阅读此文请先阅读上文：[大数据]-Elasticsearch5.3.1 IK分词，同义词/联想搜索设置，前面介绍了ES，Kibana5.3.1的安装配置，以及IK分词的安装和同义词设置，这里主要记录Logstash导入mysql数据到Elasticsearch5.3.1并设置IK分词和同义词。由于logstash配置好JDBC，ES连接之后运行脚本一站式创建index，mapping，导入数据。但是如果我们要配置IK分词器就需要修改创建index,mapping的配置，下面详细介绍。
一、Logstash-5.3.1下载安装:
- 下载：https://www.elastic.co/cn/downloads/logstash
- 解压：tar -zxf logstash-5.3.1.tar.gz
- 启动：bin/logstash -e 'input { stdin { } } output { stdout {} }' （参数表示终端输入输出）如下则成功。
-
Sending Logstash's logs to /home/rzxes/logstash-5.3.1/logs which is now configured via log4j2.properties [2017-05-16T10:27:36,957][INFO ][logstash.setting.writabledirectory] Creating directory {:setting=>"path.queue", :path=>"/home/rzxes/logstash-5.3.1/data/queue"} [2017-05-16T10:27:37,041][INFO ][logstash.agent ] No persistent UUID file found. Generating new UUID {:uuid=>"c987803c-9b18-4395-bbee-a83a90e6ea60", :path=>"/home/rzxes/logstash-5.3.1/data/uuid"} [2017-05-16T10:27:37,581][INFO ][logstash.pipeline ] Starting pipeline {"id"=>"main", "pipeline.workers"=>1, "pipeline.batch.size"=>125, "pipeline.batch.delay"=>5, "pipeline.max_inflight"=>125} [2017-05-16T10:27:37,682][INFO ][logstash.pipeline ] Pipeline main started The stdin plugin is now waiting for input: [2017-05-16T10:27:37,886][INFO ][logstash.agent ] Successfully started Logstash API endpoint {:port=>9600}
二、Logstash-5.3.1连接mysql作为数据源，ES作为数据输出端：
- 由于此版本的logstash已经集成了jdbc插件，我们只需要添加一个配置文件xxx.conf。内容如下test.conf：
-
input { stdin { } jdbc { # 数据库地址 端口 数据库名 jdbc_connection_string => "jdbc:mysql://IP:3306/dbname" # 数据库用户名 jdbc_user => "user" # 数据库密码 jdbc_password => "pass" # mysql java驱动地址 jdbc_driver_library => "/home/rzxes/logstash-5.3.1/mysql-connector-java-5.1.17.jar" jdbc_driver_class => "com.mysql.jdbc.Driver" jdbc_paging_enabled => "true" jdbc_page_size => "100000" # sql 语句文件，也可以直接写SQL，如statement => "select * from table1" statement_filepath => "/home/rzxes/logstash-5.3.1/test.sql" schedule => "* * * * *" type => "jdbc" } } output { stdout { codec => json_lines } elasticsearch { hosts => "192.168.230.150:9200" index => "test-1" #索引名称 document_type => "form" #type名称 document_id => "%{id}" #id必须是待查询的数据表的序列字段} } - 创建一个SQL文件：如上配置test.sql内容: select * from table1
- test.conf,test.sql文件都在logstash的根目录下。
- 运行logstash脚本导入数据： bin/logstash -f test.conf 启动如下;
- 等待数据导入完成。开启Es-head,访问9100端口如下：
-
可以看到已经导入了11597条数据。
- 更多详细的配置参考官方文档：plugins-inputs-jdbc-jdbc_driver_library
三、logstash是如何创建index，mapping，并导入数据？
ES导入数据必须先创建index，mapping，但是在logstash中并没有直接创建，我们只传入了index，type等参数，logstash是通过es的mapping template来创建的，这个模板文件不需要指定字段，就可以根据输入自动生成。在logstash启动的时候这个模板已经输出了如下log:
[2017-05-23T15:58:45,801][WARN ][logstash.outputs.elasticsearch] Restored connection to ES instance {:url=>#<URI::HTTP:0x68f0d43b URL:http://192.168.230.150:9200/>}
[2017-05-23T15:58:45,805][INFO ][logstash.outputs.elasticsearch] Using mapping template from {:path=>nil}
[2017-05-23T15:58:45,979][INFO ][logstash.outputs.elasticsearch] Attempting to install template {:manage_template=>{"template"=>"logstash-*", "version"=>50001, "settings"=>{"index.refresh_interval"=>"5s"}, "mappings"=>{"_default_"=>{"_all"=>{"enabled"=>true, "norms"=>false}, "dynamic_templates"=>[{"message_field"=>{"path_match"=>"message", "match_mapping_type"=>"string", "mapping"=>{"type"=>"text", "norms"=>false}}}, {"string_fields"=>{"match"=>"*", "match_mapping_type"=>"string", "mapping"=>{"type"=>"text", "norms"=>false, "fields"=>{"keyword"=>{"type"=>"keyword"}}}}}], "properties"=>{"@timestamp"=>{"type"=>"date", "include_in_all"=>false}, "@version"=>{"type"=>"keyword", "include_in_all"=>false}, "geoip"=>{"dynamic"=>true, "properties"=>{"ip"=>{"type"=>"ip"}, "location"=>{"type"=>"geo_point"}, "latitude"=>{"type"=>"half_float"}, "longitude"=>{"type"=>"half_float"}}}}}}}}
- 添加IK分词，只需要创建一个json文件： vim /home/rzxes/logstash-5.3.1/template/logstash.json 添加如下内容：
-
