K8s控制器使用的详解

最新推荐文章于 2025-03-17 22:39:23 发布

0.0-0

最新推荐文章于 2025-03-17 22:39:23 发布

阅读量555

点赞数 16

文章标签： kubernetes 容器云原生

本文链接：https://blog.youkuaiyun.com/HUHUD521/article/details/142965429

版权

replicaset控制器

replicaset功能

ReplicaSet 是下一代的 Replication Controller，官方推荐使用ReplicaSet

ReplicaSet和Replication Controller的唯一区别是选择器的支持，ReplicaSet支持新的基于集合的选择器需求

ReplicaSet 确保任何时间都有指定数量的 Pod 副本在运行

虽然 ReplicaSets 可以独立使用，但今天它主要被Deployments 用作协调 Pod 创建、删除和更新的机制

replicaset参数说明

参数名称	字段类型	参数说明
spec	Object	详细定义对象，固定值就写Spec
spec.replicas	integer	指定维护pod数量
spec.selector	Object	Selector是对pod的标签查询，与pod数量匹配
spec.selector.matchLabels	string	指定Selector查询标签的名称和值，以key：value方式指定
spec.template	Object	指定对pod的描述信息，比如lab标签，运行容器的信息等
spec.template.metadata	Object	指定pod属性
spec.template.metadata.labels	string	指定pod标签
spec.template.spec	Object	详细定义对象
spec.template.spec.containers	list	Spec对象的容器列表定义
spec.template.spec.containers.name	string	指定容器名称
spec.template.spec.containers.image	string	指定容器镜像

replicaset 示例

[root@k8s-master ~]# kubectl create deployment replicaset --image myapp:v1 --dry-run=client -o yaml > replicaset.yml

[root@k8s-master ~]# vim replicaset.yml
apiVersion: apps/v1
kind: ReplicaSet
metadata:
  name: replicaset		#指定pod名称，一定小写，如果出现大写报错
spec:
  replicas: 2			#指定维护pod数量为2
  selector:				#指定检测匹配方式
    matchLabels:		#指定匹配方式为匹配标签
      app: myapp		#指定匹配的标签为app=myapp

  template:				#模板，当副本数量不足时，会根据下面的模板创建pod副本
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - image: myapp:v1
        name: myapp

[root@k8s-master ~]# kubectl apply -f replicaset.yml
replicaset.apps/replicaset created


[root@k8s-master ~]# kubectl get pods  --show-labels
NAME               READY   STATUS    RESTARTS   AGE   LABELS
replicaset-l4xnr   1/1     Running   0          96s   app=myapp
replicaset-t2s5p   1/1     Running   0          96s   app=myapp


#replicaset是通过标签匹配pod
[root@k8s-master ~]# kubectl label pod replicaset-l4xnr app=timinglee --overwrite
pod/replicaset-l4xnr labeled
[root@k8s-master ~]# kubectl get pods  --show-labels
NAME               READY   STATUS    RESTARTS   AGE     LABELS
replicaset-gd5fh   1/1     Running   0          2s      app=myapp		#新开启的pod
replicaset-l4xnr   1/1     Running   0          3m19s   app=timinglee
replicaset-t2s5p   1/1     Running   0          3m19s   app=myapp

#恢复标签后
[root@k8s2 pod]# kubectl label pod replicaset-example-q2sq9 app- 
[root@k8s2 pod]# kubectl get pod --show-labels
NAME                       READY   STATUS    RESTARTS   AGE     LABELS
replicaset-example-q2sq9   1/1     Running   0          3m14s   app=nginx
replicaset-example-th24v   1/1     Running   0          3m14s   app=nginx
replicaset-example-w7zpw   1/1     Running   0          3m14s   app=nginx

#replicaset自动控制副本数量，pod可以自愈
[root@k8s-master ~]# kubectl delete pods replicaset-t2s5p
pod "replicaset-t2s5p" deleted

[root@k8s-master ~]# kubectl get pods --show-labels
NAME               READY   STATUS    RESTARTS   AGE     LABELS
replicaset-l4xnr   1/1     Running   0          5m43s   app=myapp
replicaset-nxmr9   1/1     Running   0          15s     app=myapp


