通过springboot 同步pgsql数据到kafka 发送到logstash 再同步到es

安装 es

下载安装包 es安装包地址

kafka安装

brew install kafka 

安装的配置文件位置

/usr/local/etc/kafka/server.properties
/usr/local/etc/kafka/zookeeper.properties

启动kafka

运行kafka需要使用Zookeeper，所以你需要先启动Zookeeper，如果你没有Zookeeper，你可以使用kafka自带打包和配置好的Zookeeper

启动zookeeper

zookeeper-server-start /usr/local/etc/kafka/zookeeper.properties

启动kafka

kafka-server-start /usr/local/etc/kafka/server.properties

安装logstash

安装步骤略

重点：配置文件

# Sample Logstash configuration for creating a simple
# Kafka -> Logstash -> Elasticsearch pipeline.

input {
    tcp {
        host => "127.0.0.1"
        port => 5544
        type => tcp
        codec => json
    }
    kafka {
        bootstrap_servers => "localhost:9092"
        topics => ["BookList","BookDuplicate"]
        decorate_events => true #可向事件添加Kafka元数据比如 topic，消费者组
    }
    jdbc {
        jdbc_connection_string => "jdbc:postgresql://127.0.0.1:5432/folio"
        jdbc_user => "folio"
        jdbc_password => "folio123" #更换自己的数据库用户名和密码
        jdbc_driver_library => "/Users/yanziyu/logstash/logstash-7.7.1/bin/postgresql-42.2.14.jar" #更换自己的驱动地址
        jdbc_driver_class => "org.postgresql.Driver"
        jdbc_paging_enabled => "true"
        jdbc_page_size => "300000"
        use_column_value => "true"
        tracking_column => "id"
        #这是要操作的sql 表，下面会提供
        statement => "select jsonb::TEXT  from shlibrary_mod_shl_inventory.booklist_add"
        schedule => "* * * * *"
        type => "jdbc_booklist"
        jdbc_default_timezone =>"Asia/Shanghai"
    }
    jdbc {
        jdbc_connection_string => "jdbc:postgresql://127.0.0.1:5432/folio"
        jdbc_user => "folio"
        jdbc_password => "folio123" #更换自己的数据库用户名和密码
        jdbc_driver_library => "/Users/yanziyu/logstash/logstash-7.7.1/bin/postgresql-42.2.14.jar" #更换自己的驱动地址
        jdbc_driver_class => "org.postgresql.Driver"
        jdbc_paging_enabled => "true"
        jdbc_page_size => "300000"
        use_column_value => "true"
        tracking_column => "id"
         #这是要操作的sql 表，下面会提供
        statement => "select jsonb::TEXT  from shlibrary_mod_shl_inventory.bookduplicate_add"
        schedule => "* * * * *"
        type => "jdbc_bookduplicate"
        jdbc_default_timezone =>"Asia/Shanghai"
    }
}

filter {

    json {
        source => "message"  #将message中数据分成对应的field
        #remove_field => ["message"]
    }
    mutate {
          remove_field => ["@version","@timestamp"] #移除字段
    }

}

output {
    if [type] == "tcp" {
        elasticsearch {
            hosts => ["http://localhost:9200"]
            index => "dataloader-es-log"
        }
    }
    if [type]== "jdbc_booklist" {
        kafka {
            bootstrap_servers => "localhost:9092" #kafka服务器地址
            topic_id => "BookList_Add"
            batch_size => 5
            codec => "json" #写入的时候使用json编码，因为logstash收集后会转换成json格式
        }
    }
    if [type]== "jdbc_bookduplicate" {
        kafka {
            bootstrap_servers => "localhost:9092" #kafka服务器地址
            topic_id => "BookDuplicate_Add"
            batch_size => 5
            codec => "json" #写入的时候使用json编码，因为logstash收集后会转换成json格式
        }
    }
    if [@metadata][kafka][topic] == "BookList" {
        elasticsearch {
            hosts => ["http://localhost:9200"]
            index => "booklist"
            document_id => "%{id}" #将message中的id设置为_doc的Id
        }
    }
    if [@metadata][kafka][topic] == "BookDuplicate" {
        elasticsearch {
            hosts => ["http://localhost:9200"]
            index => "bookduplicate"
            document_id => "%{id}" #将message中的id设置为_doc的Id
        }
    }
}

本文地址：https://blog.csdn.net/yanziyu0721730/article/details/107456723