k8s中的存储

目录

一、configmap

1、configmap的功能

2、configmap的使用场景

3、configmap创建方式

(1)字面值创建

(2)通过文件创建

(3)通过目录创建

(4)通过yaml文件创建

(5)configmap的使用方法

使用configmap填充环境变量

通过数据卷使用configmap

利用configMap填充pod的配置文件

通过热更新cm修改配置

二、secrets配置管理

1、secrets功能介绍

2、secrets的创建

(1)从文件创建

(2)编写yaml文件

3、secrets使用方法

(1)将secrets挂载到volume中

(2)向指定路径映射secrets密钥

(3)将secrets设置为环境变量

[(4)存储docker registry的认证信息](#(4)存储docker registry的认证信息)

三、volumes配置管理

1、kubernets支持的卷类型

2、emptyDir卷

3、hostpath卷

4、nfs卷

(1)部署一台nfs共享主机并在所有k8s节点中安装nfs-utils

(2)部署nfs卷

5、persistentVolume持久卷

(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)

(2)volumes访问模式

(3)volumes回收策略

(4)volumes状态说明

示例:静态pv

四、存储类storageclass

1、storageclass说明

2、storageclass的属性

[3、存储分配器NFS Client Provisioner](#3、存储分配器NFS Client Provisioner)

[4、部署NFS Client Provisioner](#4、部署NFS Client Provisioner)

(1)创建sa并授权

(2)部署应用

(3)创建存储类

(4)创建pvc

(5)创建测试pod

(6)设置默认存储类

五、statefulset控制器

1、功能特性

2、statefulset的组成部分

3、构建方法

4、测试

5、statefulset的弹缩


一、configmap

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

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

3、configmap创建方式

(1)字面值创建

[root@k8s-master ~]# kubectl create cm zx-config --from-literal fname=zx --from-literal lname=zhou
configmap/zx-config created
[root@k8s-master ~]# kubectl describe cm zx-config 
Name:         zx-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
fname:
----
zx
lname:
----
zhou

BinaryData
====

Events:  <none>
[root@k8s-master ~]# kubectl get configmaps 
NAME               DATA   AGE
kube-root-ca.crt   1      23h
zx-config          2      35s
[root@k8s-master ~]# 

(2)通过文件创建

[root@k8s-master ~]# cat /etc/resolv.conf 
# Generated by NetworkManager
search zx.org
nameserver 114.114.114.114
[root@k8s-master ~]# kubectl create cm zx2-config --from-file /etc/resolv.conf 
configmap/zx2-config created
[root@k8s-master ~]# kubectl describe cm zx2-config 
Name:         zx2-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
resolv.conf:
----
# Generated by NetworkManager
search zx.org
nameserver 114.114.114.114


BinaryData
====

Events:  <none>
[root@k8s-master ~]# 

(3)通过目录创建

[root@k8s-master ~]# mkdir zxconfig
[root@k8s-master ~]# cp /etc/fstab /etc/rc.d/rc.local zxconfig/
[root@k8s-master ~]# kubectl create cm zx3-config --from-file zxconfig/
configmap/zx3-config created
[root@k8s-master ~]# kubectl describe cm zx3-config 
Name:         zx3-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
fstab:
----

(4)通过yaml文件创建

[root@k8s-master ~]# kubectl create cm zx4-config --from-literal db_host=172.25.254.200 --from-literal db_port=3306 --dry-run=client -o yaml > zx-config.yml
[root@k8s-master ~]# vim zx-config.yml 
[root@k8s-master ~]# kubectl apply -f zx-config.yml 
configmap/zx4-config created
[root@k8s-master ~]# kubectl describe cm zx4-config 
Name:         zx4-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
db_host:
----
172.25.254.200
db_port:
----
3306

BinaryData
====

Events:  <none>
[root@k8s-master ~]# cat zx-config.yml 
apiVersion: v1
data:
  db_host: 172.25.254.200
  db_port: "3306"
kind: ConfigMap
metadata:
  creationTimestamp: null
  name: zx4-config

(5)configmap的使用方法

  • 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

使用configmap填充环境变量
#将cm中的内容映射为指定变量
[root@k8s-master ~]# vim testpod1.yml
[root@k8s-master ~]# kubectl apply -f testpod1.yml 
pod/testpod created

