k8s中的存储

目录

[一 configmap](#一 configmap)

[1.1 configmap的功能](#1.1 configmap的功能)

[1.2 configmap的使用场景](#1.2 configmap的使用场景)

[1.3 configmap创建方式](#1.3 configmap创建方式)

[1.3.1 字面值创建](#1.3.1 字面值创建)

[1.3.2 通过文件创建](#1.3.2 通过文件创建)

[1.3.3 通过目录创建](#1.3.3 通过目录创建)

[1.3.4 通过yaml文件创建](#1.3.4 通过yaml文件创建)

[1.3.5 configmap的使用方式](#1.3.5 configmap的使用方式)

[1.3.5.1 使用configmap填充环境变量](#1.3.5.1 使用configmap填充环境变量)

[1.3.5.2 通过数据卷使用configmap](#1.3.5.2 通过数据卷使用configmap)

[1.3.5.3 利用configMap填充pod的配置文件](#1.3.5.3 利用configMap填充pod的配置文件)

[1.3.5.4 通过热更新cm修改配置](#1.3.5.4 通过热更新cm修改配置)

[二 secrets配置管理](#二 secrets配置管理)

[2.1 secrets的功能介绍](#2.1 secrets的功能介绍)

[2.2 secrets的创建](#2.2 secrets的创建)

2.2.1从文件创建

[2.2.2 编写yaml文件](#2.2.2 编写yaml文件)

[2.3 Secret的使用方法](#2.3 Secret的使用方法)

[2.3.1 将Secret挂载到Volume中](#2.3.1 将Secret挂载到Volume中)

[2.3.2 向指定路径映射 secret 密钥](#2.3.2 向指定路径映射 secret 密钥)

[2.3.3 将Secret设置为环境变量](#2.3.3 将Secret设置为环境变量)

[2.3.4 存储docker registry的认证信息](#2.3.4 存储docker registry的认证信息)

[三 volumes配置管理](#三 volumes配置管理)

[3.1 kubernets支持的卷的类型](#3.1 kubernets支持的卷的类型)

[3.2 emptyDir卷](#3.2 emptyDir卷)

[3.3 hostpath卷](#3.3 hostpath卷)

[3.4 nfs卷](#3.4 nfs卷)

[3.4.1 部署一台nfs共享主机并在所有k8s节点中安装nfs-utils](#3.4.1 部署一台nfs共享主机并在所有k8s节点中安装nfs-utils)

[3.4.2 部署nfs卷](#3.4.2 部署nfs卷)

[3.5 PersistentVolume持久卷](#3.5 PersistentVolume持久卷)

[3.5.1 静态持久卷pv与静态持久卷声明pvc](#3.5.1 静态持久卷pv与静态持久卷声明pvc)

PersistentVolume(持久卷,简称PV)

PersistentVolumeClaim(持久卷声明,简称PVC)

volumes访问模式

volumes回收策略

volumes状态说明

静态pv实例:

在pod中使用pvc

[四 存储类storageclass](#四 存储类storageclass)

[4.1 StorageClass说明](#4.1 StorageClass说明)

[4.2 StorageClass的属性](#4.2 StorageClass的属性)

[4.3 存储分配器NFS Client Provisioner](#4.3 存储分配器NFS Client Provisioner)

[4.4 部署NFS Client Provisioner](#4.4 部署NFS Client Provisioner)

[4.4.1 创建sa并授权](#4.4.1 创建sa并授权)

[4.4.2 部署应用](#4.4.2 部署应用)

[4.4.3 创建存储类](#4.4.3 创建存储类)

[4.4.4 创建pvc](#4.4.4 创建pvc)

[4.4.5 创建测试pod](#4.4.5 创建测试pod)

[4.4.6 设置默认存储类](#4.4.6 设置默认存储类)

[五 statefulset控制器](#五 statefulset控制器)

[5.1 功能特性](#5.1 功能特性)

[5.2 StatefulSet的组成部分](#5.2 StatefulSet的组成部分)

[5.3 构建方法](#5.3 构建方法)

[5.4 测试:](#5.4 测试:)

[5.5 statefulset的弹缩](#5.5 statefulset的弹缩)


一 configmap

1.1 configmap的功能

  • configMap用于保存配置数据,以键值对形式存储。

  • configMap 资源提供了向 Pod 注入配置数据的方法。

  • 镜像和配置文件解耦,以便实现镜像的可移植性和可复用性。

  • etcd限制了文件大小不能超过1M

1.2 configmap的使用场景

  • 填充环境变量的值

  • 设置容器内的命令行参数

  • 填充卷的配置文件

1.3 configmap创建方式

1.3.1 字面值创建

#创建一个名为 "superhowe-config" 的 ConfigMap,其中包含两个键值对:"fname=super" 和 "lname=howe"

[root@k8s-master ~]# kubectl create cm superhowe-config --from-literal fname=super --from-literal lname=howe
configmap/superhowe-config created

[root@k8s-master ~]# kubectl describe cm superhowe-config
Name:         superhowe-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data		#键值信息显示
====
fname:
----
super
lname:
----
howe

BinaryData
====

Events:  <none>

1.3.2 通过文件创建

[root@k8s-master ~]# cat /etc/resolv.conf 
# Generated by NetworkManager
search exam.com
nameserver 172.25.250.2

[root@k8s-master ~]# kubectl create cm howe-config --from-file /etc/resolv.conf 
configmap/howe-config created

[root@k8s-master ~]# kubectl describe cm howe-config 
Name:         howe-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
resolv.conf:
----
# Generated by NetworkManager
search exam.com
nameserver 172.25.250.2


BinaryData
====

Events:  <none>

1.3.3 通过目录创建

[root@k8s-master ~]# mkdir config
[root@k8s-master ~]# cp /etc/fstab /etc/rc.d/rc.local config/
[root@k8s-master ~]# kubectl create cm howe2-config --from-file config/
configmap/howe2-config created
[root@k8s-master ~]# kubectl describe cm howe2-config 
Name:         howe2-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
rc.local:
----
#!/bin/bash
# THIS FILE IS ADDED FOR COMPATIBILITY PURPOSES
#
# It is highly advisable to create own systemd services or udev rules
# to run scripts during boot instead of using this file.
#
# In contrast to previous versions due to parallel execution during boot
# this script will NOT be run after all other services.
#
# Please note that you must run 'chmod +x /etc/rc.d/rc.local' to ensure
# that this script will be executed during boot.

touch /var/lock/subsys/local
mount /dev/sr0   /rhel9

fstab:
----

#
# /etc/fstab
# Created by anaconda on Wed Jul 31 03:08:58 2024
#
# Accessible filesystems, by reference, are maintained under '/dev/disk/'.
# See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info.
#
# After editing this file, run 'systemctl daemon-reload' to update systemd
# units generated from this file.
#
/dev/mapper/rhel-root   /                       xfs     defaults        0 0
UUID=15f86ce4-5a55-499e-8510-36a08ab8db17 /boot                   xfs     defaults        0 0
#/dev/mapper/rhel-swap   none                    swap    defaults        0 0


