我们基于Solr4.10.3版本进行安装配置SolrCloud集群,通过实践来实现索引数据的分布存储和检索。

1、准备工作

1.服务器

solrcloud01 192.168.137.171

solrcloud02 192.168.137.172

solrcloud03 192.168.137.173

2.资源文件

上传到/home/test

apache-solr-dataimportscheduler.jar 数据导入定时器

IK Analyzer 2012FF_hf1.zip 中文分词器

solr-4.10.3.tgz.tar

zookeeper-3.4.5.tar.gz

mysql-connector-java-3.1.13-bin.jar

以上资源可以从百度网盘获取。

链接:http://pan.baidu.com/s/1mhNbTxA 密码:8xnl

3.安装目录准备

mkdir -p /home/solr/cloud

cd /home/solr/cloud

mkdir lib multicore data

2、zookeeper集群搭建

在3台服务器中搭建zookeeper服务。

    进行解压:        tar -zxvf zookeeper-3.4.5.tar.gz 
    重命名:        mv zookeeper-3.4.5 /usr/local/zookeeper
    修改环境变量:    vi /etc/profile
                    export ZOOKEEPER_HOME=/usr/local/zookeeper
                    export PATH=.:$ZOOKEEPER_HOME/bin:$JAVA_HOME/bin:$PATH
    刷新:          source /etc/profile
    到zookeeper下修改配置文件  
                    cd /usr/local/zookeeper/conf
                    mv zoo_sample.cfg zoo.cfg
    修改conf:     vi zoo.cfg  修改两处
                    dataDir=/usr/local/zookeeper/data            
                    最后面添加 
                            server.0=bhz:2888:3888
                        server.1=hadoop1:2888:3888
                        server.2=hadoop2:2888:3888
    服务器标识配置:cd /usr/local/zookeeper/
                    创建文件夹:mkdir data
                    cd data
                    创建文件myid并填写内容为0:vi myid (内容为服务器标识 : 0)
                    其他两个服务器的值修改为1和2            
    启动zookeeper:
      路径:/usr/local/zookeeper/bin
     执行:zkServer.sh start (注意这里3台机器都要进行启动)
     状态:zkServer.sh status(在三个节点上检验zk的mode,一个leader和俩个follower)
     停止:zkServer.sh stop
     启动验证:输入jps,存在 QuorumPeerMain 进程的话,就说明 Zookeeper 启动成功了。

3、SolrCloud集群搭建

首先在一个节点上对SOLR进行配置,我们选择192.168.137.171节点。
1.SOLR基本配置

(1)解压solr tar文件并把solr.war文件copy到tomcat的webapps下面

cd /home/test

tar -zxvf solr-4.10.3.tgz.tar -C /usr/local

cp /usr/local/solr-4.10.3/example/webapps/solr.war /home/tomcat6/webapps/

cd /home/tomcat6/webapps/ && mkdir solr-cloud && unzip solr.war -d solr-cloud && rm -rf solr.war

(2)copy日志关联jar到solr工程的lib下面

cp /usr/local/solr-4.10.3/example/lib/ext/*.jar /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/
cp /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/*.jar /home/solr/cloud/lib/

(3)增加log配置文件

mkdir -p /home/tomcat6/webapps/solr-cloud/WEB-INF/classes
cp /usr/local/solr-4.10.3/example/resources/log4j.properties /home/tomcat6/webapps/solr-cloud/WEB-INF/classes/

(4)修改编码

vim/home/tomcat6/conf/server.xml里面加上编码设置

 

(5)slor core配置文件

 

首先建立关联目录

cd /home/solr/cloud/multicore && mkdir collection1 && touch solr.xml zoo.cfg && cd collection1 && mkdir conf data && cd conf && touch schema.xml solrconfig.xml solrcore.properties

其次写好配置文件

vim/home/solr/cloud/multicore/collection1/conf/schema.xml

<?xml version="1.0" ?>

<schema name="example core two" version="1.1">
    <field name="id" type="string" indexed="true" stored="true" multiValued="false" required="true" />
    <field name="stu_name" type="text_ik" indexed="true"  stored="true"  multiValued="false" /> 
    <field name="stu_sex"  type="int" indexed="true"  stored="true"  multiValued="false" /> 
    <field name="stu_address" type="text_ik" indexed="true"  stored="true"  multiValued="false" />
    <field name="_version_" type="long" indexed="true" stored="true" />
    
