应用介绍

zookeeper 3.4.5 安装包 windows、linux下安装是一样的，ZooKeeper是Hadoop的正式子项目，它是一个针对大型分布式系统的可靠协调系统，提供的功能包括：配置维护、名字服务、分布式同步、组服务等。ZooKeeper的目标就是封装好复杂易出错的关键服务，将简单易用的接口和性能高效、功能稳定的系统提供给用户。它也是Google的Chubby一个开源的实现，是高有效和可靠的协同工作系统，Zookeeper能够用来leader选举，配置信息维护等，在一个分布式的环境中，需要一个Master实例或存储一些配置信息，确保文件写入的一致性等。

下载溜溜体育(中国)官方网站:如何安装使用：

1，下载zookeeper-3.4.5安装包

2，解压：tar -zxvf zookeeper-3.4.5.tar.gz

3，配置环境变量：
#java环境变量
vim /etc/profile.d/java.sh
#set java environment
JAVA_HOME=/opt/modules/jdk
CLASSPATH=.:$JAVA_HOME/lib.tools.jar
PATH=$JAVA_HOME/bin:$PATH
#zookeeper环境变量
vim /etc/profile.d/zookeeper.sh
export ZOOKEEPERER_HOME=/opt/modules/zookeeper-3.4.5
export PATH=$ZOOKEEPER_HOME/bin:$PATH

4，主机名称到ip地址映射配置（这个步骤再安装hadoop的时候已经配置好了）
ZooKeeper集群中具有两个关键的角色：Leader和Follower。集群中所有的结点作为一个整体对分布式应用提供服务，集群中每个结点之间都互相连接，所以，在配置的ZooKeeper集群的时候，每一个结点的host到IP地址的映射都要配置上集群中其它结点的映射信息。
例如，我的ZooKeeper集群中每个结点的配置，以master为例，/etc/hosts内容如下所示：
192.168.209.134 master
192.168.209.135 slaver1
192.168.209.136 slaver2
ZooKeeper采用一种称为Leader election的选举算法。在整个集群运行过程中，只有一个Leader，其他的都是Follower，如果ZooKeeper集群在运行过程中 Leader出了问题，系统会采用该算法重新选出一个Leader。因此，各个结点之间要能够保证互相连接，必须配置上述映射。

5，修改/opt/modules/zookeeper-3.4.5/conf/zoo.cfg(mv zoo_sample.cfg zoo.cfg)

tickTime=2000
initLimit=5
syncLimit=2
dataDir=/home/hadoop/storage/zookeeper
clientPort=2181
server.1=master:2888:3888
server.2=slaver1:2888:3888
server.3=slaver2:2888:3888

6，传输到slaver
为了提高传输速度，我这里先将master上配置好的zookeeper进行压缩。
tar -zcvf zookeeper.tar zookeeper-3.4.5
[hadoop@master modules]$ scp zookeeper.tar hadoop@slaver1:/opt/modules/
zookeeper.tar                                                                                                                               100% 16MB  15.7MB/s 00:00
[hadoop@master modules]$ scp zookeeper.tar hadoop@slaver2:/opt/modules/
zookeeper.tar                                                                                                                               100% 16MB  15.7MB/s 00:00
[hadoop@master modules]$

7，登陆slaver解压zookeeper.tar
ssh slaver1
tar -xvf /opt/modules/zookeeper.tar
ctrl+d
ssh slaver2
tar -xvf /opt/modules/zookeeper.tar

8，设置myid
在我们配置的dataDir指定的目录下面，创建一个myid文件，里面内容为一个数字，用来标识当前主机，conf/zoo.cfg文件中配置的server.X中X为什么数字，则myid文件中就输入这个数字，例如：
[root@master ~]# su - hadoop
[hadoop@master ~]$ cd /opt/modules/zookeeper-3.4.5/conf/
[hadoop@master conf]$ vim zoo.cfg
[hadoop@master conf]$ echo "1" > /home/hadoop/storage/zookeeper/myid
-bash: /home/hadoop/storage/zookeeper/myid: No such file or directory
[hadoop@master conf]$ mkdir -p /home/hadoop/storage/zookeeper/
[hadoop@master conf]$ echo "1" > /home/hadoop/storage/zookeeper/myid
[hadoop@master conf]$ ssh slaver1
Last login: Sun Mar 23 23:18:50 2014 from master
[hadoop@slaver1 ~]$  mkdir -p /home/hadoop/storage/zookeeper/
[hadoop@slaver1 ~]$  echo "2" > /home/hadoop/storage/zookeeper/myid
[hadoop@slaver1 ~]$ logout
[hadoop@master conf]$ ssh slaver2
Last login: Sun Mar 23 23:19:25 2014 from master
[hadoop@slaver2 ~]$ mkdir -p /home/hadoop/storage/zookeeper/
[hadoop@slaver2 ~]$ echo "3" > /home/hadoop/storage/zookeeper/myid
9，启动zookeeper集群
在每个节点上执行启动任务的脚本：
[hadoop@master zookeeper-3.4.5]$ bin/zkServer.sh start
[hadoop@slaver1 zookeeper-3.4.5]$ bin/zkServer.sh start
[hadoop@slaver2 zookeeper-3.4.5]$ bin/zkServer.sh start
10，验证安装
[hadoop@master zookeeper-3.4.5]$ bin/zkServer.sh status
JMX enabled by default
Using config: /opt/modules/zookeeper-3.4.5/bin/../conf/zoo.cfg
Mode: follower
[hadoop@slaver1 zookeeper-3.4.5]$ bin/zkServer.sh status
JMX enabled by default
Using config: /opt/modules/zookeeper-3.4.5/bin/../conf/zoo.cfg
Mode: leader
[hadoop@slaver2 zookeeper-3.4.5]$ bin/zkServer.sh status
JMX enabled by default
Using config: /opt/modules/zookeeper-3.4.5/bin/../conf/zoo.cfg
Mode: follower