BinaryData
====

Events:  <none>

1.3.4 通过yaml文件创建

[root@k8s-master ~]# kubectl create cm howe3-config --from-literal db_host=172.25.250.100 --from-literal db_port=3306 --dry-run=client -o yaml > howe-config.yaml

[root@k8s-master ~]# vim howe-config.yaml 
apiVersion: v1
kind: ConfigMap
metadata:
  name: howe3-config
data:
  db_host: 172.25.250.100
  db_port: "3306"
  
[root@k8s-master ~]# kubectl apply -f howe-config.yaml 
configmap/howe3-config created
[root@k8s-master ~]# kubectl describe cm howe3-config 
Name:         howe3-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
db_host:
----
172.25.250.100
db_port:
----
3306

BinaryData
====

Events:  <none>

1.3.5 configmap的使用方式

  • 通过环境变量的方式直接传递给pod

  • 通过pod的 命令行运行方式

  • 作为volume的方式挂载到pod内

1.3.5.1 使用configmap填充环境变量
#讲cm中的内容映射为指定变量
[root@k8s-master ~]# vim testpod1.yml
apiVersion: v1
kind: Pod 
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - env
    env:
    - name: key1
      valueFrom:
        configMapKeyRef:
          name: howe-config
          key: db_host
    - name: key2
      valueFrom:
        configMapKeyRef:
          name: howe-config
          key: db_port
  restartPolicy: Never

[root@k8s-master ~]# kubectl apply -f testpod1.yml 
pod/testpod created

#查看日志
[root@k8s-master ~]# kubectl logs pods/testpod 
KUBERNETES_SERVICE_PORT=443
KUBERNETES_PORT=tcp://10.96.0.1:443
MYAPP_V1_SERVICE_HOST=10.102.186.55
HOSTNAME=testpod
SHLVL=1
MYAPP_V2_SERVICE_HOST=10.96.232.137
HOME=/
MYAPP_V1_SERVICE_PORT=80
MYAPP_V1_PORT=tcp://10.102.186.55:80
MYAPP_V2_PORT=tcp://10.96.232.137:80
MYAPP_V2_SERVICE_PORT=80
MYAPP_V1_PORT_80_TCP_ADDR=10.102.186.55
MYAPP_V2_PORT_80_TCP_ADDR=10.96.232.137
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
MYAPP_V1_PORT_80_TCP_PORT=80
MYAPP_V1_PORT_80_TCP_PROTO=tcp
MYAPP_V2_PORT_80_TCP_PORT=80
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
MYAPP_V2_PORT_80_TCP_PROTO=tcp
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
key1=172.25.250.100
key2=3306
MYAPP_V1_PORT_80_TCP=tcp://10.102.186.55:80
MYAPP_V2_PORT_80_TCP=tcp://10.96.232.137:80
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
PWD=/
KUBERNETES_SERVICE_HOST=10.96.0.1

#把cm中的值直接映射为变量
[root@k8s-master ~]# vim testpod2.yml 
apiVersion: v1
kind: Pod 
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - env
    envFrom:
    - configMapKeyRef:
          name: howe3-config
  restartPolicy: Never
  
#查看日志
[root@k8s-master ~]# kubectl logs pods/testpod 
KUBERNETES_SERVICE_PORT=443
KUBERNETES_PORT=tcp://10.96.0.1:443
MYAPP_V1_SERVICE_HOST=10.102.186.55
HOSTNAME=testpod
SHLVL=1
MYAPP_V2_SERVICE_HOST=10.96.232.137
HOME=/
MYAPP_V1_SERVICE_PORT=80
MYAPP_V1_PORT=tcp://10.102.186.55:80
MYAPP_V2_PORT=tcp://10.96.232.137:80
MYAPP_V2_SERVICE_PORT=80
MYAPP_V1_PORT_80_TCP_ADDR=10.102.186.55
MYAPP_V2_PORT_80_TCP_ADDR=10.96.232.137
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
MYAPP_V1_PORT_80_TCP_PORT=80
MYAPP_V1_PORT_80_TCP_PROTO=tcp
MYAPP_V2_PORT_80_TCP_PORT=80
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
MYAPP_V2_PORT_80_TCP_PROTO=tcp
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
key1=172.25.250.100
key2=3306
MYAPP_V1_PORT_80_TCP=tcp://10.102.186.55:80
MYAPP_V2_PORT_80_TCP=tcp://10.96.232.137:80
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
PWD=/
KUBERNETES_SERVICE_HOST=10.96.0.1

#在pod命令行中使用变量
[root@k8s-master ~]# vim testpod3.yml
apiVersion: v1
kind: Pod 
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - echo ${db_host} ${db_port}
    envFrom:
    - configMapKeyRef:
          name: howe3-config
  restartPolicy: Never
  
#查看日志
[root@k8s-master ~]# kubectl logs pods/testpod
1.3.5.2 通过数据卷使用configmap
[root@k8s-master ~]# vim testpod4.yml
apiVersion: v1
kind: Pod 
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - cat /config/db_host
    volumeMounts:				#调用卷策略
    - name: config-volume		 #卷名称
      mountPath: /config
    volumes:					#声明卷的配置
    - name: config-volume		 #卷名称
      configMap:
        name: howe3-config
    restartPolicy: Never
    
#查看日志
[root@k8s-master ~]# kubectl logs testpod
key1=172.25.250.100
key2=3306
1.3.5.3 利用configMap填充pod的配置文件
#建立配置文件模板
[root@k8s-master ~]# vim nginx.conf
server {
  listen 8000;
  server_name _;
  root /usr/share/nginx/html;
  index index.html;
}

#利用模板生成cm
[root@k8s-master ~]# kubectl create cm nginx-conf --from-file nginx.conf 
configmap/nginx.conf created

[root@k8s-master ~]# kubectl get cm nginx-conf 
NAME         DATA   AGE
nginx-conf   1      32s

[root@k8s-master ~]# kubectl describe cm nginx-conf 
Name:         nginx-conf
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
nginx.conf:
----
server {
  listen 8000;
  server_name _;
  root /usr/share/nginx/html;
  index index.html;
}


BinaryData
====

Events:  <none>


#建立nginx控制器文件
[root@k8s-master ~]# kubectl create deployment nginx --image nginx:latest --replicas 1 --dry-run=client -o yaml > nginx.yml

#设定nginx.yml中的卷
[root@k8s-master ~]# vim nginx.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: nginx
  name: nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - image: nginx:latest
        name: nginx
        volumeMounts:
        - name: config-volume
          mountPath: /etc/nginx/conf.d

      volumes:
        - name: config-volume
          configMap:
            name: nginx-conf
            

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

#测试
[root@k8s-master ~]# kubectl get deployments.apps 
NAME       READY   UP-TO-DATE   AVAILABLE   AGE
nginx      1/1     1            1           16m

