[root@master01 ~]# cat daemon.yaml 
apiVersion: apps/v1
kind: DaemonSet
metadata:
  creationTimestamp: null
  labels:
    app: my-dae
  name: my-dae
spec:
  selector:
    matchLabels:
      app: my-daemon
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: my-daemon
    spec:
      containers:
      - image: nginx
        name: nginx
        imagePullPolicy: IfNotPresent 
        resources: {}
      nodeSelector:   #node筛选器
        kubernetes.io/os: linux
      tolerations:   #容忍污点
      - key: "node-role.kubernetes.io/master"
        operator: "Exists"
        value: ""
        effect: "NoSchedule"

升级回滚 同deployment一样

[root@master01 ~]# kubectl set image daemonset.apps my-dae nginx=nginx:1.16.1 --record 
Flag --record has been deprecated, --record will be removed in the future
daemonset.apps/my-dae image updated
[root@master01 ~]# kubectl describe  daemonsets.apps my-dae 
Name:           my-dae
Selector:       app=my-daemon
Node-Selector:  kubernetes.io/os=linux
Labels:         app=my-dae
Annotations:    deprecated.daemonset.template.generation: 2
                kubernetes.io/change-cause: kubectl set image daemonset.apps my-dae nginx=nginx:1.16.1 --record=true
Desired Number of Nodes Scheduled: 3
Current Number of Nodes Scheduled: 3
Number of Nodes Scheduled with Up-to-date Pods: 2
Number of Nodes Scheduled with Available Pods: 2
Number of Nodes Misscheduled: 0
Pods Status:  2 Running / 1 Waiting / 0 Succeeded / 0 Failed
Pod Template:
  Labels:  app=my-daemon
  Containers:
   nginx:
    Image:        nginx:1.16.1
    Port:         <none>
    Host Port:    <none>
    Environment:  <none>
    Mounts:       <none>
  Volumes:        <none>
Events:
  Type    Reason            Age    From                  Message
  ----    ------            ----   ----                  -------
  Normal  SuccessfulCreate  3m24s  daemonset-controller  Created pod: my-dae-lkhs7
  Normal  SuccessfulCreate  3m24s  daemonset-controller  Created pod: my-dae-km5jw
  Normal  SuccessfulCreate  3m24s  daemonset-controller  Created pod: my-dae-xnp6q
  Normal  SuccessfulDelete  16s    daemonset-controller  Deleted pod: my-dae-lkhs7
  Normal  SuccessfulCreate  15s    daemonset-controller  Created pod: my-dae-g5rwb
  Normal  SuccessfulDelete  13s    daemonset-controller  Deleted pod: my-dae-xnp6q
  Normal  SuccessfulCreate  12s    daemonset-controller  Created pod: my-dae-zscbh
  [root@master01 ~]# kubectl rollout  history daemonset my-dae 
daemonset.apps/my-dae 
REVISION  CHANGE-CAUSE
1         <none>
2         kubectl set image daemonset.apps my-dae nginx=nginx:1.16.1 --record=true
回滚参数：
  strategy:
    rollingUpdate:
      maxSurge: 25%  初识创建的pod 这个值调的越大，副本更新速度越快 
      maxUnavailable: 25% 初识删除的pod  这个值越小，越能保证服务稳定，更新越平滑

daemonset一般做为代理程序和单机agent使用