• Kubernetes ---- 存储卷(emptyDir、hostPath、NFS)


    存储卷

      Pod是有生命周期的,当Pod出现故障时,数据会随着Pod的终结就结束了.
      针对K8s集群,我们应该使用脱离节点的存储设备,共享存储设备.
      如果使用docker持久化数据的那种方法,那么Pod重构后就不能更换节点,否则,挂载的目录位置肯定就访问不到了.

    可用存储卷
      1. emptyDir: 临时存储目录,随着Pod删除也会被删除.
      2. hostPath: 节点路径,直接在宿主机上找一个目录,与Pod进行关联.
    网络存储卷:
      脱离主机节点本地的存储设备:
      1. SAN: ISCSI,...
      2. NAS: NFS、CIFS
    分布式存储:
      1. GlusterFS
      2. cephfs: ceph文件系统存储
      3. rbd: ceph块存储
    云端存储:
      1. EBS(AWS): 弹性块存储
      2. Azure Disk(微软)

    # 查看k8s支持的存储类型
    $ kubectl explain pods.spec.volumes

    pods.spec.volumes.emptyDir <Object>
      emptyDir类型的volume在pod分配到node上时被创建,kubernetes会在node上自动分配一个目录,因此无需指定宿主机node上对应的目录文件。这个目录的初始内容为空,
    当Pod从node上移除时,emptyDir中的数据会被永久删除.

    # 以下为了演示同一Pod的容器可以共享存储空间,虽然挂载目录指定的不一样,但是创建数据最终会在各个定义的目录中出现,实则是一样的.

    $ vim pod-vol-demo.yaml
      apiVersion: v1
      kind: Pod
      metadata:
        name: pod-vol-demo
        namespace: default
      spec:
        containers:
        - name: myapp
          image: ikubernetes/myapp:v1
          imagePullPolicy: IfNotPresent
          ports:
          - name: http
            containerPort: 80
          volumeMounts:
          - name: disk1
            mountPath: /usr/share/nginx/html/
        - name: busybox
          image: busybox:latest
          imagePullPolicy: IfNotPresent
          volumeMounts:
          - name: disk1
            mountPath: /data/
          command: ["/bin/sh","-c","while true;do echo $(date) >> /data/index.html;done"]
        volumes:
        - name: disk1
          emptyDir: {}
    $ kubectl apply -f pod-vol-demo.yaml
    $ kubectl get pods -o wide
    NAME     READY STATUS RESTARTS   AGE   IP     NODE   NOMINATED NODE READINESS GATES
    pod-vol-demo 2/2 Running   0     43m 10.244.2.124 node2   <none>   <none>
    
    $ curl 10.244.2.124

    pods.spec.volumes.hostPath <Object> 宿主机路径
      把Pod所在的宿主机之上的宿主机的文件系统目录与Pod建立关联关系,在Pod被删除时,这个目录是不会被删除的,所以在Pod被删除之后重新被调度到这个节点上的话
    对应的数据依然是存在的,如果宿主机上的目录默认不存在的话是否自动创建取决于type的定义,但如果节点宕机了,那么数据还是会丢失的.

    path <string> -required-
    type <string>

             空字符串(默认)用于向后兼容,这意味着在安装 hostPath 卷之前不会执行任何检查。
    DirectoryOrCreate 如果在给定路径上什么都不存在,那么将根据需要创建空目录,权限设置为 0755,具有与 Kubelet 相同的组和所有权。
    Directory     在给定路径上必须存在的目录。
    FileOrCreate    如果在给定路径上什么都不存在,那么将在那里根据需要创建空文件,权限设置为 0644,具有与 Kubelet 相同的组和所有权。
    File        在给定路径上必须存在的文件。
    Socket       在给定路径上必须存在的 UNIX 套接字。
    CharDevice     在给定路径上必须存在的字符设备。
    BlockDevice    在给定路径上必须存在的块设备。

    $ vim pod-hostpath.yaml
      apiVersion: v1   kind: Pod   metadata:    name: pod
    -hostpath    namespace: default   spec:    containers:    - name: hostpath-container    image: ikubernetes/myapp:v1    imagePullPolicy: IfNotPresent    ports:    - name: http    containerPort: 80    volumeMounts:    - name: disk2    mountPath: /usr/share/nginx/html/    lifecycle:    postStart:    exec:    command: ["/bin/sh","-c","echo $(hostname) > /usr/share/nginx/html/index.html"]    volumes:    - name: disk2    hostPath:    path: /data/pod/volume/pod-hostpath    type: DirectoryOrCreate

    $ kubectl apply -f pod-hostpath.yaml
    $ kubectl get pods -o wide 
    NAME        READY   STATUS RESTARTS AGE     IP    NODE NOMINATED NODE READINESS GATES
    pod-hostpath   1/1   Running   0   4m34s 10.244.2.127 node2 <none>   <none>