回收资源
[root@k8s2 pod]# kubectl delete -f rs-example.yml

deployment 控制器

deployment控制器的功能

为了更好的解决服务编排的问题，kubernetes在V1.2版本开始，引入了Deployment控制器。

Deployment控制器并不直接管理pod，而是通过管理ReplicaSet来间接管理Pod

Deployment管理ReplicaSet，ReplicaSet管理Pod

Deployment 为 Pod 和 ReplicaSet 提供了一个申明式的定义方法

在Deployment中ReplicaSet相当于一个版本

deployment控制器示例

#生成yaml文件
[root@k8s-master ~]# kubectl create deployment deployment --image myapp:v1  --dry-run=client -o yaml > deployment.yml

[root@k8s-master ~]# vim deployment.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  replicas: 4
  selector:
    matchLabels:
      app: myapp
  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - image: myapp:v1
        name: myapp
#建立pod
root@k8s-master ~]# kubectl apply -f deployment.yml
deployment.apps/deployment created

#查看pod信息
[root@k8s-master ~]# kubectl get pods --show-labels
NAME                          READY   STATUS    RESTARTS   AGE   LABELS
deployment-5d886954d4-2ckqw   1/1     Running   0          23s   app=myapp,pod-template-hash=5d886954d4
deployment-5d886954d4-m8gpd   1/1     Running   0          23s   app=myapp,pod-template-hash=5d886954d4
deployment-5d886954d4-s7pws   1/1     Running   0          23s   app=myapp,pod-template-hash=5d886954d4
deployment-5d886954d4-wqnvv   1/1     Running   0          23s   app=myapp,pod-template-hash=5d886954d4

版本迭代

[root@k8s-master ~]# kubectl get pods -o wide
NAME                          READY   STATUS    RESTARTS   AGE     IP            NODE        NOMINATED NODE   READINESS GATES
deployment-5d886954d4-2ckqw   1/1     Running   0          2m40s   10.244.2.14   k8s-node2   <none>           <none>
deployment-5d886954d4-m8gpd   1/1     Running   0          2m40s   10.244.1.17   k8s-node1   <none>           <none>
deployment-5d886954d4-s7pws   1/1     Running   0          2m40s   10.244.1.16   k8s-node1   <none>           <none>
deployment-5d886954d4-wqnvv   1/1     Running   0          2m40s   10.244.2.15   k8s-node2   <none>           <none>

#pod运行容器版本为v1
[root@k8s-master ~]# curl  10.244.2.14
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

[root@k8s-master ~]# kubectl describe deployments.apps deployment
Name:                   deployment
Namespace:              default
CreationTimestamp:      Sun, 01 Sep 2024 23:19:10 +0800
Labels:                 <none>
Annotations:            deployment.kubernetes.io/revision: 1
Selector:               app=myapp
Replicas:               4 desired | 4 updated | 4 total | 4 available | 0 unavailable
StrategyType:           RollingUpdate
MinReadySeconds:        0
RollingUpdateStrategy:  25% max unavailable, 25% max surge			#默认每次更新25%


#更新容器运行版本
[root@k8s-master ~]# vim deployment.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  replicas: 4
  selector:
    matchLabels:
      app: myapp

  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - image: myapp:v2				#更新为版本2
        name: myapp

[root@k8s2 pod]# kubectl apply -f deployment-example.yaml

#更新过程
[root@k8s-master ~]# kubectl get pods   -w
NAME                          READY   STATUS    RESTARTS   AGE
deployment-5d886954d4-8kb28   1/1     Running   0          48s
#测试更新效果
[root@k8s-master ~]# kubectl get pods -o wide
NAME                          READY   STATUS    RESTARTS   AGE   IP            NODE        NOMINATED NODE   READINESS GATES
deployment-7f4786db9c-967fk   1/1     Running   0          10s   10.244.1.26   k8s-node1   <none>           <none>
deployment-7f4786db9c-cvb9k   1/1     Running   0          10s   10.244.2.24   k8s-node2   <none>           <none>
deployment-7f4786db9c-kgss4   1/1     Running   0          9s    10.244.1.27   k8s-node1   <none>           <none>
deployment-7f4786db9c-qts8c   1/1     Running   0          9s    10.244.2.25   k8s-node2   <none>           <none>

[root@k8s-master ~]# curl  10.244.1.26
Hello MyApp | Version: v2 | <a href="hostname.html">Pod Name</a>

版本回滚

[root@k8s-master ~]# vim deployment.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  replicas: 4
  selector:
    matchLabels:
      app: myapp

  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - image: myapp:v1				#回滚到之前版本
        name: myapp


[root@k8s-master ~]# kubectl apply -f deployment.yml
deployment.apps/deployment configured

