一次同步多張表是開發中的一般需求。之前研究了很久找到方法，但沒有詳細總結。 
博友前天線上提問，說明這塊理解的還不夠透徹。 
我整理下， 
一是為了儘快解決博友問題， 
二是加深記憶，便於未來產品開發中快速上手。

1、同步原理

原有ES專欄中有詳解，不再贅述。詳細請參考我的專欄： 
深入詳解Elasticsearch 
以下是通過ES5.4.0， logstash5.4.1 驗證成功。 
可以確認的是2.X版本同樣可以驗證成功。

2、核心配置檔案

input {
  stdin {
  }

  jdbc {
  type => "cxx_article_info"
  # mysql jdbc connection string to our backup databse 後面的test對應mysql中的test資料庫
 

  jdbc_connection_string => "jdbc:mysql://110.10.15.37:3306/cxxwb"
  # the user we wish to excute our statement as
  jdbc_user => "root"
  jdbc_password => "xxxxx"

  record_last_run => "true"
  use_column_value => "true"
  tracking_column => "id"
  last_run_metadata_path => "/opt/logstash/bin/logstash_xxy/cxx_info"
 

  clean_run => "false"

  # the path to our downloaded jdbc driver
  jdbc_driver_library => "/opt/elasticsearch/lib/mysql-connector-java-5.1.38.jar"
  # the name of the driver class for mysql
  jdbc_driver_class => "com.mysql.jdbc.Driver"
  jdbc_paging_enabled => "true"
  jdbc_page_size => "500"
 

  statement => "select * from cxx_article_info where id > :sql_last_value"
#定時欄位 各欄位含義（由左至右）分、時、天、月、年，全部為*預設含義為每分鐘都更新
  schedule => "* * * * *"
#設定ES索引型別
  }

  jdbc {
  type => "cxx_user"
  # mysql jdbc connection string to our backup databse 後面的test對應mysql中的test資料庫
  jdbc_connection_string => "jdbc:mysql://110.10.15.37:3306/cxxwb"
  # the user we wish to excute our statement as
  jdbc_user => "root"
  jdbc_password => "xxxxxx"

  record_last_run => "true"
  use_column_value => "true"
  tracking_column => "id"
  last_run_metadata_path => "/opt/logstash/bin/logstash_xxy/cxx_user_info"
  clean_run => "false"

  # the path to our downloaded jdbc driver
  jdbc_driver_library => "/opt/elasticsearch/lib/mysql-connector-java-5.1.38.jar"
  # the name of the driver class for mysql
  jdbc_driver_class => "com.mysql.jdbc.Driver"
  jdbc_paging_enabled => "true"
  jdbc_page_size => "500"
  statement => "select * from cxx_user_info where id > :sql_last_value"
#以下對應著要執行的sql的絕對路徑。
#statement_filepath => "/opt/logstash/bin/logstash_mysql2es/department.sql"
#定時欄位 各欄位含義（由左至右）分、時、天、月、年，全部為*預設含義為每分鐘都更新
schedule => "* * * * *"
#設定ES索引型別
  }

}

filter {
mutate {
  convert => [ "publish_time", "string" ]
 }

date {
  timezone => "Europe/Berlin"
  match => ["publish_time" , "ISO8601", "yyyy-MM-dd HH:mm:ss"]
}
#date {
 # match => [ "publish_time", "yyyy-MM-dd HH:mm:ss,SSS" ]
  # remove_field => [ "publish_time" ]
  # }
json {
  source => "message"
  remove_field => ["message"]
  }
}

output {

if [type]=="cxxarticle_info" {
  elasticsearch {
#ESIP地址與埠
  hosts => "10.100.11.231:9200"
#ES索引名稱（自己定義的）
  index => "cxx_info_index"
#自增ID編號
 # document_id => "%{id}"
  }
}

if [type]=="cxx_user" {
  elasticsearch {
#ESIP地址與埠
  hosts => "10.100.11.231:9200"
#ES索引名稱（自己定義的）
  index => "cxx_user_index"
#自增ID編號
 # document_id => "%{id}"
  }
}