[root@k8s-master ~]# kubectl get pods -o wide 
NAME                        READY   STATUS      RESTARTS       AGE     IP           NODE                 NOMINATED NODE   READINESS GATES
nginx-8487c65cfc-v8q2g      1/1     Running     0              17m     10.244.1.9   k8s-node1.exam.com   <none>           <none>

[root@k8s-master ~]# curl 10.244.1.9:8000
1.3.5.4 通过热更新cm修改配置
#如需更改端口
[root@k8s-master ~]# kubectl edit cm nginx-conf 
9       listen 8080;

[root@k8s-master ~]# kubectl exec pods/nginx-8487c65cfc-v8q2g -- cat /etc/nginx/conf.d/nginx.conf
server {
  listen 8080;
  server_name _;
  root /usr/share/nginx/html;
  index index.html;
}

[root@k8s-master ~]# kubectl delete pods nginx-8487c65cfc-v8q2g 
pod "nginx-8487c65cfc-v8q2g" deleted

[root@k8s-master ~]# kubectl get pods -o wide 
NAME                        READY   STATUS      RESTARTS       AGE     IP           NODE                 NOMINATED NODE   READINESS GATES
nginx-8487c65cfc-7s4lh      1/1     Running     0              92s     10.244.2.6   k8s-node2.exam.com   <none>           <none>

[root@k8s-master ~]# curl 10.244.2.6:8080

二 secrets配置管理

2.1 secrets的功能介绍

  • Secret 对象类型用来保存敏感信息,例如密码、OAuth 令牌和 ssh key。

  • 敏感信息放在 secret 中比放在 Pod 的定义或者容器镜像中来说更加安全和灵活

  • Pod 可以用两种方式使用 secret:

    • 作为 volume 中的文件被挂载到 pod 中的一个或者多个容器里。

    • 当 kubelet 为 pod 拉取镜像时使用。

  • Secret的类型:

    • Service Account:Kubernetes 自动创建包含访问 API 凭据的 secret,并自动修改 pod 以使用此类型的 secret。

    • Opaque:使用base64编码存储信息,可以通过base64 --decode解码获得原始数据,因此安全性弱。

    • kubernetes.io/dockerconfigjson:用于存储docker registry的认证信息

2.2 secrets的创建

在创建secrets时我们可以用命令的方法或者yaml文件的方法

2.2.1从文件创建

[root@k8s-master ~]# echo -n howe > username.txt
[root@k8s-master ~]# echo -n www > password.txt
[root@k8s-master ~]# kubectl create secret generic userlist --from-file username.txt --from-file password.txt 
secret/userlist created
[root@k8s-master ~]# kubectl get secrets userlist -o yaml 
apiVersion: v1
data:
  password.txt: d3d3
  username.txt: aG93ZQ==
kind: Secret
metadata:
  creationTimestamp: "2024-09-08T13:02:06Z"
  name: userlist
  namespace: default
  resourceVersion: "222543"
  uid: 30dfc207-da48-47d4-98db-fe4aa1dca592
type: Opaque

2.2.2 编写yaml文件

#howe的Base64编码表示
[root@k8s-master ~]# echo -n howe | base64 	#用户名
aG93ZQ==
[root@k8s-master ~]# echo -n www | base64 	#密码
d3d3

[root@k8s-master ~]# kubectl create secret generic userlist --dry-run=client -o yaml > userlist.yml

[root@k8s-master ~]# vim userlist.yml
apiVersion: v1
kind: Secret
metadata:
  name: userlist
type: Opaque
data:
  username: aG93ZQ==
  password: d3d3
  
[root@k8s-master ~]# kubectl delete secrets userlist 
secret "userlist" deleted

[root@k8s-master ~]# kubectl get secrets userlist -o yaml
apiVersion: v1
data:
  password: d3d3
  username: aG93ZQ==

2.3 Secret的使用方法

2.3.1 将Secret挂载到Volume中

[root@k8s-master ~]# kubectl run nginx --image nginx --dry-run=client -o yaml > pod1.yaml

#向固定路径映射
[root@k8s-master ~]# vim pod1.yaml 
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: nginx
  name: nginx
spec:
  containers:
  - image: nginx
    name: nginx
    volumeMounts:
    - name: secrets
      mountPath: /secret
      readOnly: true

  volumes:
  - name: secrets
    secret:
      secretName: userlist
      
[root@k8s-master ~]# kubectl apply -f pod1.yaml 
pod/nginx created

[root@k8s-master ~]# kubectl exec pods/nginx -it -- /bin/bash
root@nginx:/# cd /secret/
root@nginx:/secret# ls
password  username
root@nginx:/secret# cat password 
wwwroot@nginx:/secret# cat username 
howeroot@nginx:/secret# 

2.3.2 向指定路径映射 secret 密钥

#向指定路径映射
[root@k8s-master ~]# vim pod2.yaml
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: nginx1
  name: nginx1
spec:
  containers:
  - image: nginx
    name: nginx1
    volumeMounts:
    - name: secrets
      mountPath: /secret
      readOnly: true

  volumes:
  - name: secrets
    secret:
      secretName: userlist
      items:
      - key: username
        path: my-users/username
        
[root@k8s-master ~]# kubectl apply -f pod2.yaml 
pod/nginx1 created

[root@k8s-master ~]# kubectl exec pods/nginx1 -it -- /bin/bash
root@nginx1:/# cd secret/
root@nginx1:/secret# ls
my-users
root@nginx1:/secret# cd my-users
root@nginx1:/secret/my-users# ls
username
root@nginx1:/secret/my-users# cat username 
howeroot@nginx1:/secret/my-users# 

2.3.3 将Secret设置为环境变量

[root@k8s-master ~]# vim pod3.yaml
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: busybox
  name: busybox
spec:
  containers:
  - image: busybox
    name: busybox
    command:
    - /bin/sh
    - -c
    - env
    env:
    - name: USERNAME
      valueFrom:
        secretKeyRef:
          name: userlist
          key: username
    - name: PASS
      valueFrom:
        secretKeyRef:
          name: userlist
          key: password
  restartPolicy: Never
  
[root@k8s-master ~]# kubectl apply -f pod3.yaml 
pod/busybox created
[root@k8s-master ~]# kubectl logs pods/busybox 
KUBERNETES_PORT=tcp://10.96.0.1:443
KUBERNETES_SERVICE_PORT=443
HOSTNAME=busybox
MYAPP_V1_SERVICE_HOST=10.102.186.55
MYAPP_V2_SERVICE_HOST=10.96.232.137
SHLVL=1
HOME=/root
MYAPP_V1_SERVICE_PORT=80
MYAPP_V1_PORT=tcp://10.102.186.55:80
MYAPP_V2_SERVICE_PORT=80
MYAPP_V2_PORT=tcp://10.96.232.137:80
MYAPP_V1_PORT_80_TCP_ADDR=10.102.186.55
USERNAME=howe
MYAPP_V2_PORT_80_TCP_ADDR=10.96.232.137
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
MYAPP_V1_PORT_80_TCP_PORT=80
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
MYAPP_V2_PORT_80_TCP_PORT=80
MYAPP_V1_PORT_80_TCP_PROTO=tcp
MYAPP_V2_PORT_80_TCP_PROTO=tcp
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
MYAPP_V1_PORT_80_TCP=tcp://10.102.186.55:80
MYAPP_V2_PORT_80_TCP=tcp://10.96.232.137:80
PASS=www
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_SERVICE_HOST=10.96.0.1
PWD=/

