Kubernetes(k8s)集群部署五、控制器(rs,rc,Deployment,DaemonSet、StatefulSet、Job、CronJob、HPA)

控制器

一、 Pod 的分类：

自主式 Pod：Pod 退出后不会被创建（没有控制器）
控制器管理的 Pod：在控制器的生命周期里，始终要维持 Pod 的副本数目

二、控制器类型：

Replication Controller和ReplicaSet
Deployment
DaemonSet
StatefulSet:前提：唯一的网络标识，持久化存储
Job
CronJob
HPA全称Horizontal Pod Autoscaler #自动化扩容

1 Replication Controller（rc）和ReplicaSet（rs）

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。

1.1ReplicaSet控制器示例：

官方文档
https://kubernetes.io/zh/docs/concepts/workloads/controllers/replicaset/

[root@server2 ~]# cat rs.yml 
apiVersion: apps/v1
kind: ReplicaSet
metadata:
    name: replicaset-example
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: myapp:v1

控制副本数
在这里插入图片描述

2.Deployment

Deployment 为 Pod 和 ReplicaSet 提供了一个申明式的定义方法。
典型的应用场景：
用来创建Pod和ReplicaSet
滚动更新和回滚
扩容和缩容
暂停与恢复

2.1Deployment控制器示例

deployment 连rs ，rs控制副本数
本身rs也是一个控制器，deployment主要更新，灰度发布

[root@server2 ~]# cat deployment.yml 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  replicas: 3
  selector:
    matchLabels:    ###pod 上匹配标签是nginx，才会进行操作
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: myapp:v1

[root@server2 ~]# kubectl get pod
NAME                          READY   STATUS    RESTARTS   AGE
demo                          1/1     Running   6          22h
deployment-6456d7c676-p4x6p   1/1     Running   0          17s
deployment-6456d7c676-rspg8   1/1     Running   0          17s
deployment-6456d7c676-t4bgd   1/1     Running   0          17s

[root@server2 ~]# kubectl  get pod --show-labels 
NAME                          READY   STATUS    RESTARTS   AGE     LABELS
demo                          1/1     Running   6          23h     run=demo
deployment-6456d7c676-p4x6p   1/1     Running   0          3m34s   app=nginx,pod-template-hash=6456d7c676

当修改标签的时候，会有一个新的pod，nginx标签依然保持3个副本

[root@server2 ~]# kubectl label pod deployment-6456d7c676-p4x6p app=myapp --overwrite 
pod/deployment-6456d7c676-p4x6p labeled
[root@server2 ~]# kubectl get pod
NAME                          READY   STATUS    RESTARTS   AGE
demo                          1/1     Running   6          23h
deployment-6456d7c676-p4x6p   1/1     Running   0          5m25s
deployment-6456d7c676-qgttz   1/1     Running   0          7s
deployment-6456d7c676-rspg8   1/1     Running   0          5m25s
deployment-6456d7c676-t4bgd   1/1     Running   0          5m25s
[root@server2 ~]# kubectl get pod -L app
NAME                          READY   STATUS    RESTARTS   AGE     APP
demo                          1/1     Running   6          23h     
deployment-6456d7c676-p4x6p   1/1     Running   0          6m16s   myapp
deployment-6456d7c676-qgttz   1/1     Running   0          58s     nginx
deployment-6456d7c676-rspg8   1/1     Running   0          6m16s   nginx
deployment-6456d7c676-t4bgd   1/1     Running   0          6m16s   nginx
[root@server2 ~]#

当把文件里面的镜像myapp:v1改变为myapp:v2时
回收rs里面所有的副本，创建新的rs，在新的rs创建新的pod，原本的rs不会删除，以便回滚，除非删除控制器，通过标签匹配pod，一旦改变标签，不受管控，因为控制器已经定义了标签

[root@server2 ~]# cat rs.yml 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: deployment
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: myapp:v2    ###进行更新

[root@server2 ~]# kubectl  get pod -L app
NAME                          READY   STATUS        RESTARTS   AGE     APP
demo                          1/1     Running       6          23h     
deployment-6456d7c676-p4x6p   1/1     Running       0          9m19s   myapp
deployment-6456d7c676-rspg8   0/1     Terminating   0          9m19s   nginx
deployment-6456d7c676-t4bgd   0/1     Terminating   0          9m19s   nginx
deployment-6d4f5bf58f-4qfbq   1/1     Running       0          13s     nginx
deployment-6d4f5bf58f-b6whd   1/1     Running       0          15s     nginx
deployment-6d4f5bf58f-mk949   1/1     Running       0          12s     nginx
[root@server2 ~]# kubectl  get pod -L app
NAME                          READY   STATUS    RESTARTS   AGE     APP
demo                          1/1     Running   6          23h     
deployment-6456d7c676-p4x6p   1/1     Running   0          9m28s   myapp
deployment-6d4f5bf58f-4qfbq   1/1     Running   0          22s     nginx
deployment-6d4f5bf58f-b6whd   1/1     Running   0          24s     nginx
deployment-6d4f5bf58f-mk949   1/1     Running   0          21s     nginx

