首页 > 代码库 > solrCloud 4.9 分布式集群部署及注意事项
solrCloud 4.9 分布式集群部署及注意事项
环境搭建
一.zookeeper
参考：http://blog.chinaunix.net/uid-25135004-id-4214399.html
现有4台机器 10.14.2.201 10.14.2.202 10.14.2.203 10.14.2.204安装zookeeper集群 在所有机器上进行1、下载安装包解压 tar xvf zookeeper-3.4.5.tar.gz -C /export/ cd /export/ ln -s zookeeper-3.4.5 zookeeper2、设置环境变量 vim /etc/profile#Set ZooKeeper Enviromentexport ZOOKEEPER_HOME=/export/zookeeperexport PATH=$PATH:$ZOOKEEPER_HOME/bin3、加载环境变量 source /etc/profile4、配置zookeepercd /export/zookeeper/confcp zoo_sample.cfg zoo.cfgvim zoo.cfg tickTime=2000initLimit=10syncLimit=5dataDir=/export/zookeeper/datadataLogDir=/export/zookeeper/logclientPort=2181server.1=10.14.2.201:2888:3888server.2=10.14.2.202:2888:3888server.3=10.14.2.203:2888:3888server.4=10.14.2.204:2888:38885、dataDir和日志目录dataLogDirmkdir -p /export/zookeeper/datamkdir -p /export/zookeeper/log6、在10.14.2.201上在dataDir目录下创建myid文件echo 1 > /export/zookeeper/data/myid myid中的数字与zoo.cfg中的"server.数字=ip或者域名:port1:port2"相对应，配置文件中10.14.2.201对应的server.数字是1 所以myid的数字应为17、在10.14.2.202上在dataDir目录下创建myid文件echo 2 > /export/zookeeper/data/myid 8、在10.14.2.203上在dataDir目录下创建myid文件echo 3 > /export/zookeeper/data/myid 9、在10.14.2.204上在dataDir目录下创建myid文件echo 4 > /export/zookeeper/data/myid 在所用机器上执行10、修改权限chown hadoop.hadoop -R /export/zookeeper/chown hadoop.hadoop -R /export/zookeeper11、启动zookeeper/export/zookeeper/bin/zkServer.sh start12、观察zookeeper状态 1> 10.14.2.201 # /export/zookeeper/bin/zkServer.sh status JMX enabled by default Using config: /export/zookeeper/bin/../conf/zoo.cfg Mode: follower 2> 10.14.2.202 # /export/zookeeper/bin/zkServer.sh status JMX enabled by default Using config: /export/zookeeper/bin/../conf/zoo.cfg Mode: follower 3> 10.14.2.203 # /export/zookeeper/bin/zkServer.sh status JMX enabled by default Using config: /export/zookeeper/bin/../conf/zoo.cfg Mode: leader 4> 10.14.2.204 # /export/zookeeper/bin/zkServer.sh status JMX enabled by default Using config: /export/zookeeper/bin/../conf/zoo.cfg Mode: follower 13、此时zookeeper已安装并启动完毕
二.solrCloud
参考：http://lucien-zzy.iteye.com/blog/2002463
1.安装tomcat，配置端口。部署solr.war，将solr\example\lib\ext下的jar拷贝到solr\WEB-INF\lib下。
2.创建
/home/solr/solrcloud
/home/solr/solrcloud/config-files (放置solr\example\solr\collection1\conf 下的所有文件)
/home/solr/solrcloud/solr-lib (放置solr\WEB-INF\lib下的所有jar包)
3.配置中文分词器
4.创建solr数据目录
/home/solr/solr-cores
生成solr.xml，如下：
<?xml version="1.0" encoding="UTF-8" ?><solr persistent="true"><logging enabled="true"><watcher size="100" threshold="INFO" /></logging><cores defaultCoreName="collection1" adminPath="/admin/cores" host="${host:}" hostPort="8080" hostContext="${hostContext:solr}" zkClientTimeout="${zkClientTimeout:15000}"></cores></solr>
5.创建
tomcat/conf/Catalina
tomcat/conf/Catalina/localhost
并创建solr.xml
<?xml version="1.0" encoding="UTF-8"?><Context docBase="/soft/apache-tomcat-7.0.37/webapps/solr" debug="0" crossContext="true"><Environment name="solr/home" type="java.lang.String" value=http://www.mamicode.com/"/home/solr/solr-cores" override="true"/></Context>
docBase根据实际情况配置；
solr/home为之前创建的目录：/home/solr/solr-cores
6.修改tomcat/bin/cataina.sh 文件，在最上方加入：JAVA_OPTS="-DzkHost=master:2181,slave1:2181,slave2:2181"，或直接使用ip。
将以上配置发布到集群其他机器上。
7.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。执行如下操作：
java -classpath .:/home/solr/solrcloud/solr-lib/* org.apache.solr.cloud.ZkCLI -cmd upconfig -zkhost 192.168.91.128:2181,192.168.91.129:2181,192.168.91.130:2181 -confdir /home/solr/solrcloud/config-files/ -confname myconf
-cmd upconfig:上传配置文件
-confdir:配置文件的目录
-confname:指定对应的名称
java -classpath .:/home/solr/solrcloud/solr-lib/* org.apache.solr.cloud.ZkCLI -cmd linkconfig -collection collection1 -confname myconf -zkhost 192.168.91.128:2181,192.168.91.129:2181,192.168.91.130:2181
-cmd linkconfig:为指定collection"绑定"配置文件
-collection
-confname
8.检查zookeeper的配置信息
zkCli.sh -server ip:port
ls /configs/myconf
ls /live_nodes
9.启动tomcat，创建collection
curl ‘http://192.168.91.128:8080/solr/admin/collections?action=CREATE&name=mycollection&numShards=3&replicationFactor=1‘
action：CREATE
name：collection名称
numShards：分片数量
replicationFactor：副本集数量
也可以通过zk查询collection的情况：ls /collections/mycollection
-------------------------------
1.ZK CLI
ZooKeeper‘s Command Line Utility (CLI).