[root@k8s-master ~]# kubectl logs pods/testpod
KUBERNETES_PORT=tcp://10.96.0.1:443
KUBERNETES_SERVICE_PORT=443
MYAPP1_PORT_80_TCP=tcp://10.99.26.240:80
HOSTNAME=testpod
SHLVL=1
HOME=/
MYAPP1_SERVICE_HOST=10.99.26.240
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
key1=172.25.254.200
key2=3306
MYAPP1_SERVICE_PORT=80
MYAPP1_PORT=tcp://10.99.26.240:80
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
MYAPP1_PORT_80_TCP_ADDR=10.99.26.240
PWD=/
KUBERNETES_SERVICE_HOST=10.96.0.1
MYAPP1_PORT_80_TCP_PORT=80
MYAPP1_PORT_80_TCP_PROTO=tcp

# 把cm中的值直接映射为变量
[root@k8s-master ~]# vim testpod1.yml 
[root@k8s-master ~]# cat testpod1.yml 
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: reg.zx.org/library/busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - env
    envFrom:
    - configMapRef:
        name: zx4-config
  restartPolicy: Never

[root@k8s-master ~]# kubectl delete -f testpod1.yml 
pod "testpod" deleted
[root@k8s-master ~]# kubectl apply -f testpod1.yml 
pod/testpod created
[root@k8s-master ~]# kubectl logs pods/testpod
KUBERNETES_SERVICE_PORT=443
KUBERNETES_PORT=tcp://10.96.0.1:443
MYAPP1_PORT_80_TCP=tcp://10.99.26.240:80
HOSTNAME=testpod
SHLVL=1
HOME=/
db_port=3306
MYAPP1_SERVICE_HOST=10.99.26.240
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
MYAPP1_PORT=tcp://10.99.26.240:80
MYAPP1_SERVICE_PORT=80
KUBERNETES_SERVICE_PORT_HTTPS=443
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
MYAPP1_PORT_80_TCP_ADDR=10.99.26.240
PWD=/
KUBERNETES_SERVICE_HOST=10.96.0.1
MYAPP1_PORT_80_TCP_PORT=80
MYAPP1_PORT_80_TCP_PROTO=tcp
db_host=172.25.254.200


# pod命令行中使用变量
[root@k8s-master ~]# vim testpod1.yml 
[root@k8s-master ~]# kubectl apply -f testpod1.yml 
pod/testpod created
[root@k8s-master ~]# kubectl logs pods/testpod
172.25.254.200 3306
[root@k8s-master ~]# cat testpod1.yml 
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: reg.zx.org/library/busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - echo ${db_host} ${db_port}
    envFrom:
    - configMapRef:
        name: zx4-config
  restartPolicy: Never
通过数据卷使用configmap
[root@k8s-master ~]# cp testpod1.yml testpod2.yml
[root@k8s-master ~]# vim testpod2.yml 
[root@k8s-master ~]# kubectl apply -f testpod2.yml 
pod/testpod created
[root@k8s-master ~]# kubectl logs testpod 
172.25.254.200

[root@k8s-master ~]# cat testpod2.yml 
apiVersion: v1
kind: Pod
metadata:
  labels:
    run: testpod
  name: testpod
spec:
  containers:
  - image: reg.zx.org/library/busyboxplus:latest
    name: testpod
    command:
    - /bin/sh
    - -c
    - cat /config/db_host
    volumeMounts:
      - name: config-volume
        mountPath: /config
  volumes:
  - name: config-volume
    configMap:
      name: zx4-config
  restartPolicy: Never
利用configMap填充pod的配置文件
[root@k8s-master ~]# vim nginx.conf
[root@k8s-master ~]# cat nginx.conf 
server {
  listen 8000;
  server_name _;
  root /usr/share/nginx/html;
  index index.html;
}
[root@k8s-master ~]# kubectl create cm nginx-conf --from-file nginx.conf 
configmap/nginx.conf created
[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>
[root@k8s-master ~]# kubectl create deployment nginx.conf --image reg.zx.org/library/nginx:latest --replicas 1 --dry-run=client -o yaml > nginx.yml
[root@k8s-master ~]# vim nginx.yml 
[root@k8s-master ~]# kubectl apply -f nginx.yml 

[root@k8s-master ~]# kubectl get deployments.apps 
NAME    READY   UP-TO-DATE   AVAILABLE   AGE
nginx   1/1     1            1           63s

[root@k8s-master ~]# kubectl get pods -o wide
NAME                     READY   STATUS    RESTARTS   AGE    IP            NODE               NOMINATED NODE   READINESS GATES
nginx-688685cfd4-8cmbw   1/1     Running   0          105s   10.244.1.96   k8s-node1.zx.org   <none>           <none>

