写在前面
1. 创建PV和PVC
这里我使用的是K8S中自带的PV，没有使用动态存储，因此我用的是nfs存储，同时集群的数据都存在了一个目录下，如果你的需求不一样请注意
--- 1. 安装nfs 服务端
[root@k8s-master ~]# yum install -y rpcbind nfs-utils
## 启动rpcbind
[root@k8s-master ~]# systemctl enable --now rpcbind
## 启动nfs
[root@k8s-master ~]# systemctl enable --now nfs
## 配置
[root@k8s-master ~]# cat /etc/exports
/data/nfs/zookeeper/ 192.168.0.0/24(rw) ##共享/data/nfs/zookeeper/目录给192.168 的网段并赋予读写权限
##目录与权限
[root@k8s-master ~]# chown -R nfsnobody.nfsnobody /data/
[root@k8s-master ~]# chown -R 1000:1000 /data/nfs/
[root@k8s-master ~]# chmod -R 755 /data/nfs/
[root@k8s-master ~]# systemctl reload nfs
##挂载存储测试
mount -t nfs 192.168.0.160:/data/ /mnt/
##客户端安装
[root@k8s-master ~]#yum install -y nfs-utils
--- 2. 创建PV
[root@k8s-master /manifests/pv]# cat zk-pv.yaml
apiVersion: v1
kind: PersistentVolume
metadata:
name: zookeeper-pv
namespace: zookeeper
spec:
accessModes:
- ReadWriteMany
nfs:
server: 192.168.0.160
path: /data/nfs/zookeeper
persistentVolumeReclaimPolicy: Retain
capacity:
storage: 2Gi
--- 3. 创建PVC
[root@k8s-master /manifests/pv]# cat zk-pvc.yaml
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: zookeeper-pvc
namespace: zookeeper
spec:
## 声明资源的访问模式
accessModes:
- ReadWriteMany
## 声明资源的使用量
resources:
requests:
storage: 2Gi
## 绑定上面的pv
volumeName: zookeeper-pv2. 编写 Kubernetes 资源清单
2.1 ConfigMap 用于挂载配置文件
## 这里的主要挂在的是zoo.cfg的配置文件,如果有需要请自行添加
[root@k8s-master /manifests]# cat configmap/zoo.cfg
apiVersion: v1
kind: ConfigMap
metadata:
name: zk-config
namespace: zookeeper
data:
zoo.cfg: |
tickTime=2000
dataDir=/data
clientPort=2181
initLimit=10
syncLimit=5
server.1=zk-0.zk-headless:2888:3888
server.2=zk-1.zk-headless:2888:3888
server.3=zk-2.zk-headless:2888:38882.2 Headless Service 用于节点间互联
## 根据官方推荐使用Headless Service
[root@k8s-master /manifests]# cat service/zookeeper.service
apiVersion: v1
kind: Service
metadata:
name: zk-headless
namespace: zookeeper
labels:
app: zookeeper
spec:
selector:
app: zookeeper
ports:
- port: 2181
name: client
- port: 2888
name: follower
- port: 3888
name: election2.3 StatefulSet 用于部署有状态的 Zookeeper 节点
[root@k8s-master /manifests]# cat StatefulSet/zookeeper.yaml
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: zk
namespace: zookeeper
spec:
serviceName: zk-headless
replicas: 3
selector:
matchLabels:
app: zookeeper
template:
metadata:
labels:
app: zookeeper
spec:
securityContext:
runAsUser: 1000
fsGroup: 1000
initContainers:
- name: set-myid
image: busybox:1.28
command:
- sh
- -c
- |
idx=${HOSTNAME##*-}
echo $((idx + 1)) > /data/myid
volumeMounts:
- name: datadir
mountPath: /data
containers:
- name: zookeeper
image: 192.168.0.77:32237/k8s/zookeeper:latest
ports:
- containerPort: 2181
name: client
- containerPort: 2888
name: follower
- containerPort: 3888
name: election
env:
- name: ZOO_MY_ID
valueFrom:
fieldRef:
fieldPath: metadata.name
volumeMounts:
- name: datadir
mountPath: /data
- name: config
mountPath: /conf/zoo.cfg
subPath: zoo.cfg
volumes:
- name: config
configMap:
name: zk-config
- name: datadir
persistentVolumeClaim:
## 申明应用的PVC
claimName: zookeeper-pvc2.4 部署Zookeeper 集群
--- 1. 创建名称空间
[root@k8s-master /manifests]# kubectl create namespace zookeeper
--- 2. 部署 ConfigMap
[root@k8s-master /manifests]# kubectl apply -f configmap/zoo.cfg
--- 3. 部署Service
[root@k8s-master /manifests]# kubectl apply -f service/zookeeper.yaml
--- 4. 部署StatefulSet
[root@k8s-master /manifests]# kubectl apply -f StatefulSet/zookeeper.yaml
--- 5. 查看状态
[root@k8s-master /manifests]# kubectl get pod -n zookeeper 2.5 进入 Pod 测试客户端端口
## 进入客户端
[root@k8s-master /manifests]# kubectl exec -it zk-0 -n zookeeper -- zkCli.sh -server 127.0.0.1:2181
[zk: 127.0.0.1:2181(CONNECTED) 0] ls /
[zookeeper]
[zk: 127.0.0.1:2181(CONNECTED) 1] create /test "hello"
Created /test
[zk: 127.0.0.1:2181(CONNECTED) 2] get /test
hello
[zk: 127.0.0.1:2181(CONNECTED) 3] delete /test[zk: 127.0.0.1:2181(CONNECTED) 3] delete /test
## 查看集群状态
[root@k8s-master /manifests]# kubectl exec -it zk-0 -n zookeeper -- zkServer.sh status
[root@k8s-master /manifests]# kubectl exec -it zk-1 -n zookeeper -- zkServer.sh status
[root@k8s-master /manifests]# kubectl exec -it zk-2 -n zookeeper -- zkServer.sh status
Mode: leader
Mode: follower
Mode: follower
## 说明集群已经选出 Leader，其他是 Follower
## 使用nslookup来验证DNS
[root@k8s-master ~]# kubectl run busybox --image busybox:1.28 --restart=Never --rm -it busybox -n zookeeper -- sh
If you don't see a command prompt, try pressing enter.
/ # nslookup zk-0.zk-headless
Server: 10.200.0.10
Address 1: 10.200.0.10 kube-dns.kube-system.svc.xingzhibang.top
Name: zk-0.zk-headless
Address 1: 10.100.135.151 zk-0.zk-headless.zookeeper.svc.xingzhibang.top3. Zookeeper的使用
3.1 基本概念
ZooKeeper 是一个开源的 分布式协调服务，主要用于分布式系统中管理配置、同步状态、命名服务、选主等问题。它的核心价值是 一致性、可靠性和高可用性。
它不像数据库那样主要存储数据，而是提供一个 轻量级的数据结构（类似树的文件系统）和分布式锁/通知机制，方便多个节点协同工作。
3.2 未授权漏洞修复以及Zookeeper密码的修改
--- 1. 修改配置文件
vim zoo.cfg
## 添加参数
skipAcl=true
--- 2. 重启zk 保证配置生效
--- 3. 使用客户端进入,添加账号和密码
zkCli.sh
## 这里的zookeeper 是你的账号 Cosmic@2514 是密码
ddauth digest zookeeper:Cosmic@2514
setAcl -R / auth:zookeeper:Cosmic@2514:cdrwa
getAcl /
评论