zkcli Parameters
Short | Parameter Usage | Meaning |
|---|---|---|
|
| CLI Command to be executed: |
|
| ZooKeeper host address. |
|
| For |
|
| For |
|
| Display help text. |
|
| For |
|
| Run ZooKeeper internally by passing the Solr run port; only for clusters on one machine. |
|
| For |
ZooKeeper CLI Examples
Below are some examples of using the zkcli CLI:
Uploading a Configuration Directory
java -classpath example/solr-webapp/WEB-INF/lib/*org.apache.solr.cloud.ZkCLI -cmd upconfig -zkhost 127.0.0.1:9983-confdir example/solr/collection1/conf -confname conf1 -solrhome example/solr
Put arbitrary data into a new ZK file
java -classpath example/solr-webapp/WEB-INF/lib/*org.apache.solr.cloud.ZkCLI -zkhost 127.0.0.1:9983 -put /data.txt ‘some data‘
Put a local file into a new ZK file
java -classpath example/solr-webapp/WEB-INF/lib/*org.apache.solr.cloud.ZkCLI -zkhost 127.0.0.1:9983 -putfile /data.txt /some/local/file.txt
Linking a Collection to a Configuration Set
java -classpath example/solr-webapp/webapp/WEB-INF/lib/*org.apache.solr.cloud.ZkCLI -cmd linkconfig -zkhost 127.0.0.1:9983-collection collection1 -confname conf1 -solrhome example/solr
Bootstrapping All the Configuration Directories in solr.xml
java -classpath example/solr-webapp/webapp/WEB-INF/lib/*org.apache.solr.cloud.ZkCLI -cmd bootstrap -zkhost 127.0.0.1:9983-solrhome example/solr
Scripts
There are scripts in example/cloud-scripts that handle the classpath and class name for you if you are using Solr out of the box with Jetty. Commands then become:
sh zkcli.sh -cmd linkconfig -zkhost 127.0.0.1:9983-collection collection1 -confname conf1 -solrhome example/solr
2.Managing collections via the collections API
CREATE---
http://localhost:8983/solr/admin/collections?action=CREATE&name=mycollection&numShards=3&replicationFactor=4
name：collection的名称
numShards：指定分片数量（slices）
replicationFactor：副本数量
maxShardsPerNode：默认值为1，注意三个数值：numShards、replicationFactor、liveSolrNode，一个正常的solrCloud集群不容许同一个liveSolrNode上部署同一个shard的多个replic，因此当maxShardsPerNode=1时，numShards*replicationFactor>liveSolrNode时，报错。因此正确时因满足以下条件：
numShards*replicationFactor<liveSolrNode*maxShardsPerNode
createNodeSet：
collection.configName：指定该collection使用那份config，这份config必须存在于zk中。
DELETE---
http://localhost:8983/solr/admin/collections?action=DELETE&name=mycollection
RELOAD---
http://localhost:8983/solr/admin/collections?action=RELOAD&name=mycollection
SPLIT SHARD---
http://localhost:8983/solr/admin/collections?action=SPLITSHARD&collection=<collection_name>&shard=shardId
3.Creating cores via CoreAdmin
curl ‘http://localhost:8983/solr/admin/cores?action=CREATE&name=mycore&collection=collection1&shard=shard2‘
-------------------------------
Q1:Multiple schemas in the same SolrCloud ?
java -classpath .:/home/solrcloud/solr-lib/* org.apache.solr.cloud.ZkCLI -cmd upconfig -zkhost 10.200.51.190:2224,10.200.51.192:2224,10.200.51.194:2224 -confdir /home/solrcloud/fox-config-files/ -confname foxconf
指定配置文件，上传到zk。
curl ‘http://10.200.51.194:2223/solr/admin/collections?action=CREATE&name=fox4&numShards=1&replicationFactor=5&collection.configName=foxconf‘
为创建的collection指定collection.configName。