    
    

    $ curl 10.244.2.127
    pod-hostpath

    k8s配置NFS共享存储:

    1. 在指定一台服务器安装NFS服务以提供共享存储服务

    nfs node:

      # yum -y install nfs-utils
      # mkdir /data/volumes -pv
      # vim /etc/exports
      /data/volumes 192.168.222.0/24(rw,no_root_squash)
      # systemctl start nfs
      # ss -nlt | grep 2049


    2. 手动测试k8s的node是否支持挂载NFS文件系统,如果不支持则进行解决.
    node2:

      # mount -t nfs 192.168.222.103:/data/volumes/ /mnt
    
      mount: wrong fs type, bad option, bad superblock on 192.168.222.103:/data/volumes,
      missing codepage or helper program, or other error
      (for several filesystems (e.g. nfs, cifs) you might
      need a /sbin/mount.<type> helper program)
    
      In some cases useful info is found in syslog - try
      dmesg | tail or so.
      # yum -y install nfs-utils
      # mount -t nfs 192.168.222.103:/data/volumes /mnt
      # mount | grep 192.168.222.103
      # umount /mnt

    3. 定义Pod进行测试

    $ vim pod-nfs-demo.yaml

      apiVersion: v1   kind: Pod   metadata:    name: nfs
    -pod    namespace: default   spec:    containers:    - name: nfs-container    image: ikubernetes/myapp:v1    imagePullPolicy: IfNotPresent    ports:    - name: http    containerPort: 80    volumeMounts:    - name: nfs-disk    mountPath: /usr/share/nginx/html/    lifecycle:    postStart:    exec:    command: ["/bin/sh","-c","echo $(hostname) > /usr/share/nginx/html/index.html"]    volumes:    - name: nfs-disk    nfs:    path: /data/volumes/    server: 192.168.222.103

    $ kubectl apply -f pod-nfs-demo.yaml


    $ kubectl get pods -o wide
    NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
    pod-nfs-vol 1/1 Running 0 4m41s 10.244.2.129 node2 <none> <none>

    
    

    $ curl 10.244.2.129
    pod-nfs-vol

    3.2 定义Deployment多个Pod同时挂载至NFS
    $ vim deploy-nfs-demo.yaml

    apiVersion: v1
    Kind: Service
    metadata:
      name: nfs-svc
      namespace: default
    spec:
      selector:
        app: nfs-pod
        release: dev
      type: NodePort
      ports:
      - name: http
        port: 80
        targetPort: 80
        nodePort: 30180
    ---
    apiVersion: apps/v1
    kind: Deployment
    metadata:
      name: nfs-deploy
      namespace: default
    spec:
      replicas: 3
      selector:
        matchLabels:
          app: nfs-pod
          release: dev
      template:
        meatadata:
          name: nfs-pod
          namespace: defualt
          labels:
            app: nfs-pod
            release: dev
        spec:
          containers:
         - name: nfs-container
           image: ikubernetes/myapp:v2
           imagePullPolicy: IfNotPresent
           ports:
           - name: http
             port: 80
             containerPort: 80
           volumeMounts:
           - name: nfs-disk
             mountPath: /usr/share/nginx/html/
           lifecycle:
             postStart:
               exec:
                 command: ["/bin/sh","-c","echo $(hostname) > /usr/share/nginx/html/index.html"]
           livenessProbe:
             exec:
               command: ["/bin/sh","-c","ps aux | grep nginx"]
             initialDelaySeconds: 2
             periodSeconds: 3
           readinessProbe:
             tcpSocket:
             port: 80
           initialDelaySeconds: 2
           periodSeconds: 3
         volumes:
         - name: nfs-disk
           nfs:
             path: /data/volumes/
             server: 192.168.222.103

    $ kubectl apply -f deploy-nfs-demo.yaml
    $ kubectl get svc -o wide
    NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR
    nfs-svc NodePort 10.101.149.48 <none> 80:30180/TCP 12m app=nfs-pod,release=dev

    
    

    客户端访问"http://192.168.222.101:30180"

    4. 登录至nfs node查看指定目录下是否存在文件

    # cat /data/volumes/index.html
    pod-nfs-vol
  • 相关阅读:
    python gevent(协程模块)
    python基础之socket与socketserver
    python 使用 with open() as 读写文件
    Python logger模块
    python二维码操作:QRCode和MyQR入门
    常见的端口号及其用途
    python中hasattr()、getattr()、setattr()函数的使用
    mysql数据库自带数据库介绍
    关于BeautifulSoup4 解析器的说明
    学习opencv(1)
  • 原文地址:https://www.cnblogs.com/k-free-bolg/p/13155970.html
Copyright © 2020-2023  润新知