之前的rs并不会被删除deployment-6456d7c676

[root@server2 ~]# kubectl get rs
NAME                    DESIRED   CURRENT   READY   AGE
deployment-6456d7c676   0         0         0       10m
deployment-6d4f5bf58f   3         3         3       66s

回滚
当再次改变文件----》 myapp:v1 又恢复为之前的状态

[root@server2 ~]# kubectl  get pod 
NAME                          READY   STATUS        RESTARTS   AGE
demo                          1/1     Running       6          23h
deployment-6456d7c676-jbdx9   1/1     Running       0          9s
deployment-6456d7c676-mzggd   1/1     Running       0          12s
deployment-6456d7c676-wpzbn   1/1     Running       0          10s
deployment-6d4f5bf58f-4qfbq   0/1     Terminating   0          101s
deployment-6d4f5bf58f-b6whd   0/1     Terminating   0          103s
deployment-6d4f5bf58f-mk949   0/1     Terminating   0          100s
[root@server2 ~]# kubectl  get pod 
NAME                          READY   STATUS    RESTARTS   AGE
demo                          1/1     Running   6          23h
deployment-6456d7c676-jbdx9   1/1     Running   0          24s
deployment-6456d7c676-mzggd   1/1     Running   0          27s
deployment-6456d7c676-wpzbn   1/1     Running   0          25s
[root@server2 ~]# kubectl get rs
NAME                    DESIRED   CURRENT   READY   AGE
deployment-6456d7c676   3         3         3       11m
deployment-6d4f5bf58f   0         0         0       2m11s

补充：标签
创建svc

[root@server2 ~]# kubectl expose deployment deployment --port=80
service/deployment exposed
[root@server2 ~]# kubectl get svc
NAME         TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)   AGE
deployment   ClusterIP   10.108.76.242   <none>        80/TCP    3s
kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP   9d
[root@server2 ~]# kubectl describe svc deployment 
Name:              deployment
Namespace:         default
Labels:            <none>
Annotations:       <none>
Selector:          app=nginx
Type:              ClusterIP
IP Families:       <none>
IP:                10.108.76.242
IPs:               10.108.76.242
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.1.57:80,10.244.1.58:80,10.244.1.59:80    ###三个
Session Affinity:  None
Events:            <none>

[root@server2 ~]# kubectl label pod deployment-6456d7c676-jbdx9  app=myapp --overwrite 
pod/deployment-6456d7c676-jbdx9 labeled
[root@server2 ~]# kubectl  get pod -L app
NAME                          READY   STATUS    RESTARTS   AGE   APP
demo                          1/1     Running   6          23h   
deployment-6456d7c676-b478b   1/1     Running   0          3s    nginx
deployment-6456d7c676-jbdx9   1/1     Running   0          14m   myapp
deployment-6456d7c676-mzggd   1/1     Running   0          14m   nginx
deployment-6456d7c676-wpzbn   1/1     Running   0          14m   nginx

[root@server2 ~]# kubectl edit svc deployment
service/deployment edited

selector:
app: myapp

[root@server2 ~]# kubectl describe svc deployment 
Name:              deployment
Namespace:         default
Labels:            <none>
Annotations:       <none>
Selector:          app=myapp
Type:              ClusterIP
IP Families:       <none>
IP:                10.108.76.242
IPs:               10.108.76.242
Port:              <unset>  80/TCP
TargetPort:        80/TCP
Endpoints:         10.244.1.59:80    ###根据标签选择
Session Affinity:  None
Events:            <none>

3.DaemonSet

DaemonSet 确保全部（或者某些）节点上运行一个 Pod 的副本。当有节点加入集群时，也会为他们新增一个 Pod 。当有节点从集群移除时，这些 Pod 也会被回收。删除 DaemonSet 将会删除它创建的所有 Pod。