    <fieldType name="int" class="solr.TrieIntField" precisionStep="0" positionIncrementGap="0"/>
<fieldType name="long" class="solr.TrieLongField" precisionStep="0" positionIncrementGap="0"/>
<fieldType name="string" class="solr.StrField" sortMissingLast="true" />
    <fieldType name="text_ik" class="solr.TextField">
          <analyzer type="index" class="org.wltea.analyzer.lucene.IKAnalyzer" isMaxWordLength="false"/>  
          <analyzer type="query" class="org.wltea.analyzer.lucene.IKAnalyzer" isMaxWordLength="true"/> 
    </fieldType>

    <uniqueKey>id</uniqueKey>
    <defaultSearchField>stu_name</defaultSearchField>
    <solrQueryParser defaultOperator="OR" />
</schema>

vim/home/solr/cloud/multicore/collection1/conf/solrconfig.xml

<?xml version="1.0" encoding="UTF-8" ?>

<config>
  <luceneMatchVersion>4.10.3</luceneMatchVersion>

  <directoryFactory name="DirectoryFactory" class="${solr.directoryFactory:solr.StandardDirectoryFactory}"/>

  <dataDir>${solr.shard.data.dir:}</dataDir>

  <schemaFactory class="ClassicIndexSchemaFactory"/>

  <updateHandler class="solr.DirectUpdateHandler2">
    <updateLog>
      <str name="dir">${solr.shard.data.dir:}</str>
    </updateLog>
  </updateHandler>
    <requestHandler name="/update/csv" class="solr.CSVRequestHandler" startup="lazy">
    <lst name="defaults">
       <str name="separator">,</str>
       <str name="header">true</str>
       <str name="encapsulator">"</str>
    </lst>
    </requestHandler>

  <!-- realtime get handler, guaranteed to return the latest stored fields 
    of any document, without the need to commit or open a new searcher. The current 
    implementation relies on the updateLog feature being enabled. -->
  <requestHandler name="/get" class="solr.RealTimeGetHandler">
    <lst name="defaults">
      <str name="omitHeader">true</str>
    </lst>
  </requestHandler>  
  
  <requestHandler name="/replication" class="solr.ReplicationHandler" startup="lazy" /> 

  <requestDispatcher handleSelect="true" >
    <requestParsers enableRemoteStreaming="false" multipartUploadLimitInKB="2048" formdataUploadLimitInKB="2048" />
  </requestDispatcher>
  
  <requestHandler name="standard" class="solr.StandardRequestHandler" default="true" />
  <requestHandler name="/analysis/field" startup="lazy" class="solr.FieldAnalysisRequestHandler" />
  <requestHandler name="/update" class="solr.UpdateRequestHandler"  />
  <requestHandler name="/admin/" class="org.apache.solr.handler.admin.AdminHandlers" />
     
  <requestHandler name="/admin/ping" class="solr.PingRequestHandler">
    <lst name="invariants">
      <str name="q">solrpingquery</str>
    </lst>
    <lst name="defaults">
      <str name="echoParams">all</str>
    </lst>
  </requestHandler>
   <requestHandler name="/dataimport" class="org.apache.solr.handler.dataimport.DataImportHandler">
    <lst name="defaults">
        <str name="config">data-config.xml</str>
    </lst>
    </requestHandler>

   <query>
    <maxBooleanClauses>1024</maxBooleanClauses>
    <filterCache class="solr.FastLRUCache"
                 size="512"
                 initialSize="512"
                 autowarmCount="0"/>

    <queryResultCache class="solr.LRUCache"
                     size="512"
                     initialSize="512"
                     autowarmCount="0"/>
   
    <documentCache class="solr.LRUCache"
                   size="512"
                   initialSize="512"
                   autowarmCount="0"/>
    
    <!-- custom cache currently used by block join --> 
    <cache name="perSegFilter"
      class="solr.search.LRUCache"
      size="10"
      initialSize="0"
      autowarmCount="10"
      regenerator="solr.NoOpRegenerator" />

    <enableLazyFieldLoading>true</enableLazyFieldLoading>

   <queryResultWindowSize>20</queryResultWindowSize>

   <queryResultMaxDocsCached>200</queryResultMaxDocsCached>

    <listener event="newSearcher" class="solr.QuerySenderListener">
      <arr name="queries">
        
      </arr>
    </listener>
    <listener event="firstSearcher" class="solr.QuerySenderListener">
      <arr name="queries">
        <lst>
          <str name="q">static firstSearcher warming in solrconfig.xml</str>
        </lst>
      </arr>
    </listener>