}

• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13
• 14
• 15
• 16
• 17
• 18
• 19
• 20
• 21
• 22
• 23
• 24
• 25
• 26
• 27
• 28
• 29
• 30
• 31
• 32
• 33
• 34
• 35
• 36
• 37
• 38
• 39
• 40
• 41
• 42
• 43
• 44
• 45
• 46
• 47
• 48
• 49
• 50
• 51
• 52
• 53
• 54
• 55
• 56
• 57
• 58
• 59
• 60
• 61
• 62
• 63
• 64
• 65
• 66
• 67
• 68
• 69
• 70
• 71
• 72
• 73
• 74
• 75
• 76
• 77
• 78
• 79
• 80
• 81
• 82
• 83
• 84
• 85
• 86
• 87
• 88
• 89
• 90
• 91
• 92
• 93
• 94
• 95
• 96
• 97
• 98
• 99
• 100
• 101
• 102
• 103
• 104

3、同步成功結果

[2017-07-19T15:08:05,438][INFO ][logstash.pipeline ] Pipeline main started
The stdin plugin is now waiting for input:
[2017-07-19T15:08:05,491][INFO ][logstash.agent ] Successfully started Logstash API endpoint {:port=>9600}
[2017-07-19T15:09:00,721][INFO ][logstash.inputs.jdbc ] (0.007000s) SELECT count(*) AS `count` FROM (select * from cxx_article_info where id > 0) AS `t1` LIMIT 1
[2017-07-19T15:09:00,721][INFO ][logstash.inputs.jdbc ] (0.008000s) SELECT count(*) AS `count` FROM (select * from cxx_user_info where id > 0) AS `t1` LIMIT 1
[2017-07-19T15:09:00,730][INFO ][logstash.inputs.jdbc ] (0.004000s) SELECT * FROM (select * from cxx_user_info where id > 0) AS `t1` LIMIT 500 OFFSET 0
[2017-07-19T15:09:00,731][INFO ][logstash.inputs.jdbc ] (0.007000s) SELECT * FROM (select * from cxx_article_info where id > 0) AS `t1` LIMIT 500 OFFSET 0
[2017-07-19T15:10:00,173][INFO ][logstash.inputs.jdbc ] (0.002000s) SELECT count(*) AS `count` FROM (select * from cxx_article_info where id > 3) AS `t1` LIMIT 1
[2017-07-19T15:10:00,174][INFO ][logstash.inputs.jdbc ] (0.003000s) SELECT count(*) AS `count` FROM (select * from cxx_user_info where id > 2) AS `t1` LIMIT 1
[2017-07-19T15:11:00,225][INFO ][logstash.inputs.jdbc ] (0.001000s) SELECT count(*) AS `count` FROM (select * from cxx_article_info where id > 3) AS `t1` LIMIT 1
[2017-07-19T15:11:00,225][INFO ][logstash.inputs.jdbc ] (0.002000s) SELECT count(*) AS `count` FROM (select * from cxx_user_info where id > 2) AS `t1` LIMIT 1

• 1
• 2
• 3
• 4
• 5
• 6
• 7
• 8
• 9
• 10
• 11
• 12
• 13

4、擴充套件

1）多個表無非就是在input裡面多加幾個型別，在output中多加基礎 
型別判定。 
舉例：

if [type]=="cxx_user"  

• 1

2）input裡的type和output if判定的type**保持一致**，該type對應ES中的type。

後記

死磕ES，有問題歡迎大家提問探討！

—————————————————————————————————— 
更多ES相關實戰乾貨經驗分享，請掃描下方【銘毅天下】微信公眾號二維碼關注。 
（每週至少更新一篇！）

2017年07月19日 23:32 於家中床前