DaemonSet 的典型用法：
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。

3.1DaemonSet控制器示例

一个节点布一个但只有一个server3节点，效果不太明显
[root@server1 harbor]# docker pull zabbix/zabbix-agent
docker tag zabbix/zabbix-agent:latest reg.westos.org/library/zabbix-agent:latest
docker push reg.westos.org/library/zabbix-agent:latest

[root@server2 ~]# cat daemonset.yml 
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: daemonset-example
  labels:
    k8s-app: zabbix-agent
spec:
  selector:
    matchLabels:
      name: zabbix-agent
  template:
    metadata:
      labels:
        name: zabbix-agent
    spec:
      containers:
      - name: zabbix-agent
        image: zabbix-agent

[root@server2 ~]# kubectl apply -f daemonset.yml 
daemonset.apps/daemonset-example created
[root@server2 ~]# kubectl  get pod
NAME                      READY   STATUS    RESTARTS   AGE
daemonset-example-gj4dm   1/1     Running   0          5s
demo                      1/1     Running   6          23h
[root@server2 ~]# kubectl  get pod -o wide
NAME                      READY   STATUS    RESTARTS   AGE   IP            NODE      NOMINATED NODE   READINESS GATES
daemonset-example-gj4dm   1/1     Running   0          14s   10.244.1.61   server3   <none>           <none>
demo                      1/1     Running   6          23h   10.244.1.40   server3   <none>           <none>

4.StatefulSet

StatefulSet 是用来管理有状态应用的工作负载 API 对象。实例之间有不对等关系，以及实例对外部数据有依赖关系的应用，称为“有状态应用”

StatefulSet 用来管理 Deployment 和扩展一组 Pod，并且能为这些 Pod 提供序号和唯一性保证。

StatefulSets 对于需要满足以下一个或多个需求的应用程序很有价值：
稳定的、唯一的网络标识符。
稳定的、持久的存储。
有序的、优雅的部署和缩放。
有序的、自动的滚动更新。

5.job控制器

执行批处理任务，仅执行一次任务，保证任务的一个或多个Pod成功结束。

5.1job控制器示例

私有仓库server1中准备好镜像
docker pull perl
docker tag perl:latest reg.westos.org/library/perl:latest
docker push reg.westos.org/library/perl:latest

[root@server2 ~]# cat job.yml 
apiVersion: batch/v1
kind: Job
metadata:
 name: pi
spec:
 template:
   spec:
     containers:
     - name: pi
       image: perl
       command: ["perl",  "-Mbignum=bpi", "-wle", "print bpi(2000)"]
     restartPolicy: Never  #不重启
 backoffLimit: 4   #如果发生错误，停止


[root@server2 ~]# kubectl apply -f job.yml 
job.batch/pi created
[root@server2 ~]# kubectl get pod
NAME       READY   STATUS              RESTARTS   AGE
demo       1/1     Running             6          23h
pi-59ld8   0/1     ContainerCreating   0          15s
[root@server2 ~]# kubectl  describe pod 
demo      pi-59ld8  
[root@server2 ~]# kubectl  describe pod pi-59ld8 
Name:         pi-59ld8
Namespace:    default
Priority:     0
Node:         server3/172.25.10.3
Events:
  Type    Reason     Age   From               Message
  ----    ------     ----  ----               -------
  Normal  Scheduled  34s   default-scheduler  Successfully assigned default/pi-59ld8 to server3
  Normal  Pulling    33s   kubelet            Pulling image "perl"
  Normal  Pulled     11s   kubelet            Successfully pulled image "perl" in 21.814470215s
  Normal  Created    11s   kubelet            Created container pi
  Normal  Started    11s   kubelet            Started container pi
[root@server2 ~]# kubectl get pod
NAME       READY   STATUS      RESTARTS   AGE
demo       1/1     Running     6          23h
pi-59ld8   0/1     Completed   0          47s

[root@server2 ~]# kubectl  logs pi-59ld8 
3.14159265358979323846264338327950288419716939937510582097494459230781640628620899862803482534211706798214808651328230664709384460955058223172535940812848111745028410270193852110555964462294895493038196442881097566593344612847564823378678316527120190914564856692346034861045432664821339360726024914127372458700660631558817488152092096282925409171536436789259036001133053054882046652138414695194151160943305727036575959195309218611738193261179310511854807446237996274956735188575272489122793818301194912983367336244065664308602139494639522473719070217986094370277053921717629317675238467481846766940513200056812714526356082778577134275778960917363717872146844090122495343014654958537105079227968925892354201995611212902196086403441815981

