Elasticsearch+MySQL数据同步示例

@author：wushuang

版本：Elasticsearch2.4Logstash2.4MySQL5.7

操作系统：CentOS Linux 7

前言：做个备份，免得以后忘记。在网上查阅了不少资料，官网讲的还是很清楚。

 

一、安装Elasticsearch

https://www.elastic.co/guide/en/elasticsearch/reference/current/setup.html

二、在MySQL上新建库表(若已有库表可跳过)

建库脚本如下：

SET FOREIGN_KEY_CHECKS=0;

 

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

-- Table structure for contacts

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

DROP TABLE IF EXISTS `contacts`;

CREATE TABLE `contacts` (

  `uid` bigint(20) unsigned NOT NULL AUTO_INCREMENT,

  `email` varchar(80) NOT NULL,

  `first_name` varchar(80) NOT NULL,

  `last_name` varchar(80) NOT NULL,

  UNIQUE KEY `uid` (`uid`)

) ENGINE=InnoDB AUTO_INCREMENT=7 DEFAULT CHARSET=utf8;

 

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

-- Records of contacts

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

INSERT INTO `contacts` VALUES ('1', '[e-mail]', 'Jim', 'Smith');

INSERT INTO `contacts` VALUES ('2', '', 'John', 'Smith');

INSERT INTO `contacts` VALUES ('3', '[e-mail]', 'Carole', 'Smith');

INSERT INTO `contacts` VALUES ('4', '[e-mail]', 'Sam', 'Smith');

INSERT INTO `contacts` VALUES ('5', '[e-mail]', 'Rick', 'Roses');

INSERT INTO `contacts` VALUES ('6', '[e-mail]', 'Sum', 'OKFord');

 

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

-- Table structure for user

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

DROP TABLE IF EXISTS `user`;

CREATE TABLE `user` (

  `USERID` bigint(20) NOT NULL,

  `USERNAME` varchar(20) DEFAULT NULL,

  `AGE` int(3) unsigned DEFAULT NULL,

  `SEX` tinyint(1) DEFAULT NULL,

  PRIMARY KEY (`USERID`)

) ENGINE=InnoDB DEFAULT CHARSET=utf8;

 

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

-- Records of user

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

INSERT INTO `user` VALUES ('201609210935231', 'Quejs', '24', '1');

INSERT INTO `user` VALUES ('201609210936233', 'Smith', '24', '1');

INSERT INTO `user` VALUES ('201609210938234', 'Mends', '26', '0');

INSERT INTO `user` VALUES ('201609210947235', 'wisd', '38', '0');

INSERT INTO `user` VALUES ('201609210948236', 'Losed', '35', '0');

SET FOREIGN_KEY_CHECKS=1;

 

三、安装Logstash

https://www.elastic.co/guide/en/logstash/current/installing-logstash.html

四、安装logstash-input-jdbc和logstash-output-elasticsearch插件

https://www.elastic.co/guide/en/logstash/current/plugins-inputs-jdbc.html

https://www.elastic.co/guide/en/logstash/current/plugins-outputs-elasticsearch.html

五、新建配置文件

1.需要将MYSQL的jar文件复制到Logstash安装目录下，示例中放在了/home/manager/app/logstash-2.4.0下。如图：

 

 

2.在目录/home/manager/app/logstash-2.4.0/bin下新建config文件夹。

命令：切换到/home/manager/app/logstash-2.4.0/bin目录下，mkdir config

3.在目录/home/manager/app/logstash-2.4.0/bin/config/下新建conf目录，命令：mkdir conf

4.新建两个配置文件contacts.conf和user.conf。

user.conf内容：

input {

    stdin {

    }

    jdbc {

      # mysql jdbc connection string to our backup databse

      jdbc_connection_string => "jdbc:mysql://localhost:3306/mytest"

      # the user we wish to excute our statement as

      jdbc_user => "manager"

      jdbc_password => "manager@2015"

      # the path to our downloaded jdbc driver

      jdbc_driver_library => "/home/manager/app/logstash-2.4.0/mysql-connector-java-5.1.22.jar"

      # the name of the driver class for mysql

      jdbc_driver_class => "com.mysql.jdbc.Driver"

      jdbc_paging_enabled => "true"

      jdbc_page_size => "50000"

      statement_filepath => "/home/manager/app/logstash-2.4.0/bin/config/sql/user.sql"

      schedule => "* * * * *"

      type => "user"

    }

}

 

filter {

    json {

        source => "message"

        remove_field => ["message"]

    }

}

 

output {

if[type]=="user"{

    elasticsearch {

        hosts => ["127.0.0.1:9200"]

        index => "mytest"

 document_type => "user"

        document_id => "%{userid}"

 workers => 1

        flush_size => 20000

        idle_flush_time => 10

        template_overwrite => true

    }

 

 

    stdout {

        codec => json_lines

    }

}

}

 

contacts.conf内容：

input {

    stdin {

    }

    jdbc {

      # mysql jdbc connection string to our backup databse

      jdbc_connection_string => "jdbc:mysql://localhost:3306/mytest"

      # the user we wish to excute our statement as

      jdbc_user => "manager"

      jdbc_password => "manager@2015"

      # the path to our downloaded jdbc driver

      jdbc_driver_library => "/home/manager/app/logstash-2.4.0/mysql-connector-java-5.1.22.jar"

      # the name of the driver class for mysql

 

      jdbc_driver_class => "com.mysql.jdbc.Driver"

      jdbc_paging_enabled => "true"

      jdbc_page_size => "50000"

      statement_filepath => "/home/manager/app/logstash-2.4.0/bin/config/sql/user.sql"

      schedule => "* * * * *"

      type => "user"

    }

}

 

filter {

    json {

        source => "message"

        remove_field => ["message"]

    }

}

 

output {

if[type]=="user"{

    elasticsearch {

        hosts => ["127.0.0.1:9200"]

 

        index => "mytest"

 document_type => "user"

        document_id => "%{userid}"

 workers => 1

        flush_size => 20000

        idle_flush_time => 10

        template_overwrite => true

}

 

    stdout {

        codec => json_lines

    }

}

}

六、启动，验证

1.到Logstash bin目录下启动Logstash：nohup ./logstash -f ./config/conf/ >/dev/null 2>&1 &

2.验证数据是否同步过来了：curl -XGET 'http://localhost:9200/mytest/contacts/_search?pretty'

 

七、注意事项

1.只能逻辑删除，物理删除同步不了。

2.以上示例是全量同步，增量的需要每张表加timespan字段。Stackoverflow上可以查一查。

3.性能还得进行压测，目前先实现第一步。