2.3.4 存储docker registry的认证信息

建立私有仓库并上传镜像

#建立私有仓库并上传镜像
[root@k8s-master ~]# docker login reg.exam.com

#上传镜像
[root@k8s-master ~]# docker load -i game2048.tar 
[root@k8s-master ~]# docker tag timinglee/game2048:latest reg.exam.com/howe/game2048:latest
[root@k8s-master ~]# docker push reg.exam.com/howe/game2048
Using default tag: latest
The push refers to repository [reg.exam.com/howe/game2048]
88fca8ae768a: Pushed 
6d7504772167: Pushed 
192e9fad2abc: Pushed 
36e9226e74f8: Pushed 
011b303988d2: Pushed 
latest: digest: sha256:8a34fb9cb168c420604b6e5d32ca6d412cb0d533a826b313b190535c03fe9390 size: 1364


#建立用于docker认证的secret
[root@k8s-master ~]# kubectl create secret docker-registry docker-auth --docker-server reg.exam.com --docker-username admin --docker-password redhat --docker-email howe@howe.com

#编写yml文件
[root@k8s-master ~]# vim pod4.yml
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: game2048
  name: game2048
spec:
  containers:
  - image: reg.exam.com/howe/game2048:latest
    name: game2048
  imagePullSecrets:					#不设定docker认证时无法下载镜像
  - name: docker-auth
  
[root@k8s-master ~]# kubectl apply -f pod4.yml 
pod/game2048 created

[root@k8s-master ~]# kubectl get pods 
NAME                        READY   STATUS      RESTARTS        AGE
game2048                    1/1     Running     0               11s

三 volumes配置管理

  • 容器中文件在磁盘上是临时存放的,这给容器中运行的特殊应用程序带来一些问题

  • 当容器崩溃时,kubelet将重新启动容器,容器中的文件将会丢失,因为容器会以干净的状态重建。

  • 当在一个 Pod 中同时运行多个容器时,常常需要在这些容器之间共享文件。

  • Kubernetes 卷具有明确的生命周期与使用它的 Pod 相同

  • 卷比 Pod 中运行的任何容器的存活期都长,在容器重新启动时数据也会得到保留

  • 当一个 Pod 不再存在时,卷也将不再存在。

  • Kubernetes 可以支持许多类型的卷,Pod 也能同时使用任意数量的卷。

  • 卷不能挂载到其他卷,也不能与其他卷有硬链接。 Pod 中的每个容器必须独立地指定每个卷的挂载位置。

3.1 kubernets支持的卷的类型

官网:卷 | Kubernetes

k8s支持的卷的类型如下:

  • awsElasticBlockStore 、azureDisk、azureFile、cephfs、cinder、configMap、csi

  • downwardAPI、emptyDir、fc (fibre channel)、flexVolume、flocker

  • gcePersistentDisk、gitRepo (deprecated)、glusterfs、hostPath、iscsi、local、

  • nfs、persistentVolumeClaim、projected、portworxVolume、quobyte、rbd

  • scaleIO、secret、storageos、vsphereVolume

3.2 emptyDir卷

功能:

当Pod指定到某个节点上时,首先创建的是一个emptyDir卷,并且只要 Pod 在该节点上运行,卷就一直存在。卷最初是空的。 尽管 Pod 中的容器挂载 emptyDir 卷的路径可能相同也可能不同,但是这些容器都可以读写 emptyDir 卷中相同的文件。 当 Pod 因为某些原因被从节点上删除时,emptyDir 卷中的数据也会永久删除

emptyDir 的使用场景:

  • 缓存空间,例如基于磁盘的归并排序。

  • 耗时较长的计算任务提供检查点,以便任务能方便地从崩溃前状态恢复执行。

  • 在 Web 服务器容器服务数据时,保存内容管理器容器获取的文件。

示例:

[root@k8s-master volumes]# vim pod1.yml
apiVersion: v1
kind: Pod
metadata:
  name: vol1
spec:
  containers:
  - image: busyboxplus:latest
    name: vm1
    command:
    - /bin/sh
    - -c
    - sleep 30000000
    volumeMounts:
    - mountPath: /cache
      name: cache-vol
  - image: nginx:latest
    name: vm2
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: cache-vol
  volumes:
  - name: cache-vol
    emptyDir:
      medium: Memory
      sizeLimit: 100Mi
      
[root@k8s-master volumes]# kubectl apply -f pod1.yml 
pod/vol1 created

#查看pod中卷的使用情况
[root@k8s-master volumes]# kubectl describe pods vol1 

#测试效果
[root@k8s-master volumes]# kubectl exec -it pods/vol1 -c vm1 -- /bin/sh
/ # cd cache/
/cache # ls
/cache # echo haha > index.html
/cache # curl localhost
haha
/cache # 
/cache # dd if=/dev/zero of=bigfile bs=1M count=101
dd: writing 'bigfile': No space left on device
101+0 records in
99+1 records out

3.3 hostpath卷

功能:

hostPath 卷能将主机节点文件系统上的文件或目录挂载到您的 Pod 中,不会因为pod关闭而被删除

hostPath 的一些用法

  • 运行一个需要访问 Docker 引擎内部机制的容器,挂载 /var/lib/docker 路径。

  • 在容器中运行 cAdvisor(监控) 时,以 hostPath 方式挂载 /sys。

  • 允许 Pod 指定给定的 hostPath 在运行 Pod 之前是否应该存在,是否应该创建以及应该以什么方式存在

hostPath的安全隐患

  • 具有相同配置(例如从 podTemplate 创建)的多个 Pod 会由于节点上文件的不同而在不同节点上有不同的行为。

  • 当 Kubernetes 按照计划添加资源感知的调度时,这类调度机制将无法考虑由 hostPath 使用的资源。

  • 基础主机上创建的文件或目录只能由 root 用户写入。您需要在 特权容器 中以 root 身份运行进程,或者修改主机上的文件权限以便容器能够写入 hostPath 卷。

示例:

[root@k8s-master volumes]# vim pod2.yml
apiVersion: v1
kind: Pod
metadata:
  name: vol1
spec:
  containers:
  - image: nginx:latest
    name: vm1
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: cache-vol
  volumes:
  - name: cache-vol
    hostPath:
      path: /data				
      type: DirectoryOrCreate		#当/data目录不存在时自动建立
      
#测试:
[root@k8s-master volumes]# kubectl apply -f pod2.yml 
pod/vol1 created

[root@k8s-master volumes]# kubectl get pods -o wide 
NAME                        READY   STATUS      RESTARTS       AGE     IP            NODE                 NOMINATED NODE   READINESS GATES
vol1                        1/1     Running     0              24s     10.244.2.8    k8s-node2.exam.com   <none>           <none>