[root@k8s-master ~]# cat nginx.yml 
apiVersion: apps/v1
kind: Deployment
metadata:
  creationTimestamp: null
  labels:
    app: nginx
  name: nginx
spec:
  replicas: 1
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - image: reg.zx.org/library/nginx:latest
        name: nginx
        volumeMounts:
        - name: config-volume
          mountPath: /etc/nginx/conf.d

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

[root@k8s-master ~]# curl 10.244.1.96:8000
通过热更新cm修改配置
[root@k8s-master ~]# kubectl edit cm nginx-conf     # 修改端口为8080
configmap/nginx-conf edited

[root@k8s-master ~]# 
[root@k8s-master ~]# kubectl exec pods/nginx-688685cfd4-8cmbw -- cat /etc/nginx/conf.d/nginx.conf
server {
  listen 8080;
  server_name _;
  root /usr/share/nginx/html;
  index index.html;
}

配置文件修改后不会生效,需要删除pod后控制器会重建pod,这时就生效了

[root@k8s-master ~]# kubectl delete pods/nginx-688685cfd4-8cmbw 
pod "nginx-688685cfd4-8cmbw" deleted

[root@k8s-master ~]# kubectl get pods -o wide
NAME                     READY   STATUS    RESTARTS   AGE   IP             NODE               NOMINATED NODE   READINESS GATES
nginx-688685cfd4-t4pvm   1/1     Running   0          18s   10.244.2.103   k8s-node2.zx.org   <none>           <none>
[root@k8s-master ~]# curl 10.244.2.103:8080

二、secrets配置管理

1、secrets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、secrets的创建

(1)从文件创建

[root@k8s-master ~]# mkdir secrets
[root@k8s-master ~]# cd secrets/
[root@k8s-master secrets]# echo -n zx > username.txt
[root@k8s-master secrets]# echo -n zhou > password.txt
[root@k8s-master secrets]# kubectl create secret generic userlist --from-file username.txt --from-file password.txt 
secret/userlist created
[root@k8s-master secrets]# kubectl get secrets userlist -o yaml
apiVersion: v1
data:
  password.txt: emhvdQ==
  username.txt: eng=
kind: Secret
metadata:
  creationTimestamp: "2024-09-17T03:11:20Z"
  name: userlist
  namespace: default
  resourceVersion: "37150"
  uid: aa11f642-fdbd-4ff1-84c5-31785dcb4f41
type: Opaque

(2)编写yaml文件

[root@k8s-master secrets]# echo -n zx | base64
eng=
[root@k8s-master secrets]# echo -n zhou | base64
emhvdQ==
[root@k8s-master secrets]# kubectl create secret generic userlist --dry-run=client -o yaml > userlist.yml
[root@k8s-master secrets]# vim userlist.yml 
[root@k8s-master secrets]# kubectl apply -f userlist.yml 
Warning: resource secrets/userlist is missing the kubectl.kubernetes.io/last-applied-configuration annotation which is required by kubectl apply. kubectl apply should only be used on resources created declaratively by either kubectl create --save-config or kubectl apply. The missing annotation will be patched automatically.
secret/userlist configured
[root@k8s-master secrets]# kubectl describe secrets userlist
Name:         userlist
Namespace:    default
Labels:       <none>
Annotations:  <none>

Type:  Opaque

Data
====
password:      4 bytes
password.txt:  4 bytes
username:      2 bytes
username.txt:  2 bytes

[root@k8s-master secrets]# cat userlist.yml 
apiVersion: v1
kind: Secret
metadata:
  creationTimestamp: null
  name: userlist
type: Opaque
data:
  username: eng=
  password: emhvdQ==

3、secrets使用方法

(1)将secrets挂载到volume中

[root@k8s-master secrets]# kubectl run nginx --image reg.zx.org/library/nginx:latest --dry-run=client -o yaml > pod1.yml
[root@k8s-master secrets]# vim pod1.yml 
[root@k8s-master secrets]# kubectl apply -f pod1.yml 
pod/nginx created
[root@k8s-master secrets]# kubectl exec  pods/nginx -it -- /bin/bash
root@nginx:/# cat /secret/
cat: /secret/: Is a directory
root@nginx:/# cd /secret/
root@nginx:/secret# ls
password  password.txt	username  username.txt
root@nginx:/secret# cat password
zhouroot@nginx:/secret# cat username
zxroot@nginx:/secret# 
root@nginx:/secret# exit
exit

