solrcloud，tomcat，外部zookeeper配置详解

之前由于在项目中使用的solr3.6，服务器总是莫名其妙的挂掉，各种解决未果，于是尝试换到solr4，想试试solrcloud和zookeeper集群看看效果怎么样，实际效果比3.6稳定很多，下面说一下项目中我们的具体现在一个测试的配置。 我们配置3台zookper服务器和3台solr服务器，由于没有多台机器测试，zookeeper的集群都在一台服务器上，tomcat部署在3台服务器上。

zookeeper是什么

ZooKeeper 顾名思义 动物园管理员，他是拿来管大象(Hadoop) 、 蜜蜂(Hive) 、 小猪(Pig)  的管理员， Apache Hbase和 Apache Solr 以及LinkedIn sensei  等项目中都采用到了 Zookeeper。ZooKeeper是一个分布式的，开放源码的分布式应用程序协调服务，ZooKeeper是以Fast Paxos算法为基础，实现同步服务，配置维护和命名服务等分布式应用。

zookeeper安装及配置

zookeeper下载

http://www.apache.org/dyn/closer.cgi/zookeeper/

zookeeper配置

---------------

首先先把/etc/hosts下的主机名映射127.0.1.1改成127.0.0.1。

---------------

（1）解压为zookeeper

tar -xf -C /home/myuser/zookeeper/

复制zookeeper文件夹3份，分别重名名为zookeeperA，zookeeperB，zookeeperC。

并且创建数据快照以及日志存放文件夹，命名为zooA，zooB，zooC。

（2）编辑对应的zookeeper配置文件，复制zookeeperconf下zoo_sample.cfg为zoo.cfg

cd /home/myuser/zookeeperA/conf
cp zoo_sample.cfg zoo.cfg   

（3）修改zoo.cfg

# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial 
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between 
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just 
# example sakes.
dataDir=/home/myuser/zooA/data
# the port at which the clients will connect
clientPort=2181
# ZooKeeper server and its port no. # ZooKeeper ensemble should know about every other machine in the ensemble # specify server id by creating 'myid' file in the dataDir # use hostname instead of IP address for convenient maintenance 
server.1=127.0.0.1:2888:3888
server.2=127.0.0.1:2988:3988 
server.3=127.0.0.1:2088:3088
#
# Be sure to read the maintenance section of the 
# administrator guide before turning on autopurge.
#
# http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1
dataLogDir=/home/myuser/zooA/log

tickTime：心跳时间，为了确保连接存在的，以毫秒为单位，最小超时时间为两个心跳时间

initLimit：多少个心跳时间内，允许其他server连接并初始化数据，如果ZooKeeper管理的数据较大，则应相应增大这个值

clientPort：服务的监听端口

dataDir：用于存放内存数据库快照的文件夹，同时用于集群的myid文件也存在这个文件夹里（注意：一个配置文件只能包含一个dataDir字样，即使它被注释掉了。）

dataLogDir：用于单独设置transaction log的目录，transaction log分离可以避免和普通log还有快照的竞争

syncLimit：多少个tickTime内，允许follower同步，如果follower落后太多，则会被丢弃。
（4）创建myid文件

cd /home/myuser/zooA/data
sudo sh -c 'echo "1" >> myid'  

其他文件夹类似创建myid文件，zookeeperB为2，zookeeperC为3

（5）启动zookeeper

cd /home/myuser/zookeeperA/bin
sudo sh  zkServer.sh start

查看zookeeper状态

[root@weibo bin]# sh zkServer.sh status
JMX enabled by default
Using config: /home/weibo/zookeeperA/bin/../conf/zoo.cfg
Mode: follower

启动OK，依次启动另外两台zookeeper，启动第一台zookeeper后，你可以观察bin下的zookeeper.out可以看到报错，connection refused，没有关系，zookeeper需要等待其他另个节点的加入，全部启动之后就正常了。