[root@k8s-master volumes]# curl 10.244.2.8
<html>
<head><title>403 Forbidden</title></head>
<body>
<center><h1>403 Forbidden</h1></center>
<hr><center>nginx/1.27.1</center>
</body>
</html>

#通过kubectl get pods -o wide查看挂载到了哪个主机 比如这里为node2
[root@k8s-node2 ~]# echo xixi > /data/index.html

#回到master
[root@k8s-master volumes]# curl 10.244.2.8
xixi

#当pod被删除后hostPath不会被清理
[root@k8s-master volumes]# kubectl delete -f pod2.yml 
pod "vol1" deleted
[root@k8s-node2 ~]# ls /data/
index.html

3.4 nfs卷

NFS 卷允许将一个现有的 NFS 服务器上的目录挂载到 Kubernetes 中的 Pod 中。这对于在多个 Pod 之间共享数据或持久化存储数据非常有用

例如,如果有多个容器需要访问相同的数据集,或者需要将容器中的数据持久保存到外部存储,NFS 卷可以提供一种方便的解决方案。

3.4.1 部署一台nfs共享主机并在所有k8s节点中安装nfs-utils

#部署nfs主机
[root@harbor ~]# mkdir /nfsdata
[root@harbor ~]# dnf install nfs-utils -y
[root@harbor ~]# systemctl enable --now nfs-server.service 

[root@harbor ~]# vim /etc/exports
/nfsdata   *(rw,sync,no_root_squash)	#no_root_squash因为管理用户为root
[root@harbor ~]# ls -ld /nfsdata/
drwxr-xr-x 2 root root 6 Sep  9 05:58 /nfsdata/
[root@harbor ~]# exportfs -rv			#重启
exporting *:/nfsdata

[root@k8s-master ~]# showmount -e 172.25.250.250	#每台主机都要能看到
Export list for 172.25.250.250:
/nfsdata *

#在k8s所有节点中安装nfs-utils
[root@k8s-master ~]# dnf install nfs-utils -y
[root@k8s-master ~]# systemctl enable --now nfs-server.service
[root@k8s-node1 ~]# dnf install nfs-utils -y
[root@k8s-node1 ~]# systemctl enable --now nfs-server.service
[root@k8s-node2 ~]# dnf install nfs-utils -y
[root@k8s-node2 ~]# systemctl enable --now nfs-server.service

3.4.2 部署nfs卷

[root@k8s-master volumes]# vim pod3.yml
apiVersion: v1
kind: Pod 
metadata:
  name: vol1
spec:
  containers:
  - image: nginx:latest
    name: vm1 
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: cache-vol
  volumes:
  - name: cache-vol
    nfs:
      server: 172.25.250.250		#nfs服务器IP
      path: /nfsdata			    #nfs服务器共享目录

[root@k8s-master volumes]# kubectl apply -f pod3.yml 
pod/vol1 created

#测试
[root@k8s-master volumes]# kubectl get pods -o wide 
NAME                        READY   STATUS      RESTARTS        AGE     IP            NODE                 NOMINATED NODE   READINESS GATES
vol1                        1/1     Running     0               50s     10.244.2.9    k8s-node2.exam.com   <none>           <none>

[root@k8s-master volumes]# curl 10.244.2.9
<html>
<head><title>403 Forbidden</title></head>
<body>
<center><h1>403 Forbidden</h1></center>
<hr><center>nginx/1.27.1</center>
</body>
</html>

#在nfs主机中
[root@harbor ~]# echo welcome back! > /nfsdata/index.html
[root@k8s-master volumes]# curl 10.244.2.9
welcome back!

3.5 PersistentVolume持久卷

3.5.1 静态持久卷pv与静态持久卷声明pvc

PersistentVolume(持久卷,简称PV)
  • pv是集群内由管理员提供的网络存储的一部分。

  • PV也是集群中的一种资源。是一种volume插件,

  • 但是它的生命周期却是和使用它的Pod相互独立的。

  • PV这个API对象,捕获了诸如NFS、ISCSI、或其他云存储系统的实现细节

  • pv有两种提供方式:静态和动态

    • 静态PV:集群管理员创建多个PV,它们携带着真实存储的详细信息,它们存在于Kubernetes API中,并可用于存储使用

    • 动态PV:当管理员创建的静态PV都不匹配用户的PVC时,集群可能会尝试专门地供给volume给PVC。这种供给基于StorageClass

PersistentVolumeClaim(持久卷声明,简称PVC)
  • 是用户的一种存储请求

  • 它和Pod类似,Pod消耗Node资源,而PVC消耗PV资源

  • Pod能够请求特定的资源(如CPU和内存)。PVC能够请求指定的大小和访问的模式持久卷配置

  • PVC与PV的绑定是一对一的映射。没找到匹配的PV,那么PVC会无限期得处于unbound未绑定状态

volumes访问模式
  • ReadWriteOnce -- 该volume只能被单个节点以读写的方式映射

  • ReadOnlyMany -- 该volume可以被多个节点以只读方式映射

  • ReadWriteMany -- 该volume可以被多个节点以读写的方式映射

  • 在命令行中,访问模式可以简写为:

    • RWO - ReadWriteOnce
  • ROX - ReadOnlyMany

  • RWX -- ReadWriteMany

volumes回收策略
  • Retain:保留,需要手动回收

  • Recycle:回收,自动删除卷中数据(在当前版本中已经废弃)

  • Delete:删除,相关联的存储资产,如AWS EBS,GCE PD,Azure Disk,or OpenStack Cinder卷都会被删除

注意:只有NFS和HostPath支持回收利用

AWS EBS,GCE PD,Azure Disk,or OpenStack Cinder卷支持删除操作

volumes状态说明
  • Available 卷是一个空闲资源,尚未绑定到任何申领

  • Bound 该卷已经绑定到某申领

  • Released 所绑定的申领已被删除,但是关联存储资源尚未被集群回收

  • Failed 卷的自动回收操作失败

静态pv实例:
#在nfs主机中建立实验目录
[root@harbor ~]# mkdir /nfsdata/pv{1..3}

#编写创建pv的yml文件,pv是集群资源,不在任何namespace中
[root@k8s-master pvc]# vim pv.yml
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv1
spec:
  capacity:
    storage: 1Gi
  volumeMode: Filesystem		#注意缩进
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv1
    server: 172.25.250.250

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv2
spec:
  capacity:
    storage: 3Gi
  volumeMode: Filesystem
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv2
    server: 172.25.250.250
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv3
spec:
  capacity:
    storage: 5Gi
  volumeMode: Filesystem
  accessModes:
    - ReadOnlyMany
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv3
    server: 172.25.250.250

    
[root@k8s-master pvc]# kubectl apply -f pv.yml 
persistentvolume/pv1 unchanged
persistentvolume/pv2 unchanged
persistentvolume/pv3 created