[root@k8s-master secrets]# cat pod1.yml 
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    run: nginx
  name: nginx
spec:
  containers:
  - image: reg.zx.org/library/nginx:latest
    name: nginx
    volumeMounts:
    - name: secrets
      mountPath: /secret
      readOnly: true

  volumes:
  - name: secrets
    secret:
      secretName: userlist

(2)向指定路径映射secrets密钥

[root@k8s-master secrets]# kubectl delete -f pod1.yml 
pod "nginx" deleted
[root@k8s-master secrets]# 
[root@k8s-master secrets]# cp pod1.yml pod2.yml
[root@k8s-master secrets]# vim pod2.yml 
[root@k8s-master secrets]# kubectl apply -f pod2.yml 
pod/nginx created
[root@k8s-master secrets]# kubectl exec pods/nginx -it -- /bin/bash
root@nginx:/# cd secret/
root@nginx:/secret# ls
my-users
root@nginx:/secret# cd my-users
root@nginx:/secret/my-users# ls
username
root@nginx:/secret/my-users# cat username 
zxroot@nginx:/secret/my-users# 
root@nginx:/secret/my-users# exit
exit
[root@k8s-master secrets]# cat pod2.yml 
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    run: nginx
  name: nginx
spec:
  containers:
  - image: reg.zx.org/library/nginx:latest
    name: nginx
    volumeMounts:
    - name: secrets
      mountPath: /secret
      readOnly: true

  volumes:
  - name: secrets
    secret:
      secretName: userlist
      items:
      - key: username
        path: my-users/username

(3)将secrets设置为环境变量

[root@k8s-master secrets]# cp pod1.yml pod3.yml
[root@k8s-master secrets]# vim pod3.yml 
[root@k8s-master secrets]# kubectl apply -f pod3.yml 
pod/busybox created

[root@k8s-master secrets]# kubectl logs pods/busybox
KUBERNETES_PORT=tcp://10.96.0.1:443
KUBERNETES_SERVICE_PORT=443
MYAPP1_PORT_80_TCP=tcp://10.99.26.240:80
HOSTNAME=busybox
SHLVL=1
HOME=/
MYAPP1_SERVICE_HOST=10.99.26.240
USERNAME=zx
KUBERNETES_PORT_443_TCP_ADDR=10.96.0.1
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
KUBERNETES_PORT_443_TCP_PORT=443
KUBERNETES_PORT_443_TCP_PROTO=tcp
MYAPP1_SERVICE_PORT=80
MYAPP1_PORT=tcp://10.99.26.240:80
KUBERNETES_PORT_443_TCP=tcp://10.96.0.1:443
MYAPP1_PORT_80_TCP_ADDR=10.99.26.240
KUBERNETES_SERVICE_PORT_HTTPS=443
PASS=zhou
PWD=/
KUBERNETES_SERVICE_HOST=10.96.0.1
MYAPP1_PORT_80_TCP_PORT=80
MYAPP1_PORT_80_TCP_PROTO=tcp

[root@k8s-master secrets]# cat pod3.yml 
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    run: busybox
  name: busybox
spec:
  containers:
  - image: reg.zx.org/library/busyboxplus:latest
    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

(4)存储docker registry的认证信息

[root@k8s-master secrets]# 
[root@k8s-master secrets]# docker login reg.zx.org
Authenticating with existing credentials...
WARNING! Your password will be stored unencrypted in /root/.docker/config.json.
Configure a credential helper to remove this warning. See
https://docs.docker.com/engine/reference/commandline/login/#credential-stores

Login Succeeded
[root@k8s-master secrets]# docker load -i game2048.tar 
011b303988d2: Loading layer   5.05MB/5.05MB
36e9226e74f8: Loading layer  51.46MB/51.46MB
192e9fad2abc: Loading layer  3.584kB/3.584kB
6d7504772167: Loading layer  4.608kB/4.608kB
88fca8ae768a: Loading layer  629.8kB/629.8kB
Loaded image: timinglee/game2048:latest
[root@k8s-master secrets]# docker tag timinglee/game2048:latest reg.zx.org/zx/game2048:latest
[root@k8s-master secrets]# docker push reg.zx.org/zx/game2048:latest
The push refers to repository [reg.zx.org/zx/game2048]
88fca8ae768a: Pushed 
6d7504772167: Pushed 
192e9fad2abc: Pushed 
36e9226e74f8: Pushed 
011b303988d2: Pushed 
latest: digest: sha256:8a34fb9cb168c420604b6e5d32ca6d412cb0d533a826b313b190535c03fe9390 size: 1364

# 退出登录无法pull
[root@k8s-master secrets]# docker logout reg.zx.org
Removing login credentials for reg.zx.org
[root@k8s-master secrets]# docker pull reg.zx.org/zx/game2048:latest
Error response from daemon: unauthorized: unauthorized to access repository: zx/game2048, action: pull: unauthorized to access repository: zx/game2048, action: pull