在master输入jps，显示如下：
4096 SecondaryNameNode
4181 JobTracker
4607 Jps
3919 NameNode
4381 QuorumPeerMain
在slaver1，slaver2输入jps，显示如下：
3539 Jps
3021 DataNode
3224 QuorumPeerMain
3127 TaskTracker

另外，可以通过客户端脚本，连接到ZooKeeper集群上。对于客户端来说，ZooKeeper是一个整体（ensemble），连接到ZooKeeper集群实际上感觉在独享整个集群的服务，所以，你可以在任何一个结点上建立到服务集群的连接，例如：
[hadoop@slaver1 zookeeper-3.4.5]$ bin/zkCli.sh -server master:2181
Connecting to master:2181
2014-03-24 00:24:22,321 [myid:] - INFO  [main:Environment@100] - Client environment:zookeeper.version=3.4.5-1392090, built on 09/30/2012 17:52 GMT
2014-03-24 00:24:22,325 [myid:] - INFO  [main:Environment@100] - Client environment:host.name=slaver1
2014-03-24 00:24:22,326 [myid:] - INFO  [main:Environment@100] - Client environment:java.version=1.6.0_45
2014-03-24 00:24:22,327 [myid:] - INFO  [main:Environment@100] - Client environment:java.vendor=Sun Microsystems Inc.
2014-03-24 00:24:22,327 [myid:] - INFO  [main:Environment@100] - Client environment:java.home=/opt/modules/jdk/jre
2014-03-24 00:24:22,328 [myid:] - INFO  [main:Environment@100] - Client environment:java.class.path=/opt/modules/zookeeper-3.4.5/bin/../build/classes:/opt/modules/zookeeper-3.4.5/bin/../build/lib/*.jar:/opt/modules/zookeeper-3.4.5/bin/../lib/slf4j-log4j12-1.6.1.jar:/opt/modules/zookeeper-3.4.5/bin/../lib/slf4j-api-1.6.1.jar:/opt/modules/zookeeper-3.4.5/bin/../lib/netty-3.2.2.Final.jar:/opt/modules/zookeeper-3.4.5/bin/../lib/log4j-1.2.15.jar:/opt/modules/zookeeper-3.4.5/bin/../lib/jline-0.9.94.jar:/opt/modules/zookeeper-3.4.5/bin/../zookeeper-3.4.5.jar:/opt/modules/zookeeper-3.4.5/bin/../src/java/lib/*.jar:/opt/modules/zookeeper-3.4.5/bin/../conf:.:/opt/modules/jdk/lib.tools.jar
2014-03-24 00:24:22,328 [myid:] - INFO  [main:Environment@100] - Client environment:java.library.path=/opt/modules/jdk/jre/lib/amd64/server:/opt/modules/jdk/jre/lib/amd64:/opt/modules/jdk/jre/../lib/amd64:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib
2014-03-24 00:24:22,335 [myid:] - INFO  [main:Environment@100] - Client environment:java.io.tmpdir=/tmp
2014-03-24 00:24:22,335 [myid:] - INFO  [main:Environment@100] - Client environment:java.compiler=<NA>
2014-03-24 00:24:22,336 [myid:] - INFO  [main:Environment@100] - Client environment: os.name=Linux
2014-03-24 00:24:22,336 [myid:] - INFO  [main:Environment@100] - Client environment: os.arch=amd64
2014-03-24 00:24:22,337 [myid:] - INFO  [main:Environment@100] - Client environment: os.version=2.6.32-358.el6.x86_64
2014-03-24 00:24:22,337 [myid:] - INFO  [main:Environment@100] - Client environment:user.name=hadoop
2014-03-24 00:24:22,338 [myid:] - INFO  [main:Environment@100] - Client environment:user.home=/home/hadoop
2014-03-24 00:24:22,338 [myid:] - INFO  [main:Environment@100] - Client environment:user.dir=/opt/modules/zookeeper-3.4.5
2014-03-24 00:24:22,343 [myid:] - INFO  [main:ZooKeeper@438] - Initiating client connection, connectString=master:2181 sessionTimeout=30000 watcher=org.apache.zookeeper.ZooKeeperMain$MyWatcher@24a37368
Welcome to ZooKeeper!
2014-03-24 00:24:22,404 [myid:] - INFO  [main-SendThread(master:2181):ClientCnxn$SendThread@966] - Opening socket connection to server master/192.168.209.134:2181. Will not attempt to authenticate using SASL (Unable to locate a login configuration)
2014-03-24 00:24:22,418 [myid:] - INFO  [main-SendThread(master:2181):ClientCnxn$SendThread@849] - Socket connection established to master/192.168.209.134:2181, initiating session
JLine support is enabled
2014-03-24 00:24:22,545 [myid:] - INFO  [main-SendThread(master:2181):ClientCnxn$SendThread@1207] - Session establishment complete on server master/192.168.209.134:2181, sessionid = 0x144f2dc538b0000, negotiated timeout = 30000

WATCHER::

WatchedEvent state:SyncConnected type:None path:null
[zk: master:2181(CONNECTED) 0]

到此，分布式的zookeeper集群就算安装好了。稍后开始安装hbase。