[root@k8s-master pvc]# kubectl get pv
NAME   CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM          STORAGECLASS   VOLUMEATTRIBUTESCLASS   REASON   AGE
pv1    1Gi        RWO            Retain           Bound       default/pvc1   nfs            <unset>                          3m26s
pv2    3Gi        RWX            Retain           Available                  nfs            <unset>                          3m26s
pv3    5Gi        ROX            Retain           Available                  nfs            <unset>                          3m8s


#建立pvc,pvc是pv使用的申请,需要保证和pod在一个namesapce中
[root@k8s-master pvc]# vim pvc.ym
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: pvc1
spec:
  storageClassName: nfs
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 1Gi

---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: pvc2
spec:
  storageClassName: nfs
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 3Gi

---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: pvc3
spec:
  storageClassName: nfs
  accessModes:
    - ReadOnlyMany
  resources:
    requests:
      storage: 5Gi

[root@k8s-master pvc]# kubectl apply -f pvc.ym 
persistentvolumeclaim/pvc1 created
persistentvolumeclaim/pvc2 created
persistentvolumeclaim/pvc3 created

[root@k8s-master pvc]# kubectl get pv
NAME   CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS   CLAIM          STORAGECLASS   VOLUMEATTRIBUTESCLASS   REASON   AGE
pv1    1Gi        RWO            Retain           Bound    default/pvc1   nfs            <unset>                          3m20s
pv2    3Gi        RWX            Retain           Bound    default/pvc2   nfs            <unset>                          3m20s
pv3    5Gi        ROX            Retain           Bound    default/pvc3   nfs            <unset>                          3m20s

[root@k8s-master pvc]# kubectl get pvc
NAME         STATUS    VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
pvc1         Bound     pv1      1Gi        RWO            nfs            <unset>                 18s
pvc2         Bound     pv2      3Gi        RWX            nfs            <unset>                 18s
pvc3         Bound     pv3      5Gi        ROX            nfs            <unset>                 18s
test-claim   Pending                                      nfs-client     <unset>                 2m35s
www-web-0    Pending                                      nfs-client     <unset>                 3h29m


#在其他namespace中无法应用
[root@k8s-master pvc]# kubectl -n kube-system get pvc
No resources found in kube-system namespace.
在pod中使用pvc
[root@k8s-master pvc]# vim pod.yml
apiVersion: v1
kind: Pod
metadata:
  name: howe
spec:
  containers:
  - image: nginx
    name: nginx
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: vol1
  volumes:
  - name: vol1
    persistentVolumeClaim:
      claimName: pvc1
      
[root@k8s-master pvc]# kubectl apply -f pod.yml 
pod/howe created

[root@k8s-master pvc]# kubectl get pods -o wide 
NAME                        READY   STATUS      RESTARTS       AGE     IP           NODE                 NOMINATED NODE   READINESS GATES
howe                        1/1     Running     0              46s     10.244.1.8   k8s-node1.exam.com   <none>           <none>

[root@k8s-master pvc]# kubectl exec -it pods/howe -- /bin/bash
root@howe:~# cd /usr/share/nginx/html/
root@howe:/usr/share/nginx/html# ls

#nfs主机写入共享目录
[root@harbor nfsdata]# echo hehehe > /nfsdata/pv1/index.html

root@howe:/usr/share/nginx/html# ls
index.html
root@howe:/usr/share/nginx/html# curl localhost
hehehe

四 存储类storageclass

官网: GitHub - kubernetes-sigs/nfs-subdir-external-provisioner: Dynamic sub-dir volume provisioner on a remote NFS server.

4.1 StorageClass说明

  • StorageClass提供了一种描述存储类(class)的方法,不同的class可能会映射到不同的服务质量等级和备份策略或其他策略等。

  • 每个 StorageClass 都包含 provisioner、parameters 和 reclaimPolicy 字段, 这些字段会在StorageClass需要动态分配 PersistentVolume 时会使用到

4.2 StorageClass的属性

属性说明:存储类 | Kubernetes

Provisioner(存储分配器):用来决定使用哪个卷插件分配 PV,该字段必须指定。可以指定内部分配器,也可以指定外部分配器。外部分配器的代码地址为: kubernetes-incubator/external-storage,其中包括NFS和Ceph等。

Reclaim Policy(回收策略):通过reclaimPolicy字段指定创建的Persistent Volume的回收策略,回收策略包括:Delete 或者 Retain,没有指定默认为Delete。

4.3 存储分配器NFS Client Provisioner

源码地址:GitHub - kubernetes-sigs/nfs-subdir-external-provisioner: Dynamic sub-dir volume provisioner on a remote NFS server.

  • NFS Client Provisioner是一个automatic provisioner,使用NFS作为存储,自动创建PV和对应的PVC,本身不提供NFS存储,需要外部先有一套NFS存储服务。

  • PV以 {namespace}-{pvcName}-${pvName}的命名格式提供(在NFS服务器上)

  • PV回收的时候以 archieved-{namespace}-{pvcName}-${pvName} 的命名格式(在NFS服务器上)

4.4 部署NFS Client Provisioner

#上传镜像仓库
[root@k8s-master ~]# docker load -i nfs-subdir-external-provisioner-4.0.2.tar
Loaded image: registry.k8s.io/sig-storage/nfs-subdir-external-provisioner:v4.0.2

[root@k8s-master ~]# docker tag registry.k8s.io/sig-storage/nfs-subdir-external-provisioner:v4.0.2 reg.exam.com/sig-storage/nfs-subdir-external-provisioner:v4.0.2

[root@k8s-master ~]# docker push reg.exam.com/sig-storage/nfs-subdir-external-provisioner:v4.0.2
The push refers to repository [reg.exam.com/sig-storage/nfs-subdir-external-provisioner]
ad321585b8f5: Pushed 
1a5ede0c966b: Pushed 
v4.0.2: digest: sha256:f741e403b3ca161e784163de3ebde9190905fdbf7dfaa463620ab8f16c0f6423 size: 739

4.4.1 创建sa并授权

[root@k8s-master nfs]# vim rbac.yml
apiVersion: v1
kind: Namespace
metadata:
  name: nfs-client-provisioner
---
apiVersion: v1
kind: ServiceAccount
metadata:
  name: nfs-client-provisioner
  namespace: nfs-client-provisioner
---
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: nfs-client-provisioner-runner
rules:
  - apiGroups: [""]
    resources: ["nodes"]
    verbs: ["get", "list", "watch"]
  - apiGroups: [""]
    resources: ["persistentvolumes"]
    verbs: ["get", "list", "watch", "create", "delete"]
  - apiGroups: [""]
    resources: ["persistentvolumeclaims"]
    verbs: ["get", "list", "watch", "update"]
  - apiGroups: ["storage.k8s.io"]
    resources: ["storageclasses"]
    verbs: ["get", "list", "watch"]
  - apiGroups: [""]
    resources: ["events"]
    verbs: ["create", "update", "patch"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: run-nfs-client-provisioner
subjects:
  - kind: ServiceAccount
    name: nfs-client-provisioner
    namespace: nfs-client-provisioner
roleRef:
  kind: ClusterRole
  name: nfs-client-provisioner-runner
  apiGroup: rbac.authorization.k8s.io
---
kind: Role
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: leader-locking-nfs-client-provisioner
  namespace: nfs-client-provisioner
rules:
  - apiGroups: [""]
    resources: ["endpoints"]
    verbs: ["get", "list", "watch", "create", "update", "patch"]
---
kind: RoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
  name: leader-locking-nfs-client-provisioner
  namespace: nfs-client-provisioner
subjects:
  - kind: ServiceAccount
    name: nfs-client-provisioner
    namespace: nfs-client-provisioner
roleRef:
  kind: Role
  name: leader-locking-nfs-client-provisioner
  apiGroup: rbac.authorization.k8s.io
  