6.CronJob控制器

Cron Job 创建基于时间调度的 Jobs。

一个 CronJob 对象就像 crontab (cron table) 文件中的一行，它用 Cron 格式进行编写，并周期性地在给定的调度时间执行 Job。
https://kubernetes.io/zh/docs/concepts/workloads/controllers/cron-jobs/

Cron 时间表语法

# ┌───────────── 分钟 (0 - 59)
# │ ┌───────────── 小时 (0 - 23)
# │ │ ┌───────────── 月的某天 (1 - 31)
# │ │ │ ┌───────────── 月份 (1 - 12)
# │ │ │ │ ┌───────────── 周的某天 (0 - 6) （周日到周一；在某些系统上，7 也是星期日）
# │ │ │ │ │                                   
# │ │ │ │ │
# │ │ │ │ │
# * * * * *

6.1 CronJob控制器示例

每隔一分钟执行

[root@server2 ~]# cat cronjob.yml 
apiVersion: batch/v1beta1
kind: CronJob
metadata:
  name: hello
spec:
  schedule: "*/1 * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hello
            image: busyboxplus
            imagePullPolicy: IfNotPresent
            args:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure

[root@server2 ~]# kubectl get pod
NAME                     READY   STATUS      RESTARTS   AGE
demo                     1/1     Running     6          24h
hello-1614713340-rsrj5   0/1     Completed   0          4s
pi-59ld8                 0/1     Completed   0          6m11s
[root@server2 ~]# kubectl get  all
NAME                     READY   STATUS      RESTARTS   AGE
demo                     1/1     Running     6          24h
hello-1614713340-rsrj5   0/1     Completed   0          4s
pi-59ld8                 0/1     Completed   0          6m11s
[root@server2 ~]# kubectl get  all
NAME                         READY   STATUS      RESTARTS   AGE
pod/demo                     1/1     Running     6          24h
pod/hello-1614713340-rsrj5   0/1     Completed   0          14s
pod/pi-59ld8                 0/1     Completed   0          6m21s

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP   9d

NAME                         COMPLETIONS   DURATION   AGE
job.batch/hello-1614713340   1/1           1s         14s
job.batch/pi                 1/1           28s        6m21s

NAME                  SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
cronjob.batch/hello   */1 * * * *   False     0        19s             28s

每隔一分钟调用一个

[root@server2 ~]# kubectl get pod
NAME                     READY   STATUS      RESTARTS   AGE
demo                     1/1     Running     6          24h
hello-1614713340-rsrj5   0/1     Completed   0          97s
hello-1614713400-5s9fg   0/1     Completed   0          37s
pi-59ld8                 0/1     Completed   0          7m44s
[root@server2 ~]# kubectl  logs hello-1614713340-rsrj5
Tue Mar  2 19:29:06 UTC 2021
[root@server2 ~]# kubectl  logs hello-1614713400-5s9fg 
Tue Mar  2 19:30:06 UTC 2021

7. HPA

根据资源利用率自动调整service中Pod数量，实现Pod水平自动缩放。

posted on 2022-06-29 17:48 DoubleLi 阅读(320) 评论(0) 编辑收藏举报

刷新页面返回顶部

登录后才能查看或发表评论，立即登录或者逛逛博客园首页

相关博文：

· Kubernetes 之Pod学习

· DaemonSet 与 StatefulSet 的使用

· k8s 中 Pod 的控制器

· Kubernetes 篇章（六）

· K8s-控制器

阅读排行：
· 阿里最新开源QwQ-32B，效果媲美deepseek-r1满血版，部署成本又又又降低了！
· 单线程的Redis速度为什么快？
· SQL Server 2025 AI相关能力初探
· AI编程工具终极对决：字节Trae VS Cursor，谁才是开发者新宠？
· 展开说说关于C#中ORM框架的用法！

历史上的今天：
2017-06-29 在Windows下搭建基于nginx的视频直播和点播系统
2017-06-29 Linux下视频流媒体直播服务器搭建详解
2017-06-29 四款好用的免费直播编码推流软件
2017-06-29 如何生成HLS协议的M3U8文件
2017-06-29 Web视频播放之【HTML5 Video标签】
2017-06-29 谷歌浏览器Chrome播放rtsp视频流解决方案
2016-06-29 linux 调试利器gdb, strace, pstack, pstree, lsof

DoubleLi

公告

搜索

常用链接

我的标签

随笔分类 (6181)

随笔档案 (4734)

文章分类 (2)

参考博客

阅读排行榜

评论排行榜

推荐排行榜

最新评论

控制器