一、pod概念

1. pod类型

① 自主式pod：由yaml文件来创建的pod，也是pod自己去控制自己，防止pod备控制器杀死（一般很少用到）

② 控制器管理的pod：由Kubernetes控制器管理的pod，controllers在集群上管理和运行pod对象，通过label-selector相关联，pod通过控制器实现应用运维，如伸缩，升级等

kubernetes 监控pod请求数 kubernetes中controller定义的pod特性有_docker

2. pod控制类型

① ReplicationController用来确保容器应用的副本数始终保持在用户定义的副本数，即如果有容器异常退出，会自动创建新的pod来替代；而如果异常多出来的容器也会自动回收。建议是用ReplicaSet替代ReplicationController

② ReplicaSet跟ReplicationController没有本质的不同，只是名字不一样，并且ReplicaSet支持集合式的selector，虽然ReplicaSet可以独立使用，但一般还是建议使用Deployment来自动管理ReplicaSet，这样无需担心跟其他机制的不兼容问题（比如ReplicaSet不支持rolling-update但Deployment支持）

③ Deployment为pod和ReplicaSet提供了一个声明式定义方法，用来替代以前的ReplicationController来方便的管理应用。典型的应用场景包括：

  • 定义Deployment来创建pod和ReplicaSet
  • 滚动升级和回滚应用
  • 扩容和伸缩
  • 暂停和继续Deployment
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
  namespace: default
spec:
  replicas: 2
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: daocloud.io/library/nginx:1.13.0-alpine
        ports:
        - containerPort: 80
root@master:/home/pod-yaml# kubectl get pods,deploy,rs -o wide
NAME                                   READY   STATUS    RESTARTS   AGE    IP           NODE    NOMINATED NODE   READINESS GATES
pod/nginx-deployment-d68cf6b55-dgxk9   1/1     Running   0          143m   10.244.2.9   node2   <none>           <none>
pod/nginx-deployment-d68cf6b55-pddmb   1/1     Running   0          143m   10.244.2.8   node2   <none>           <none>

NAME                               READY   UP-TO-DATE   AVAILABLE   AGE    CONTAINERS   IMAGES                                    SELECTOR
deployment.apps/nginx-deployment   2/2     2            2           143m   nginx        daocloud.io/library/nginx:1.13.0-alpine   app=nginx

NAME                                         DESIRED   CURRENT   READY   AGE    CONTAINERS   IMAGES                                    SELECTOR
replicaset.apps/nginx-deployment-d68cf6b55   2         2         2       143m   nginx        daocloud.io/library/nginx:1.13.0-alpine   app=nginx,pod-template-hash=d68cf6b55

③ Horizontal Pod Autoscaling仅适用于Deployment和ReplicaSet，在v1版本中仅支持根据pod的cpu利用率扩所容

④ StatefulSet是为了解决有状态服务的问题（对应Deployments和ReplicaSets是为无状态服务而设计），其应用场景包括：

  • 稳定的持久化存储，即pod重新调度后还是能访问到相同的持久化数据，基于PVC来实现
  • 稳定的网络标志，即pod重新调度后其PodName和HostName不变，基于Headless Service（即没有Cluster IP的Service）来实现
  • 有序部署，有序扩展，即pod是有顺序的，在部署或者扩展的时候要依据定义的顺序依次进行（即从0到N-1，在下一个pod运行之前所有之前的pod必须都是Running和Ready状态），基于init containers实现
  • 有序收缩，有序删除（即从N-1到0）

StatefulSet身份（唯一网络标识符）三要素：

  • 域名
  • 主机名
  • 存储（PVC）
ClusterIP A记录格式:<service-name>.<namespace-name>.svc.cluster.local 
ClusterIP=None A记录格式:<statefulsetName-index>.<service-name>.<namespace-name>.svc.cluster.local
示例:nginx-statefulset-0.nginx.default.svc.cluster.local
apiVersion: v1
kind: Service
metadata:
  name: nginx
  labels:
    app: nginx
spec:
  ports:
  - port: 80
    name: web
  clusterIP: None
  selector:
    app: nginx