#查看rbac信息
[root@k8s-master nfs]# kubectl apply -f rbac.yml 
namespace/nfs-client-provisioner created
serviceaccount/nfs-client-provisioner created
clusterrole.rbac.authorization.k8s.io/nfs-client-provisioner-runner created
clusterrolebinding.rbac.authorization.k8s.io/run-nfs-client-provisioner created
role.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created
rolebinding.rbac.authorization.k8s.io/leader-locking-nfs-client-provisioner created

[root@k8s-master nfs]# kubectl -n nfs-client-provisioner get sa
NAME                     SECRETS   AGE
default                  0         26s
nfs-client-provisioner   0         27s

4.4.2 部署应用

[root@k8s-master nfs]# vim deployment.yml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: nfs-client-provisioner
  labels:
    app: nfs-client-provisioner
  namespace: nfs-client-provisioner
spec:
  replicas: 1
  strategy:
    type: Recreate
  selector:
    matchLabels:
      app: nfs-client-provisioner
  template:
    metadata:
      labels:
        app: nfs-client-provisioner
    spec:
      serviceAccountName: nfs-client-provisioner
      containers:
        - name: nfs-client-provisioner
          image: sig-storage/nfs-subdir-external-provisioner:v4.0.2
          volumeMounts:
            - name: nfs-client-root
              mountPath: /persistentvolumes
          env:
            - name: PROVISIONER_NAME
              value: k8s-sigs.io/nfs-subdir-external-provisioner
            - name: NFS_SERVER
              value: 172.25.250.250
            - name: NFS_PATH
              value: /nfsdata
      volumes:
        - name: nfs-client-root
          nfs:
            server: 172.25.250.250
            path: /nfsdata
            

[root@k8s-master nfs]# kubectl apply -f deployment.yml 
deployment.apps/nfs-client-provisioner created

[root@k8s-master nfs]# kubectl -n nfs-client-provisioner get deployments.apps nfs-client-provisioner 
NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
nfs-client-provisioner   1/1     1            1           4s

4.4.3 创建存储类

[root@k8s-master nfs]# vim class.yaml
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: nfs-client
provisioner: /nfs-subdir-external-provisioner
parameters:
  archiveOnDelete: "false"
  
[root@k8s-master nfs]# kubectl apply -f class.yaml 
storageclass.storage.k8s.io/nfs-client created

[root@k8s-master nfs]# kubectl get storageclasses.storage.k8s.io 
NAME         PROVISIONER                                   RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-client   k8s-sigs.io/nfs-subdir-external-provisioner   Delete          Immediate           false                  26s

4.4.4 创建pvc

[root@k8s-master nfs]# vim pvc.yml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: test-claim
spec:
  storageClassName: nfs-client
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 1G
      
[root@k8s-master nfs]# kubectl apply -f pvc.yml 
persistentvolumeclaim/test-claim created

#pending问题 可能由于没有pv资源
[root@k8s-master nfs]# kubectl get pvc
NAME         STATUS    VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
test-claim   Pending                                      nfs-client     <unset>                 10s

#检查nfs插件是否上传到镜像仓库
[root@k8s-master nfs]# kubectl get pvc
NAME         STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
test-claim   Bound    pvc-7f3fd08b-186e-42fa-9e0d-02ef7f0c9b36   1G         RWX            nfs-client     <unset>                 27s

4.4.5 创建测试pod

[root@k8s-master nfs]# vim pod.yml
apiVersion: v1
kind: Pod
metadata:
  name: test-pod
spec:
  containers:
  - name: test-pod
    image: busybox
    command:
      - "/bin/sh"
    args:
      - "-c"
      - "touch /mnt/SUCCESS && exit 0 || exit 1"
    volumeMounts:
      - name: nfs-pvc
        mountPath: "/mnt"
  restartPolicy: "Never"
  volumes:
    - name: nfs-pvc
      persistentVolumeClaim:
        claimName: test-claim

[root@k8s-master nfs]# kubectl apply -f pod.yml 
pod/test-pod created

[root@k8s-master nfs]# kubectl get pvc
NAME         STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
test-claim   Bound    pvc-7f3fd08b-186e-42fa-9e0d-02ef7f0c9b36   1G         RWX            nfs-client     <unset>                 27s

#nfs主机上查看
[root@harbor ~]# ls /nfsdata/default-test-claim-pvc-7f3fd08b-186e-42fa-9e0d-02ef7f0c9b36/
SUCCESS

[root@k8s-master nfs]# kubectl delete -f pod.yml 
pod "test-pod" deleted

4.4.6 设置默认存储类

  • 在未设定默认存储类时pvc必须指定使用类的名称

  • 在设定存储类后创建pvc时可以不用指定storageClassName

    #一次性指定多个pvc
    [root@k8s-master pvc]# vim pvc.yml
    apiVersion: v1
    kind: PersistentVolumeClaim
    metadata:
    name: pvc1
    spec:
    storageClassName: nfs-client
    accessModes:
    - ReadWriteOnce
    resources:
    requests:
    storage: 1Gi


    apiVersion: v1
    kind: PersistentVolumeClaim
    metadata:
    name: pvc2
    spec:
    storageClassName: nfs-client
    accessModes:
    - ReadWriteMany
    resources:
    requests:
    storage: 10Gi


    apiVersion: v1
    kind: PersistentVolumeClaim
    metadata:
    name: pvc3
    spec:
    storageClassName: nfs-client
    accessModes:
    - ReadOnlyMany
    resources:
    requests:
    storage: 15Gi

    [root@k8s-master pvc]# kubectl apply -f pvc.yml
    persistentvolumeclaim/pvc1 created
    persistentvolumeclaim/pvc2 created
    persistentvolumeclaim/pvc3 created

    [root@k8s-master pvc]# kubectl get pvc
    NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS VOLUMEATTRIBUTESCLASS AGE
    pvc1 Bound pvc-1d48c4c3-a6ef-487d-8d88-43297ca804a2 1Gi RWO nfs-client <unset> 12s
    pvc2 Bound pvc-27223a35-32c4-4768-8267-ad1866230b0d 10Gi RWX nfs-client <unset> 12s
    pvc3 Bound pvc-8ef40d48-991a-4b9d-afd8-3be6b4f45b89 15Gi ROX nfs-client <unset> 12s