    
    <useColdSearcher>false</useColdSearcher>

    <maxWarmingSearchers>2</maxWarmingSearchers>

  </query>
  <!-- config for the admin interface --> 
  <admin>
    <defaultQuery>solr</defaultQuery>
  </admin>

</config>

vim/home/solr/cloud/multicore/collection1/conf/solrcore.properties

solr.shard.data.dir=/home/solr/cloud/data

属性solr.shard.data.dir在solrconfig.xml文件中被引用过,指定索引数据的存放位置。

vim/home/solr/cloud/multicore/solr.xml

该文件中指定了ZooKeeper的相关配置,已经Solr Core的配置内容:

<?xml version="1.0" encoding="UTF-8" ?>

<solr persistent="true">
    <cores defaultCoreName="collection1" host="${host:}" adminPath="/admin/cores" zkClientTimeout="${zkClientTimeout:15000}" hostPort="8080" hostContext="${hostContext:solr-cloud}">
    </cores>
</solr>

注意:这里,我们并没有配置任何的core元素,这个等到整个配置安装完成之后,通过SOLR提供的REST接口,来实现Collection以及Shard的创建,从而来更新这些配置文件。

vim/home/solr/cloud/multicore/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.
# dataDir=/opt/zookeeper/data
# NOTE: Solr defaults the dataDir to <solrHome>/zoo_data

# the port at which the clients will connect
# clientPort=2181
# NOTE: Solr sets this based on zkRun / zkHost params

(6)DIH全量增量从Mysql数据库导入数据的配置

cp/usr/local/solr-4.10.3/dist/solr-dataimporthandler-4.10.3.jar /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/

cp/home/test/mysql-connector-java-3.1.13-bin.jar /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/

vim/home/solr/cloud/multicore/collection1/conf/data-config.xml

<?xml version="1.0" encoding="UTF-8"?>
<dataConfig>
    <dataSource type="JdbcDataSource" driver="com.mysql.jdbc.Driver" url="jdbc:mysql://192.168.137.171:3306/solr" user="root" password="root" batchSize="-1" />
    <document name="testDoc">
        <entity name="student" pk="id"
                query="select * from student where isDeleted=0"
                deltaImportQuery="select * from student where id='${dih.delta.id}'"
                deltaQuery="select id from student where updateTime> '${dataimporter.last_index_time}' and isDeleted=0"
                deletedPkQuery="select id from student where isDeleted=1">
            <field column="id" name="id"/>
            <field column="stu_name" name="stu_name"/>
            <field column="stu_sex" name="stu_sex"/>
            <field column="stu_address" name="stu_address"/>
        </entity>
    </document>
</dataConfig>

cp/home/test/apache-solr-dataimportscheduler.jar /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/

vim/home/tomcat6/webapps/solr-cloud/WEB-INF/web.xml
元素添加一个子元素

<listener>
        <listener-class>
    org.apache.solr.handler.dataimport.scheduler.ApplicationListener
        </listener-class>
</listener>

mkdir –p /home/solr/cloud/multicore/conf

vim/home/solr/cloud/multicore/conf/dataimport.properties

下面是最终我的自动定时更新配置文件内容:syncCores需要在各服务器单独配置

#################################################
#                                              
#       dataimport scheduler properties        
#                                              
#################################################

#  to sync or not to sync
#  1 - active; anything else - inactive
syncEnabled=1

#  which cores to schedule
#  in a multi-core environment you can decide which cores you want syncronized
#  leave empty or comment it out if using single-core deployment
#  syncCores=game,resource
syncCores=myc_shard2_replica1

#  solr server name or IP address
#  [defaults to localhost if empty]
server=

#  solr server port
#  [defaults to 80 if empty]
port=8080

#  application name/context
#  [defaults to current ServletContextListener's context (app) name]
webapp=solr-cloud

#  URLparams [mandatory]
#  remainder of URL
#http://localhost:8983/solr/collection1/dataimport?command=delta-import&clean=false&commit=true
params=/dataimport?command=delta-import&clean=false&commit=true

#  schedule interval
#  number of minutes between two runs
#  [defaults to 30 if empty]
interval=1

#  重做索引的时间间隔,单位分钟,默认7200,即1天; 
#  为空,为0,或者注释掉:表示永不重做索引
reBuildIndexInterval=0

#  重做索引的参数
reBuildIndexParams=/dataimport?command=full-import&clean=true&commit=true

#  重做索引时间间隔的计时开始时间,第一次真正执行的时间=reBuildIndexBeginTime+reBuildIndexInterval*60*1000;
#  两种格式:2012-04-11 03:10:00 或者  03:10:00,后一种会自动补全日期部分为服务启动时的日期
reBuildIndexBeginTime=03:10:00

