Zookeeper简介

Posted

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Zookeeper简介相关的知识,希望对你有一定的参考价值。

zookeeper 是一个分布式协调系统

下载链接:http://mirror.cogentco.com/pub/apache/zookeeper/


1.zookeeper集群结构

(1) leader

是zk集群的主节点,客户端向zk注册数据时,都要通过leader来对整个集群中的所有从节点做数据同步

(2)follower

是zk集群的从节点,保留数据,接收leader的请求,参与leader的选举

(3)observer

是zk集群的从节点,保留数据,接收leader的请求,不参与leader的选举


2.zookeeper的选举机制

(1)集群启动的时候

server1启动,先查看集群中有没有leader,如果没有,则选举自己为leader;

server2启动,先查看集群中有没有leader,如果没有,则选举自己为leader,

第二轮投票,server1j和server2投id较大为leader, server2当选为leader;

(2)集群运行的时候

运行中,如果leader宕机,剩余的机器会进入选举状态,重新选举的依据:

优先考虑节点所持有的数据的版本号,最新的作为leader;

如果每个节点的数据都一样新,那就选举id大的为leader;


3.安装配置

解压安装

[[email protected] zookeeper-3.4.10]# tar -zxvf zookeeper-3.4.10.tar.gz

[[email protected] zookeeper-3.4.10]# mkdir -p   /usr/local/zookeeper-3.4.10/data

[[email protected] zookeeper-3.4.10]# echo 1 > data/myid

修改配置文件

其中server.1中的1对应文件myid内容,每个zookeeper都要有myid文件

[[email protected] zookeeper-3.4.10]# cd /usr/local/zookeeper-3.4.10/conf/

[[email protected] conf]# cp zoo_sample.cfg zoo.cfg

[[email protected] conf]# vim zoo.cfg

tickTime=2000

initLimit=10

syncLimit=5

dataDir=/usr/local/zookeeper-3.4.10/data

clientPort=2181

server.1=Darren2:2888:3888  

server.2=Darren3:2888:3888

server.3=Darren4:2888:3888


#启动zookeeper

[[email protected] bin]# ./zkServer.sh start

ZooKeeper JMX enabled by default

Using config: /usr/local/zookeeper-3.4.10/bin/../conf/zoo.cfg

Starting zookeeper ... STARTED


[[email protected] bin]# ./zkServer.sh status

ZooKeeper JMX enabled by default

Using config: /usr/local/zookeeper-3.4.10/bin/../conf/zoo.cfg

Mode: Leader


[[email protected] bin]# ./zkServer.sh status

ZooKeeper JMX enabled by default

Using config: /usr/local/zookeeper-3.4.10/bin/../conf/zoo.cfg

Mode: Follower


4.zookeeper命令

zookeeper中数据本质是key-value,zookeeper中一个数据被称为一个znode,一个znode不能太大,通常在10K以内,官方要求最大不要超过1M,如果太大,会导致zookeeper集群个节点的数据无法实时同步 ,数据保持一致性。

key使用路径表示,如:/dir1   value

znode类型

(1) persistent:默认的节点类型,一旦创建就会一直存在,除非手动删除数据;

(2)ephemeral:短暂节点,创建该数据节点的客户端如果和zk服务失去联系,该数据节点会被zk服务自动删除;

(3)sequential:带自增序号的节点,在同一个节点下创建sequential子节点,zk会给子节点名字自动拼接一个自增的序列号;

#登录zk client

[[email protected] bin]# ./zkCli.sh

[zk: localhost:2181(CONNECTED) 0] help

ZooKeeper -server host:port cmd args

    stat path [watch]

    set path data [version]

    ls path [watch]

    delquota [-n|-b] path

    ls2 path [watch]

    setAcl path acl

    setquota -n|-b val path

    history

    redo cmdno

    printwatches on|off

    delete path [version]

    sync path

    listquota path

    rmr path

    get path [watch]

    create [-s] [-e] path data acl

    addauth scheme auth

    quit

    getAcl path

    close

    connect host:port


[zk: localhost:2181(CONNECTED) 13] create /dir1 a1 

[zk: localhost:2181(CONNECTED) 16] ls /

[dir1, zookeeper]


[zk: localhost:2181(CONNECTED) 15] get /dir1

a1

cZxid = 0x7

ctime = Sun Nov 26 13:05:34 CST 2017

mZxid = 0x7

mtime = Sun Nov 26 13:05:34 CST 2017

pZxid = 0x7

cversion = 0

dataVersion = 0

aclVersion = 0

ephemeralOwner = 0x0

dataLength = 2

numChildren = 0


#创建短暂的znode,一旦quit客户端,则自动删除

[zk: localhost:2181(CONNECTED) 17] create -e /dir2 b1


#创建序列znode

[zk: localhost:2181(CONNECTED) 1] create -s /dir3 c1

Created /dir30000000003

[zk: localhost:2181(CONNECTED) 2] ls /

[dir1, zookeeper, dir30000000003]

[zk: localhost:2181(CONNECTED) 3] create -s /dir3 c1

Created /dir30000000004

[zk: localhost:2181(CONNECTED) 4] ls /

[dir1, zookeeper, dir30000000003, dir30000000004]


#创建短暂序列znode

[zk: localhost:2181(CONNECTED) 5] create -s -e /dir3 c1


#事件监听

[zk: localhost:2181(CONNECTED) 4] ls /dir1 watch

WATCHER::

WatchedEvent state:SyncConnected type:NodeChildrenChanged path:/dir1


5.zookeeper的客户端api的基本使用

package zkdemo1;

import org.apache.zookeeper.ZooKeeper;

public class TestConnection {

       public static void main(String[] args) throws Exception{

             ZooKeeper zk = new ZooKeeper("192.168.163.102", 2000, null);

             byte[] data = zk.getData("/dir1", false, null);

             System.out.println(new String(data));

             zk.close();

       

       }

}

输出:

a1


以上是关于Zookeeper简介的主要内容,如果未能解决你的问题,请参考以下文章

Zookeeper简介

分布式专题——Zookeeper简介

一zookeeper简介

zookeeper系列之:zookeeper简介浅谈

ZooKeeper简介

Zookeeper简介