---

apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: nginx-statefulset
  namespace: default
spec:
  serviceName: nginx
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: daocloud.io/library/nginx:1.13.0-alpine
        ports:
        - containerPort: 80

pod NAME是不变的，这样可以保证pod有一个唯一固定的域名：

root@master:/home/pod-yaml# kubectl create -f statefulset.yaml 
service/nginx created
statefulset.apps/nginx-statefulset created
root@master:/home/pod-yaml# kubectl get pods
NAME                  READY   STATUS    RESTARTS   AGE
nginx-statefulset-0   1/1     Running   0          7s
nginx-statefulset-1   1/1     Running   0          5s
nginx-statefulset-2   1/1     Running   0          3s
root@master:/home/pod-yaml# kubectl delete pod nginx-statefulset-0
pod "nginx-statefulset-0" deleted
root@master:/home/pod-yaml# kubectl get pods
NAME                  READY   STATUS    RESTARTS   AGE
nginx-statefulset-0   1/1     Running   0          4s
nginx-statefulset-1   1/1     Running   0          70s
nginx-statefulset-2   1/1     Running   0          68s

⑤ DaemonSet确保全部（或者一些）Node上运行一个pod的副本。当有Node加入集群时，也会为他们新增一个pod。当有Node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。使用DaemonSet的一些典型用法：

  • 运行集群存储daemon，例如在每个Node上运行glusterd、ceph
  • 在每个Node上运行日志收集daemon，例如fluentd、logstash
  • 在每个Node上运行监控daemon，例如Prometheus Node Exporter
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: filebeat-daemonset
  labels:
    app: filebeat
spec:
  selector:
    matchLabels:
      app: filebeat
  template:
    metadata:
      labels:
        app: filebeat
    spec:
      containers:
      - name: logs
        image: daocloud.io/library/nginx:1.13.0-alpine
        ports:
        - containerPort: 80
        volumeMounts:
        - name: varlog
          mountPath: /tmp/log
      volumes:
      - name: varlog
        hostPath:
          path: /var/log
root@master:/home/pod-yaml# kubectl get pods -o wide
NAME                       READY   STATUS    RESTARTS   AGE    IP            NODE    NOMINATED NODE   READINESS GATES
filebeat-daemonset-bfcdd   1/1     Running   0          3m3s   10.244.1.26   node1   <none>           <none>
filebeat-daemonset-px94w   1/1     Running   0          3m3s   10.244.2.15   node2   <none>           <none>
root@master:/home/pod-yaml# kubectl exec -it filebeat-daemonset-bfcdd -- /bin/sh
/ # cd /tmp/log/
/tmp/log # ls
alternatives.log          cloud-init-output.log     faillog                   lastlog                   unattended-upgrades       vmware-network.5.log      vmware-vmsvc-root.3.log
apt                       cloud-init.log            installer                 lxd                       vmware-network.1.log      vmware-network.6.log      vmware-vmsvc-root.log
auth.log                  containers                journal                   pods                      vmware-network.2.log      vmware-network.log        vmware-vmtoolsd-root.log
bootstrap.log             dist-upgrade              kern.log                  syslog                    vmware-network.3.log      vmware-vmsvc-root.1.log   wtmp
btmp                      dpkg.log                  landscape                 tallylog                  vmware-network.4.log      vmware-vmsvc-root.2.log

⑥ Job负责批处理任务，即仅执行一次的任务，它保证批处理任务的一个或多个pod成功结束，应用场景:离线数据处理，视频解码等业务

apiVersion: batch/v1
kind: Job
metadata:
  name: pi
spec:
  template:
    spec:
      containers:
      - name: pi
        image: perl
        command: ["perl",  "-Mbignum=bpi", "-wle", "print bpi(10)"]
      restartPolicy: Never
  backoffLimit: 4
root@master:/home/pod-yaml# kubectl get pods
NAME       READY   STATUS      RESTARTS   AGE
pi-h9dr9   0/1     Completed   0          12m
root@master:/home/pod-yaml# kubectl logs pi-h9dr9
3.141592654

⑦ Cron Job管理基于时间的Job，即

  • 在给定时间点只运行一次
  • 周期性地给定时间点运行
  • 应用场景:通知，备份
apiVersion: batch/v1beta1
kind: CronJob
metadata:
  name: hello
spec:
  schedule: "*/1 * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: hello
            image: busybox
            args:
            - /bin/sh
            - -c
            - date; echo Hello from the Kubernetes cluster
          restartPolicy: OnFailure
root@master:/home/pod-yaml# kubectl create -f cronjob.yaml 
cronjob.batch/hello created
root@master:/home/pod-yaml# kubectl get cronjob
NAME    SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
hello   */1 * * * *   False     0        17s             30s
root@master:/home/pod-yaml# kubectl get pods
NAME                     READY   STATUS      RESTARTS   AGE
hello-1589135400-hzs6q   0/1     Completed   0          28s
root@master:/home/pod-yaml# kubectl logs hello-1589135400-hzs6q
Sun May 10 18:30:10 UTC 2020
Hello from the Kubernetes cluster