（6）客户端连接zookeeper

[root@weibo bin]# sh zkCli.sh 
Connecting to localhost:2181
2013-05-10 15:00:25,363 [myid:] - INFO  [main:Environment@100] - Client environment:zookeeper.version=3.4.5-1392090, built on 09/30/2012 17:52 GMT
2013-05-10 15:00:25,369 [myid:] - INFO  [main:Environment@100] - Client environment:host.name=localhost
2013-05-10 15:00:25,370 [myid:] - INFO  [main:Environment@100] - Client environment:java.version=1.6.0_33
2013-05-10 15:00:25,371 [myid:] - INFO  [main:Environment@100] - Client environment:java.vendor=Sun Microsystems Inc.
2013-05-10 15:00:25,372 [myid:] - INFO  [main:Environment@100] - Client environment:java.home=/home/weibo/jdk1.6.0_33/jre
2013-05-10 15:00:25,373 [myid:] - INFO  [main:Environment@100] - Client environment:java.class.path=/home/weibo/zookeeperA/bin/../build/classes:/home/weibo/zookeeperA/bin/../build/lib/*.jar:/home/weibo/zookeeperA/bin/../lib/slf4j-log4j12-1.6.1.jar:/home/weibo/zookeeperA/bin/../lib/slf4j-api-1.6.1.jar:/home/weibo/zookeeperA/bin/../lib/netty-3.2.2.Final.jar:/home/weibo/zookeeperA/bin/../lib/log4j-1.2.15.jar:/home/weibo/zookeeperA/bin/../lib/jline-0.9.94.jar:/home/weibo/zookeeperA/bin/../zookeeper-3.4.5.jar:/home/weibo/zookeeperA/bin/../src/java/lib/*.jar:/home/weibo/zookeeperA/bin/../conf:/home/weibo/jdk1.6.0_33/lib:.
2013-05-10 15:00:25,373 [myid:] - INFO  [main:Environment@100] - Client environment:java.library.path=/home/weibo/jdk1.6.0_33/jre/lib/amd64/server:/home/weibo/jdk1.6.0_33/jre/lib/amd64:/home/weibo/jdk1.6.0_33/jre/../lib/amd64:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib
2013-05-10 15:00:25,374 [myid:] - INFO  [main:Environment@100] - Client environment:java.io.tmpdir=/tmp
2013-05-10 15:00:25,375 [myid:] - INFO  [main:Environment@100] - Client environment:java.compiler=
2013-05-10 15:00:25,375 [myid:] - INFO  [main:Environment@100] - Client environment:os.name=Linux
2013-05-10 15:00:25,376 [myid:] - INFO  [main:Environment@100] - Client environment:os.arch=amd64
2013-05-10 15:00:25,377 [myid:] - INFO  [main:Environment@100] - Client environment:os.version=2.6.32-279.5.2.el6.x86_64
2013-05-10 15:00:25,377 [myid:] - INFO  [main:Environment@100] - Client environment:user.name=root
2013-05-10 15:00:25,378 [myid:] - INFO  [main:Environment@100] - Client environment:user.home=/root
2013-05-10 15:00:25,379 [myid:] - INFO  [main:Environment@100] - Client environment:user.dir=/home/weibo/zookeeperA/bin
2013-05-10 15:00:25,382 [myid:] - INFO  [main:ZooKeeper@438] - Initiating client connection, connectString=localhost:2181 sessionTimeout=30000 watcher=org.apache.zookeeper.ZooKeeperMain$MyWatcher@9d04653
Welcome to ZooKeeper!
2013-05-10 15:00:25,604 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@966] - Opening socket connection to server localhost/127.0.0.1:2181. Will not attempt to authenticate using SASL (无法定位登录配置)
JLine support is enabled
2013-05-10 15:00:25,649 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@849] - Socket connection established to localhost/127.0.0.1:2181, initiating session
2013-05-10 15:00:25,842 [myid:] - INFO  [main-SendThread(localhost:2181):ClientCnxn$SendThread@1207] - Session establishment complete on server localhost/127.0.0.1:2181, sessionid = 0x13e7e067764005d, negotiated timeout = 30000

WATCHER::

WatchedEvent state:SyncConnected type:None path:null
[zk: localhost:2181(CONNECTED) 0] ls /
[configs, zookeeper, clusterstate.json, aliases.json, live_nodes, overseer, overseer_elect, collections]
[zk: localhost:2181(CONNECTED) 1] 

configs：保存上传的配置文件信息

clusterstate.json：集群状态json

aliases：别名json

live_node：当solr服务器启动的时候，会注册到这里

overseer：保存shard信息

overseer_elect：节点选举

collections：所有的collection

Solr及Tomcat安装配置

（1）下载solr4.2，把solr-4.2.1\example\solr\collection1\conf 目录下的配置文件上传的zookeeper

（2）上传solr配置文件到zookeeper

java -classpath .:/home/weibo/tomcat_solr_qq/webapps/solr/WEB-INF/lib/* org.apache.solr.cloud.ZkCLI -cmd upconfig -zkhost 127.0.0.1:2181,127.0.0.1:2281,127.0.0.1:2381 -confdir /home/weibo/solr-config -confname weiboconfig

（3）把配置文件和目标collection联系起来

java -classpath .:/home/weibo/tomcat_solr_qq/webapps/solr/WEB-INF/lib/* org.apache.solr.cloud.ZkCLI -cmd linkconfig -collection weibocollection -confname weiboconfig -zkhost 127.0.0.1:2181,127.0.0.1:2281,127.0.0.1:2381

（4）复制3个tomcat，在conf/Catalina/localhost创建solr.xml（如果目录不存在，请自行创建），内容如下

crossContext不能设置为true，这里注意

（5）在你的solr home目录创建solr.xml

其他tomcat对应设置对应的端口号

（6）编辑tomcat bin/catalina.sh（windows下catalina.bat），加入

JAVA_OPTS='-DzkHost=127.0.0.1:2181,127.0.0.1:2281,127.0.0.1:2381'

这里就是solr和zookeeper交互的地方了

（7）依次启动tomcat，创建core及shard，访问solr

----------------------------------这里由于为了测试简单，我只创建了一个主节点，没有replica节点，如果你需要，请跳过这一步，直接看下面的collection配置-----------------

http://localhost:8080/solr/admin/collections?action=CREATE&name=weibocollection&numShards=3&replicationFactor=1

Collection,Shard,Replication,Core配置

首先，我们创建一个collection，并且指明它的replica的个数，shard的个数

http://localhost:8080/solr/admin/collections?action=CREATE&name=weibocollection&numShards=3&replicationFactor=3&maxShardsPerNode=3

由于solr4.2会自动创建replication，所以这里不用手动去创建replication了已经。

Last

由于zookeeper是过半存活，所以由于我这里使用的是3台zookeeper，所以最大可以死掉1台，还能继续工作，官方推荐使用奇数台服务器。