#建立用于docker认证的secret
[root@k8s-master secrets]# kubectl create secret docker-registry docker-auth --docker-server reg.zx.org --docker-username admin --docker-password 123 --docker-email zx@zx.org
secret/docker-auth created

[root@k8s-master secrets]# vim pod4.yml 
[root@k8s-master secrets]# kubectl apply -f pod4.yml 
pod/game2048 created

[root@k8s-master secrets]# kubectl delete -f pod3.yml 
pod "busybox" deleted
[root@k8s-master secrets]# kubectl get pods
NAME                     READY   STATUS    RESTARTS   AGE
game2048                 1/1     Running   0          24s
nginx-688685cfd4-t4pvm   1/1     Running   0          39m

[root@k8s-master secrets]# cat pod4.yml 
apiVersion: v1
kind: Pod
metadata:
  creationTimestamp: null
  labels:
    run: game2048
  name: game2048
spec:
  containers:
  - image: reg.zx.org/zx/game2048:latest
    name: game2048
  imagePullSecrets:        #不设定docker认证时无法下载镜像
  - name: docker-auth

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。

1、kubernets支持的卷类型

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

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。

    [root@k8s-master ~]# mkdir volumes
    [root@k8s-master ~]# cd volumes/
    [root@k8s-master volumes]# vim pod1.yml
    [root@k8s-master volumes]# kubectl apply -f pod1.yml
    pod/vol1 created
    [root@k8s-master volumes]# kubectl describe pods vol1
    Name: vol1
    Namespace: default
    Priority: 0
    Service Account: default
    Node: k8s-node1.zx.org/172.25.254.10
    Start Time: Tue, 17 Sep 2024 11:55:11 +0800
    Labels: <none>
    Annotations: <none>
    Status: Running
    IP: 10.244.1.102
    IPs:
    IP: 10.244.1.102
    Containers:
    vm1:
    Container ID: docker://7c42d2f2959de768f864b6ab1ed4ae095fc9945023550be9540406981ce9dbcb
    Image: reg.zx.org/library/busyboxplus:latest
    Image ID: docker-pullable://reg.zx.org/library/busyboxplus@sha256:9d1c242c1fd588a1b8ec4461d33a9ba08071f0cc5bb2d50d4ca49e430014ab06
    Port: <none>
    Host Port: <none>
    Command:
    /bin/sh
    -c
    sleep 30000000
    State: Running
    Started: Tue, 17 Sep 2024 11:55:13 +0800
    Ready: True
    Restart Count: 0
    Environment: <none>
    Mounts:
    /cache from cache-vol (rw)
    /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-5kmp9 (ro)
    vm2:
    Container ID: docker://49385fe0330033611426759d23cb5c9254aef4d1f17e850bcb2cfaa0d846407f
    Image: reg.zx.org/library/nginx:latest
    Image ID: docker-pullable://reg.zx.org/library/nginx@sha256:127262f8c4c716652d0e7863bba3b8c45bc9214a57d13786c854272102f7c945
    Port: <none>
    Host Port: <none>
    State: Running
    Started: Tue, 17 Sep 2024 11:55:14 +0800
    Ready: True
    Restart Count: 0
    Environment: <none>
    Mounts:
    /usr/share/nginx/html from cache-vol (rw)
    /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-5kmp9 (ro)
    Conditions:
    Type Status
    PodReadyToStartContainers True
    Initialized True
    Ready True
    ContainersReady True
    PodScheduled True
    Volumes:
    cache-vol:
    Type: EmptyDir (a temporary directory that shares a pod's lifetime)
    Medium: Memory
    SizeLimit: 100Mi
    kube-api-access-5kmp9:
    Type: Projected (a volume that contains injected data from multiple sources)
    TokenExpirationSeconds: 3607
    ConfigMapName: kube-root-ca.crt
    ConfigMapOptional: <nil>
    DownwardAPI: true
    QoS Class: BestEffort
    Node-Selectors: <none>
    Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
    node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
    Events:
    Type Reason Age From Message
    ---- ------ ---- ---- -------
    Normal Scheduled 5s default-scheduler Successfully assigned default/vol1 to k8s-node1.zx.org
    Normal Pulling 3s kubelet Pulling image "reg.zx.org/library/busyboxplus:latest"
    Normal Pulled 3s kubelet Successfully pulled image "reg.zx.org/library/busyboxplus:latest" in 377ms (377ms including waiting). Image size: 12855024 bytes.
    Normal Created 3s kubelet Created container vm1
    Normal Started 2s kubelet Started container vm1
    Normal Pulling 2s kubelet Pulling image "reg.zx.org/library/nginx:latest"
    Normal Pulled 2s kubelet Successfully pulled image "reg.zx.org/library/nginx:latest" in 283ms (283ms including waiting). Image size: 187694648 bytes.
    Normal Created 2s kubelet Created container vm2
    Normal Started 1s kubelet Started container vm2
    [root@k8s-master volumes]#
    [root@k8s-master volumes]# kubectl exec -it pods/vol1 -c vm1 -- /bin/sh
    / # cd /cache/
    /cache # curl localhost