#测试回滚效果
[root@k8s-master ~]# kubectl get pods -o wide
NAME                          READY   STATUS    RESTARTS   AGE   IP            NODE        NOMINATED NODE   READINESS GATES
deployment-5d886954d4-dr74h   1/1     Running   0          8s    10.244.2.26   k8s-node2   <none>           <none>
deployment-5d886954d4-thpf9   1/1     Running   0          7s    10.244.1.29   k8s-node1   <none>           <none>
deployment-5d886954d4-vmwl9   1/1     Running   0          8s    10.244.1.28   k8s-node1   <none>           <none>
deployment-5d886954d4-wprpd   1/1     Running   0          6s    10.244.2.27   k8s-node2   <none>           <none>

[root@k8s-master ~]# curl  10.244.2.26
Hello MyApp | Version: v1 | <a href="hostname.html">Pod Name</a>

滚动更新策略

[root@k8s-master ~]# vim deployment.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  minReadySeconds: 5		#最小就绪时间，指定pod每隔多久更新一次
  replicas: 4
  strategy:					#指定更新策略
    rollingUpdate:
      maxSurge: 1			#比定义pod数量多几个
      maxUnavailable: 0		#比定义pod个数少几个
  selector:
    matchLabels:
      app: myapp

  template:
    metadata:
      labels:
        app: myapp
    spec:
      containers:
      - image: myapp:v1
        name: myapp
[root@k8s2 pod]# kubectl apply -f deployment-example.yaml

daemonset控制器

daemonset功能

在每个节点上运行集群存储 DaemonSet，例如 glusterd、ceph。

在每个节点上运行日志收集 DaemonSet，例如 fluentd、logstash。

在每个节点上运行监控 DaemonSet，例如 Prometheus Node Exporter、zabbix agent等

一个简单的用法是在所有的节点上都启动一个 DaemonSet，将被作为每种类型的 daemon 使用

一个稍微复杂的用法是单独对每种 daemon 类型使用多个 DaemonSet，但具有不同的标志，并且对不同硬件类型具有不同的内存、CPU 要求

daemonset 示例

[root@k8s2 pod]# cat daemonset-example.yml
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: daemonset-example
spec:
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      tolerations:		#对于污点节点的容忍
      - effect: NoSchedule
        operator: Exists
      containers:
      - name: nginx
        image: nginx

[root@k8s-master ~]# kubectl get pods  -o wide
NAME              READY   STATUS    RESTARTS   AGE   IP            NODE         NOMINATED NODE   READINESS GATES
daemonset-87h6s   1/1     Running   0          47s   10.244.0.8    k8s-master   <none>           <none>
daemonset-n4vs4   1/1     Running   0          47s   10.244.2.38   k8s-node2    <none>           <none>
daemonset-vhxmq   1/1     Running   0          47s   10.244.1.40   k8s-node1    <none>           <none>


#回收
[root@k8s2 pod]# kubectl delete -f daemonset-example.yml

job 控制器

job控制器功能

当Job创建的pod执行成功结束时，Job将记录成功结束的pod数量

当成功结束的pod达到指定的数量时，Job将完成执行

job 控制器示例

[root@k8s2 pod]# vim job.yml
apiVersion: batch/v1
kind: Job
metadata:
  name: pi
spec:
  completions: 6		#一共完成任务数为6		
  parallelism: 2		#每次并行完成2个
  template:
    spec:
      containers:
      - name: pi
        image: perl:5.34.0
        command: ["perl",  "-Mbignum=bpi", "-wle", "print bpi(2000)"]	计算Π的后2000位
      restartPolicy: Never						#关闭后不自动重启
  backoffLimit: 4								#运行失败后尝试4重新运行

[root@k8s2 pod]# kubectl apply -f job.yml

cronjob 控制器

cronjob 控制器功能

Cron Job 创建基于时间调度的 Jobs。

CronJob控制器以Job控制器资源为其管控对象，并借助它管理pod资源对象，

CronJob可以以类似于Linux操作系统的周期性任务作业计划的方式控制其运行时间点及重复运行的方式。

CronJob可以在特定的时间点(反复的)去运行job任务。

cronjob 控制器示例

[root@k8s2 pod]# vim cronjob.yml
apiVersion: batch/v1
kind: CronJob
metadata:
  name: hello
spec:
  schedule: "* * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hello
            image: busybox
            imagePullPolicy: IfNotPresent
            command:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure

[root@k8s2 pod]# kubectl apply -f cronjob.yml