(7)中文分词器关联配置

cd/home/test

unzip IK\ Analyzer\ 2012FF_hf1.zip -d IK

cd/home/test/IK

cpIKAnalyzer2012FF_u1.jar /home/tomcat6/webapps/solr-cloud/WEB-INF/lib/

cpIKAnalyzer.cfg.xml stopword.dic /home/tomcat6/webapps/solr-cloud/WEB-INF/classes/

2.ZooKeeper管理监控配置文件
SolrCloud是通过ZooKeeper集群来保证配置文件的变更及时同步到各个节点上,所以,需要将配置文件上传到ZooKeeper集群中:

java -classpath .:/home/solr/cloud/lib/* org.apache.solr.cloud.ZkCLI -cmd upconfig -zkhost 192.168.137.171:2181,192.168.137.172:2181,192.168.137.173:2181 -confdir /home/solr/cloud/multicore/collection1/conf -confname myconf 
 
 
java -classpath .:/home/solr/cloud/lib/* org.apache.solr.cloud.ZkCLI -cmd linkconfig -collection collection1 -confname myconf -zkhost 192.168.137.171:2181,192.168.137.172:2181,192.168.137.173:2181

上传完成以后,我们检查一下ZooKeeper上的存储情况:

zkCli.sh -server 192.168.137.171:2181

 

3.Tomcat配置与启动

vim/home/tomcat6/bin/catalina.sh
在Tomcat的启动脚本bin/catalina.sh中,增加如下配置:

JAVA_OPTS="-server -Xmx128m -Xms128m -verbose:gc -Xloggc:solr_gc.log -Dsolr.solr.home=/home/solr/cloud/multicore -DzkHost=192.168.137.171:2181,192.168.137.172:2181,192.168.137.173:2181"
 
启动Tomcat服务器:
/home/tomcat6/bin/catalina.sh start
查看日志:
tail -100f /home/tomcat6/logs/catalina.out
我们查看一下ZooKeeper中的数据状态,如下所示:

zkCli.sh -server 192.168.137.171:2181


这时候,SolrCloud集群中只有一个活跃的节点,而且默认生成了一个collection1实例,这个实例实际上虚拟的,因为通过web界面无法访问http://192.168.137.171:8080/solr-cloud/,看不到任何有关SolrCloud的信息,如图所示:

 

4.同步数据和配置信息,启动其他节点
在另外两个节点上安装Tomcat和Solr服务器,只需要拷贝对应的目录即可:

另外两个节点rm -rf /home/tomcat6

scp -r /home/tomcat6/ root@192.168.137.172:/home/

scp -r /home/solr/ root@192.168.137.172:/home/

scp -r /home/tomcat6/ root@192.168.137.173:/home/

scp -r /home/solr/ root@192.168.137.173:/home/

启动其他Solr服务器节点:

/home/tomcat6/bin/catalina.sh start
 
查看日志:
tail -100f /home/tomcat6/logs/catalina.out

查看ZooKeeper集群中数据状态:

zkCli.sh -server 192.168.137.171:2181

 

这时,已经存在3个活跃的节点了,但是SolrCloud集群并没有更多信息,访问http://192.168.137.171:8080/solr-cloud/后,同上面的图是一样的,没有SolrCloud相关数据。

5.创建Collection、Shard和Replication

(1)创建Collection及初始Shard

直接通过REST接口来创建Collection,如下所示:

curl 'http://192.168.137.171:8080/solr-cloud/admin/collections?action=CREATE&name=mycollection&numShards=3&replicationFactor=1'

 

more /home/solr/cloud/multicore/solr.xml

 

 

(2)创建Replication

下面对已经创建的初始分片进行复制。 shard1已经在192.168.137.173上,我们复制分片到192.168.137.171和192.168.137.172上,执行如下命令:

curl 'http://192.168.137.171:8080/solr-cloud/admin/cores?action=CREATE&collection=mycollection&name=mycollection_shard1_replica_2&shard=shard1'

curl 'http://192.168.137.172:8080/solr-cloud/admin/cores?action=CREATE&collection=mycollection&name=mycollection_shard1_replica_3&shard=shard1'