{ "template": "*", "version": 50001, "settings": { "index.refresh_interval": "5s" }, "mappings": { "_default_": { "_all": { "enabled": true, "norms": false }, "dynamic_templates": [ { "message_field": { "path_match": "message", "match_mapping_type": "string", "mapping": { "type": "text", "norms": false } } }, { "string_fields": { "match": "*", "match_mapping_type": "string", "mapping": { "type": "text", "norms": false, "analyzer": "ik_max_word",#只需要添加这一行即可设置分词器为ik_max_word "fields": { "keyword": { "type": "keyword" } } } } } ], "properties": { "@timestamp": { "type": "date", "include_in_all": false }, "@version": { "type": "keyword", "include_in_all": false } } } } } -
如需配置同义词，需自定义分词器，配置同义词过滤<IK分词同义词详见上一篇文章>。修改模板logstash.json如下:
-
{ "template" : "*", "version" : 50001, "settings" : { "index.refresh_interval" : "5s", #分词，同义词配置：自定义分词器，过滤器，如不配同义词则没有index这一部分 "index": { "analysis": { "analyzer": { "by_smart": { "type": "custom", "tokenizer": "ik_smart", "filter": ["by_tfr","by_sfr"], "char_filter": ["by_cfr"] }, "by_max_word": { "type": "custom", "tokenizer": "ik_max_word", "filter": ["by_tfr","by_sfr"], "char_filter": ["by_cfr"] } }, "filter": { "by_tfr": { "type": "stop", "stopwords": [" "] }, "by_sfr": { "type": "synonym", "synonyms_path": "analysis/synonyms.txt" #同义词路径 } }, "char_filter": { "by_cfr": { "type": "mapping", "mappings": ["| => |"] } } } } # index --end-- }, "mappings" : { "_default_" : { "_all" : { "enabled" : true, "norms" : false }, "dynamic_templates" : [ { "message_field" : { "path_match" : "message", "match_mapping_type" : "string", "mapping" : { "type" : "text", "norms" : false }} }, { "string_fields" : { "match" : "*", "match_mapping_type" : "string", "mapping" : { "type" : "text", "norms" : false, #选择分词器：自定义分词器，或者ik_mmax_word "analyzer" : "by_max_word", "fields" : { "keyword" : { "type" : "keyword" } } } } } ], "properties" : { "@timestamp" : { "type" : "date", "include_in_all" : false }, "@version" : { "type" : "keyword", "include_in_all" : false } } } } } - 有了自定义模板文件，test.conf中配置模板覆盖使模板生效。test.conf最终配置如下：
-
input { stdin { } jdbc { # 数据库地址 端口 数据库名 jdbc_connection_string => "jdbc:mysql://IP:3306/dbname" # 数据库用户名 jdbc_user => "user" # 数据库密码 jdbc_password => "pass" # mysql java驱动地址 jdbc_driver_library => "/home/rzxes/logstash-5.3.1/mysql-connector-java-5.1.17.jar" jdbc_driver_class => "com.mysql.jdbc.Driver" jdbc_paging_enabled => "true" jdbc_page_size => "100000" # sql 语句文件 statement_filepath => "/home/rzxes/logstash-5.3.1/mytest.sql" schedule => "* * * * *" type => "jdbc" } } output { stdout { codec => json_lines } elasticsearch { hosts => "192.168.230.150:9200" index => "test-1" document_type => "form" document_id => "%{id}" #id必须是待查询的数据表的序列字段 template_overwrite => true template => "/home/rzxes/logstash-5.3.1/template/logstash.json" } } - 删除上次创建的index（由于数据导入时会根据原有数据的index，mapping进行索引创建），重新启动logstash。
- 最终在Kibana中检索关键词 番茄，就会发现西红柿也会被检索到。如下图：
- 致此logstash数据导入的template重写就完成了。
- 另一种方式配置IK分词:全局配置，不需要自定义模板。
-
curl -XPUT "http://192.168.230.150:9200/_template/rtf" -H 'Content-Type: application/json' -d' { "template" : "*", "version" : 50001, "settings" : { "index.refresh_interval" : "5s", "index": { "analysis": { "analyzer": { "by_smart": { "type": "custom", "tokenizer": "ik_smart", "filter": ["by_tfr","by_sfr"], "char_filter": ["by_cfr"] }, "by_max_word": { "type": "custom", "tokenizer": "ik_max_word", "filter": ["by_tfr","by_sfr"], "char_filter": ["by_cfr"] } }, "filter": { "by_tfr": { "type": "stop", "stopwords": [" "] }, "by_sfr": { "type": "synonym", "synonyms_path": "analysis/synonyms.txt" } }, "char_filter": { "by_cfr": { "type": "mapping", "mappings": ["| => |"] } } } } }, "mappings" : { "_default_" : { "_all" : { "enabled" : true, "norms" : false }, "dynamic_templates" : [ { "message_field" : { "path_match" : "message", "match_mapping_type" : "string", "mapping" : { "type" : "text", "norms" : false }} }, { "string_fields" : { "match" : "*", "match_mapping_type" : "string", "mapping" : { "type" : "text", "norms" : false, "analyzer" : "by_max_word", "fields" : { "keyword" : { "type" : "keyword" } } } } } ], "properties" : { "@timestamp" : { "type" : "date", "include_in_all" : false }, "@version" : { "type" : "keyword", "include_in_all" : false } } } } }' - 可以使用curl查看模板： curl -XGET "http://192.168.230.150:9200/_template"