    <html> <head><title>403 Forbidden</title></head> <body>

    403 Forbidden


    nginx/1.27.1
    </body> </html> /cache # /cache # echo zx > index.html /cache # curl localhost zx /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 /cache #

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。

    [root@k8s-master volumes]# vim pod2.yml
    [root@k8s-master volumes]# cat pod2.yml
    apiVersion: v1
    kind: Pod
    metadata:
    name: vol1
    spec:
    containers:
    - image: reg.zx.org/library/nginx:latest
    name: vm1
    volumeMounts:
    - mountPath: /usr/share/nginx/html
    name: cache-vol
    volumes:
    - name: cache-vol
    hostPath:
    path: /data
    type: DirectoryOrCreate

    [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
    nginx-688685cfd4-t4pvm 1/1 Running 0 60m 10.244.2.103 k8s-node2.zx.org <none> <none>
    [root@k8s-master volumes]# curl 10.244.2.104
    [root@k8s-node2 ~]# echo zx > /data/index.html
    [root@k8s-master volumes]# curl 10.244.2.104
    zx

    #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

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。

(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@docker-node1 ~]# dnf install nfs-utils -y
[root@docker-node1 ~]# systemctl enable --now nfs-server.service
[root@docker-node1 ~]# vim /etc/exports
[root@docker-node1 ~]# cat /etc/exports
/nfsdata   *(rw,sync,no_root_squash)

[root@docker-node1 ~]# mkdir /nfsdata
[root@docker-node1 ~]# exportfs -rv
exporting *:/nfsdata
[root@docker-node1 ~]# 
[root@docker-node1 ~]# showmount  -e
Export list for docker-node1.zx.org:
/nfsdata *

[root@docker-node1 ~]# ll /data/
total 8
drwxr-xr-x 2            10000 10000    6 Sep 10 18:10 ca_download
drwxr-xr-x 2 root             root    42 Sep 10 18:08 certs
drwxr-xr-x 2            10000 10000    6 Sep 10 18:10 chart_storage
drwx------ 3 systemd-coredump input   18 Sep 10 18:11 database
drwxr-xr-x 2            10000 10000    6 Sep 10 18:10 job_logs
drwxr-xr-x 2 systemd-coredump input   22 Sep 17 12:15 redis
drwxr-xr-x 3            10000 10000   20 Sep 10 18:28 registry
drwxr-xr-x 6 root             root    58 Sep 16 08:00 secret
-rw-r--r-- 1 root             root  2126 Sep 10 18:03 zx.org.crt
-rw------- 1 root             root  3272 Sep 10 18:03 zx.org.key
[root@docker-node1 ~]# 

#在k8s所有节点中安装nfs-utils
[root@k8s-master & node1 & node2  ~]# dnf install nfs-utils -y

[root@k8s-node2 ~]# showmount  -e 172.25.254.100    # 三台主机上
Export list for 172.25.254.100:
/nfsdata *

(2)部署nfs卷

[root@k8s-master volumes]# vim pod3.yml
[root@k8s-master volumes]# kubectl apply -f pod3.yml
pod/vol1 created
[root@k8s-master volumes]# cat pod3.yml 
apiVersion: v1
kind: Pod
metadata:
  name: vol1
spec:
  containers:
  - image: reg.zx.org/library/nginx:latest
    name: vm1
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: cache-vol
  volumes:
  - name: cache-vol
    nfs:
      server: 172.25.254.100
      path: /nfsdata

[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          7s    10.244.2.106   k8s-node2.zx.org   <none>           <none>

[root@k8s-master volumes]# curl 10.244.2.106
[root@docker-node1 ~]# echo zx > /nfsdata/index.html
[root@k8s-master volumes]# curl 10.244.2.106

5、persistentVolume持久卷

(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

(2)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

(3)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

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

(4)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

示例:静态pv
[root@docker-node1 ~]# mkdir  /nfsdata/pv{1..3}
[root@docker-node1 nfsdata]# ls
pv1  pv2  pv3