设定默认存储类

[root@k8s-master nfs]# kubectl edit sc nfs-client 

#测试,未指定storageClassName参数
[root@k8s-master nfs]# vim pvc.yml 
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: test-claim
spec:
  storageClassName: nfs-client
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 1Gi
      
[root@k8s-master nfs]# kubectl apply -f pvc.yml 
persistentvolumeclaim/test-claim created

[root@k8s-master nfs]# kubectl get pvc
NAME         STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
test-claim   Bound    pvc-7900365a-ca7a-4de9-bcc4-0984b4149f58   1Gi        RWX            nfs-client     <unset>                 15s

五 statefulset控制器

5.1 功能特性

  • Statefulset是为了管理有状态服务的问提设计的

  • StatefulSet将应用状态抽象成了两种情况:

  • 拓扑状态:应用实例必须按照某种顺序启动。新创建的Pod必须和原来Pod的网络标识一样

  • 存储状态:应用的多个实例分别绑定了不同存储数据。

  • StatefulSet给所有的Pod进行了编号,编号规则是:(statefulset名称)-(序号),从0开始。

  • Pod被删除后重建,重建Pod的网络标识也不会改变,Pod的拓扑状态按照Pod的"名字+编号"的方式固定下来,并且为每个Pod提供了一个固定且唯一的访问入口,Pod对应的DNS记录。

5.2 StatefulSet的组成部分

  • Headless Service:用来定义pod网络标识,生成可解析的DNS记录

  • volumeClaimTemplates:创建pvc,指定pvc名称大小,自动创建pvc且pvc由存储类供应。

  • StatefulSet:管理pod的

5.3 构建方法

#建立无头服务
[root@k8s-master statefulset]# vim headless.yml
apiVersion: v1
kind: Service
metadata:
  name: nginx-svc
  labels:
    app: nginx
spec:
  ports:
  - port: 80
    name: web
  clusterIP: None
  selector:
    app: nginx


[root@k8s-master statefulset]# kubectl apply -f headless.yml 
service/nginx-svc created


#建立statefulset
[root@k8s-master statefulset]# vim statefulset.yml
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: web
spec:
  serviceName: "nginx-svc"
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx
        volumeMounts:
        - name: www
          mountPath: /usr/share/nginx/html
  volumeClaimTemplates:
  - metadata:
      name: www
    spec:
      storageClassName: nfs-client
      accessModes:
      - ReadWriteOnce
      resources:
        requests:
          storage: 1Gi
          
[root@k8s-master statefulset]# kubectl apply -f statefulset.yml 
statefulset.apps/web created

[root@k8s-master statefulset]# kubectl get pods 
NAME	READY   STATUS             RESTARTS       AGE
web-0	1/1     Running            0              6s
web-1	1/1     Running            0              4s
web-2	1/1     Running            0              2s

[root@harbor ~]# ls /nfsdata/
default-www-web-0-pvc-a3ceae44-d7f2-4514-8591-02ab8e8bed4c
default-www-web-1-pvc-5f5756d6-5d16-4a69-89ce-beef5321ba41
default-www-web-2-pvc-31726893-6723-42ef-ab56-32a35e263fba

5.4 测试:

#在nfs主机为每个pod建立index.html文件
[root@harbor ~]# cd /nfsdata/
[root@harbor nfsdata]# echo web-0 > default-www-web-0-pvc-a3ceae44-d7f2-4514-8591-02ab8e8bed4c/index.html
[root@harbor nfsdata]# echo web-1 > default-www-web-1-pvc-5f5756d6-5d16-4a69-89ce-beef5321ba41/index.html
[root@harbor nfsdata]# echo web-2 > default-www-web-2-pvc-31726893-6723-42ef-ab56-32a35e263fba/index.html

#在master建立测试pod访问web-0~2
[root@k8s-master statefulset]# kubectl run -it testpod --image busyboxplus
/ # curl web-0.nginx-svc
web-0
/ # curl web-1.nginx-svc
web-1
/ # curl web-2.nginx-svc
web-2
/ # 

#删掉重新建立statefulset
[root@k8s-master statefulset]# kubectl delete -f statefulset.yml 
statefulset.apps "web" deleted
[root@k8s-master statefulset]# kubectl apply -f statefulset.yml 
statefulset.apps/web created

#访问依然不变
[root@k8s-master statefulset]# kubectl attach testpod -c testpod -i -t
If you don't see a command prompt, try pressing enter.
/ # curl web-0.nginx-svc
web-0
/ # curl web-1.nginx-svc
web-1
/ # curl web-2.nginx-svc
web-2
/ # 

5.5 statefulset的弹缩

首先,想要弹缩的StatefulSet. 需先清楚是否能弹缩该应用

用命令改变副本数

$ kubectl scale statefulsets <stateful-set-name> --replicas=<new-replicas>

通过编辑配置改变副本数

$ kubectl edit statefulsets.apps <stateful-set-name>

statefulset有序回收

[root@k8s-master statefulset]# kubectl scale statefulset web --replicas 0
statefulset.apps/web scaled
[root@k8s-master statefulset]# kubectl delete -f statefulset.yml 
statefulset.apps "web" deleted
[root@k8s-master statefulset]# kubectl delete pvc --all
persistentvolumeclaim "www-web-0" deleted
persistentvolumeclaim "www-web-1" deleted
persistentvolumeclaim "www-web-2" deleted

[root@k8s-master statefulset]# kubectl scale statefulsets web --replicas=0

[root@k8s-master statefulset]# kubectl delete -f statefulset.yaml

[root@k8s-master statefulset]# kubectl delete pvc --all
相关推荐
荒Huang21 分钟前
Linux挖矿病毒(kswapd0进程使cpu爆满)
linux·运维·服务器
€☞扫地僧☜€2 小时前
docker 拉取MySQL8.0镜像以及安装
运维·数据库·docker·容器
hjjdebug2 小时前
linux 下 signal() 函数的用法,信号类型在哪里定义的?
linux·signal
其乐无涯2 小时前
服务器技术(一)--Linux基础入门
linux·运维·服务器
Diamond技术流2 小时前
从0开始学习Linux——网络配置
linux·运维·网络·学习·安全·centos
斑布斑布2 小时前
【linux学习2】linux基本命令行操作总结
linux·运维·服务器·学习
Spring_java_gg2 小时前
如何抵御 Linux 服务器黑客威胁和攻击
linux·服务器·网络·安全·web安全
✿ ༺ ོIT技术༻2 小时前
Linux:认识文件系统
linux·运维·服务器
会掉头发3 小时前
Linux进程通信之共享内存
linux·运维·共享内存·进程通信
我言秋日胜春朝★3 小时前
【Linux】冯诺依曼体系、再谈操作系统
linux·运维·服务器