elasticsearch-2.2.0-集群搭建

Hadoop-2.5.1-HA搭建

注意事项： yml文件要顶格，封号后面的空格要保留；所有文件都将保留在home目录下

	NN	DN	ZK	ZKFC	JN	RM	NM（任务管理）	Elasticsearch	Kibana
Hadoop1	Y		Y	Y		Y		Y	Y
Hadoop2	Y	Y	Y	Y	Y	Y	Y	Y
Hadoop3		Y	Y		Y		Y	Y
Hadoop4		Y			Y		Y	Y

一、解压

tar -zvxf elasticsearch-2.2.0.tar.gz

mv elasticsearch-2.2.0 /home/

cd /home/elasticsearch-2.2.0

二、配置

[root@hadoop1 elasticsearch-2.2.0]# vi config/elasticsearch.yml

cluster.name: chenkl
node.name: hadoop1
network.host: 192.168.25.151
# 防止脑裂配置
discovery.zen.ping.multicast.enabled: false
discovery.zen.ping_timeout: 120s
client.transport.ping_timeout: 60s
discovery.zen.ping.unicast.hosts: ["192.168.25.151","192.168.25.152","192.168.25.153","192.168.25.154"]

[root@hadoop1 home]# scp -r elasticsearch-2.2.0/ root@hadoop2:/home/
[root@hadoop1 home]# scp -r elasticsearch-2.2.0/ root@hadoop3:/home/
[root@hadoop1 home]# scp -r elasticsearch-2.2.0/ root@hadoop4:/home/

更改相应的配置

node.name: hadoop1

network.host: 192.168.25.151

三、所有节点创建用户、并且启动

 ~]# adduser bigdata

 ~]# su bigdata

[bigdata@hadoop1 root]$ exit

[root@hadoop1 elasticsearch-2.2.0]# chown -R bigdata:bigdata ../

[root@hadoop1 elasticsearch-2.2.0]# su bigdata

[bigdata@hadoop1 elasticsearch-2.2.0]$ cd /home/elasticsearch-2.2.0

[bigdata@hadoop1 elasticsearch-2.2.0]$ bin/elasticsearch

Ctrl+c可以停止上面开启的服务

到此elasticsearch安装启动完毕

192.168.25.151:9200

192.168.25.151:9200/_cluster/health

查看集群状态，status 为yellow/green 则说明正常

四、安装Kibana

kibana-4.4.1-linux-x64.tar.gz
hadoop1, 解压安装，修改配置文件vi config/kibana.yml的elasticsearch.url属性即可

 注意配置yml结尾的配置文件都需要冒号后面加空格才行

elasticsearch安装插件，所有节点

[root@hadoop1 home]# cd elasticsearch-2.2.0/

[root@hadoop1 elasticsearch-2.2.0]# bin/plugin install license

[root@hadoop1 elasticsearch-2.2.0]# bin/plugin install marvel-agent

装有kibana的节点安装

[root@hadoop1 kibana-4.4.1-linux-x64]# bin/kibana plugin --install elasticsearch/marvel/latest
marvel-2.2.1能和kibana-4.4.1完美结合，指定插件版本安装
[root@hadoop1 kibana-4.4.1-linux-x64]# bin/kibana plugin --install elasticsearch/marvel/2.2.1
卸载或者删除插件命令
[root@hadoop1 kibana-4.4.1-linux-x64]# bin/kibana plugin -r marvel

接下来启动elasticsearch和kibana

[bigdata@hadoop1 elasticsearch-2.2.0]$ bin/elasticsearch

浏览器输入：http://192.168.25.151:9200 访问

[root@node1 kibana-4.4.1-linux-x64]# bin/kibana

index ready 说明启动成功

浏览器输入：http://192.168.25.151:5601 访问

五、集成中文分词器（已经编译过的），所有节点

ik分词器和es的版本对应关系，以及如何手动编译ik分词器到es中使用
https://github.com/medcl/elasticsearch-analysis-ik/tree/v6.2.4
 

[bigdata@hadoop1 elasticsearch-2.2.0]# mkdir plugins/ik

[bigdata@hadoop1 ik]# unzip elasticsearch-analysis-ik-1.8.0.zip

[bigdata@hadoop1 ik]# ll

要有这么几个文件

common-codec-1.9.jar

common-logging-1.2.jar

config

elastcsearch-analysis-ik-1.8.0.jar

httpclient-4.4.1.jar

httpcore-4.4.1.jar

plugin-descriptor.properties

复制到其他节点上
 

[bigdata@hadoop1 ik]# scp -r ./ root@hadoop2:/home/elasticsearch-2.2.0/plugins/ik/

[bigdata@hadoop1 ik]# scp -r ./ root@hadoop3:/home/elasticsearch-2.2.0/plugins/ik/

[bigdata@hadoop1 ik]# scp -r ./ root@hadoop4:/home/elasticsearch-2.2.0/plugins/ik/

利用ik分词器设置搜索域    创建索引库cd /home/
vi dkjhl.json

{
    "settings":{
        "number_of_shards":5,
        "number_of_replicas":0
    },
    "mappings":{
        "doc":{
            "dynamic":"strict",
            "properties":{
                "id":{"type":"integer","store":"yes"},
                "title":{"type":"string","store":"yes","index":"analyzed","analyzer": "ik_max_word","search_analyzer": "ik_max_word"},
                "describe":{"type":"string","store":"yes","index":"analyzed","analyzer": "ik_max_word","search_analyzer": "ik_max_word"},
                "author":{"type":"string","store":"yes","index":"no"}
            }
        }
    }

}

删除索引库

[root@hadoop1 home]# curl -XDELETE 'hadoop1:9200/dkjhl'

创建索引库

[root@hadoop1 home]# curl -XPOST 'hadoop1:9200/dkjhl' -d @dkjhl.json

 

下次启动elasticsearch

所有节点

cd /home/elasticsearch-2.2.0/

su bigdata

后台启动

$ bin/elasticsearch -d

hadoop1节点

cd /home/kibana-4.4.1-linux-x64/

后台启动

bin/kibana &

等待一会后，加载出status信息，接着输入“exit”回车即可