[root@k8s-node1 ~]# showmount  -e 172.25.254.100    # 节点(三个)有内容

#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
[root@k8s-master pvc]# kubectl apply -f pv.yml 
[root@k8s-master pvc]# cat pv.yml 
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv1
spec:
  capacity:
    storage: 5Gi
  volumeMode: Filesystem
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv1
    server: 172.25.254.100

---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv2
spec:
  capacity:
    storage: 15Gi
  volumeMode: Filesystem
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv2
    server: 172.25.254.100
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: pv3
spec:
  capacity:
    storage: 25Gi
  volumeMode: Filesystem
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Retain
  storageClassName: nfs
  nfs:
    path: /nfsdata/pv3
    server: 172.25.254.100

[root@k8s-master pvc]# kubectl get pv
NAME   CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS   VOLUMEATTRIBUTESCLASS   REASON   AGE
pv1    5Gi        RWO            Retain           Available           nfs            <unset>                          2m23s
pv2    15Gi       RWX            Retain           Available           nfs            <unset>                          2m23s
pv3    25Gi       RWX            Retain           Available           nfs            <unset>                          65s
#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.yml
[root@k8s-master pvc]# kubectl apply -f pvc.yml 
persistentvolumeclaim/pvc1 created
persistentvolumeclaim/pvc2 created
persistentvolumeclaim/pvc3 created
[root@k8s-master pvc]# cat pvc.yml 
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: 10Gi

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

#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.
[root@k8s-master pvc]# vim pod.yml
[root@k8s-master pvc]# kubectl apply -f pod.yml
[root@k8s-master pvc]# cat pod.yml 
apiVersion: v1
kind: Pod
metadata:
  name: zx
spec:
  containers:
  - image: reg.zx.org/library/nginx:latest
    name: nginx
    volumeMounts:
    - mountPath: /usr/share/nginx/html
      name: vol1
  volumes:
  - name: vol1
    persistentVolumeClaim:
      claimName: pvc1

四、存储类storageclass

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

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

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。

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、部署NFS Client Provisioner

(1)创建sa并授权

[root@k8s-master ~]# mkdir storageclass
[root@k8s-master ~]# cd storageclass/
[root@k8s-master storageclass]# vim rbac.yml
[root@k8s-master storageclass]# 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 storageclass]# 
[root@k8s-master storageclass]# kubectl -n nfs-client-provisioner get sa
NAME                     SECRETS   AGE
default                  0         9s
nfs-client-provisioner   0         9s

[root@k8s-master storageclass]# cat 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

(2)部署应用

[root@k8s-master ~]# docker load -i nfs-subdir-external-provisioner-4.0.2.tar 

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

[root@k8s-master storageclass]# vim deployment.yml
[root@k8s-master storageclass]# kubectl apply -f deployment.yml 
deployment.apps/nfs-client-provisioner created
[root@k8s-master storageclass]# 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           17s
[root@k8s-master storageclass]# cat 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: reg.zx.org/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: 172.25.254.100
            - name: NFS_PATH
              value: /nfsdata
      volumes:
        - name: nfs-client-root
          nfs:
            server: 172.25.254.100
            path: /nfsdata

(3)创建存储类

[root@k8s-master storageclass]# vim class.yaml
[root@k8s-master storageclass]# kubectl apply -f class.yaml 
storageclass.storage.k8s.io/nfs-client created
[root@k8s-master storageclass]# kubectl get storageclasses.storage.k8s.io
NAME         PROVISIONER                                   RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-client   k8s-sigs.io/nfs-subdir-external-provisioner   Delete          Immediate           false                  9s
[root@k8s-master storageclass]# cat class.yaml 
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: nfs-client
provisioner: k8s-sigs.io/nfs-subdir-external-provisioner
parameters:
  archiveOnDelete: "false"

(4)创建pvc

[root@k8s-master storageclass]# vim pvc.yml
[root@k8s-master storageclass]# kubectl apply -f pvc.yml
persistentvolumeclaim/test-claim created
[root@k8s-master storageclass]# kubectl get pvc
NAME         STATUS    VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   VOLUMEATTRIBUTESCLASS   AGE
pvc1         Bound     pv1                                        5Gi        RWO            nfs            <unset>                 35m
pvc2         Bound     pv2                                        15Gi       RWX            nfs            <unset>                 35m
pvc3         Pending                                                                        nfs            <unset>                 35m
test-claim   Bound     pvc-3aeba466-5930-404a-b0d6-de108ac56473   1G         RWX            nfs-client     <unset>                 7s
[root@k8s-master storageclass]# cat pvc.yml 
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: test-claim
spec:
  storageClassName: nfs-client
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: 1G

(5)创建测试pod

[root@k8s-master storageclass]# vim pod.yml
[root@k8s-master storageclass]# kubectl apply -f pod.yml 
pod/test-pod created
[root@k8s-master storageclass]# cat pod.yml 
kind: Pod
apiVersion: v1
metadata:
  name: test-pod
spec:
  containers:
  - name: test-pod
    image: reg.zx.org/library/busyboxplus:latest
    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 storageclass]# kubectl get pods
NAME       READY   STATUS      RESTARTS   AGE
test-pod   0/1     Completed   0          4m7s

(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

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

五、statefulset控制器

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。

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

3、构建方法

#建立无头服务
[root@k8s-master pvc]# vim headless.yml
[root@k8s-master pvc]# kubectl apply -f headless.yml 
service/nginx-svc created
[root@k8s-master pvc]# kubectl get svc
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP   3d4h
myapp1       ClusterIP   10.99.26.240   <none>        80/TCP    2d6h
nginx-svc    ClusterIP   None           <none>        80/TCP    116s

#建立statefulset
[root@k8s-master pvc]# vim statefulset.yml
[root@k8s-master pvc]# kubectl apply -f statefulset.yml 
statefulset.apps/web created
[root@k8s-master pvc]# kubectl get pods
NAME    READY   STATUS    RESTARTS   AGE
web-0   1/1     Running   0          26s
web-1   1/1     Running   0          21s
web-2   1/1     Running   0          16s

[root@k8s-master pvc]# cat 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 pvc]# cat 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: reg.zx.org/library/nginx:latest
     volumeMounts:
       - name: www
         mountPath: /usr/share/nginx/html
 volumeClaimTemplates:
  - metadata:
     name: www
    spec:
     storageClassName: nfs-client
     accessModes:
     - ReadWriteOnce
     resources:
      requests:
       storage: 1Gi

4、测试

#为每个pod建立index.html文件
[root@docker-node1 nfsdata]# echo web-0 > default-www-web-0-pvc-31d62985-0b9a-43fc-8d5f-9451c215f79f/index.html
[root@docker-node1 nfsdata]# echo web-1 > default-www-web-1-pvc-fde53553-f3a5-4a4a-b880-b94b38fca816/index.html
[root@docker-node1 nfsdata]# echo web-2 > default-www-web-2-pvc-7866fc97-e410-4224-be13-266ca7cf8c10/index.html

#建立测试pod访问web-0~2
[root@k8s-master pvc]# kubectl run -it testpod --image reg.zx.org/library/busyboxplus:latest
If you don't see a command prompt, try pressing enter.
/ # curl 10.244.2.109
web-0
[root@k8s-master pvc]# kubectl run -it testpod --image reg.zx.org/library/busyboxplus:latest
/ # 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
或者改变pod的数量
[root@k8s-master pvc]# kubectl scale statefulset web --replicas 0
statefulset.apps/web scaled
[root@k8s-master pvc]# kubectl get pods -o wide
No resources found in default namespace.
[root@k8s-master pvc]# kubectl scale statefulset web --replicas 3
statefulset.apps/web scaled


#访问依然不变
[root@k8s-master pvc]# 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、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>
相关推荐
嚯——哈哈3 分钟前
从入门到精通:解析如何使用亚马逊云服务器(AWS EC2)
运维·服务器·aws
编程修仙37 分钟前
Collections工具类
linux·windows·python
芝麻团坚果1 小时前
对subprocess启动的子进程使用VSCode python debugger
linux·ide·python·subprocess·vscode debugger
Elastic 中国社区官方博客1 小时前
Elasticsearch 中的热点以及如何使用 AutoOps 解决它们
大数据·运维·elasticsearch·搜索引擎·全文检索
写点什么啦1 小时前
[debug]不同的window连接ubuntu的vscode后无法正常加载kernel
linux·vscode·ubuntu·debug
wellnw1 小时前
[ubuntu]编译共享内存读取出现read.c:(.text+0x1a): undefined reference to `shm_open‘问题解决方案
linux·ubuntu
不爱学习的YY酱1 小时前
【操作系统不挂科】<CPU调度(13)>选择题(带答案与解析)
java·linux·前端·算法·操作系统
DC_BLOG1 小时前
Linux-Nginx虚拟主机
linux·运维·nginx
坐公交也用券1 小时前
使用Python3实现Gitee码云自动化发布
运维·gitee·自动化
一名路过的小码农1 小时前
ceph 18.2.4二次开发,docker镜像制作
ceph·docker·容器