宜宾网站建设公司

好程序员大数据学习路线之Logstach与flume对比

好程序员大数据学习路线之Logstach与flume对比，没有集群的概念,logstach与flume都称为组

我们提供的服务有：成都网站建设、成都网站设计、微信公众号开发、网站优化、网站认证、余干ssl等。为超过千家企事业单位解决了网站和推广的问题。提供周到的售前咨询和贴心的售后服务，是有科学管理、有技术的余干网站制作公司

logstash是用JRuby语言开发的

组件的对比:

logstach : input  filter  output

flume    : source  channel  sink  

优劣对比:

logstach :

 安装简单,安装体积小

 有filter组件,使得该工具具有数据过滤,数据切分的功能

 可以与ES无缝结合

 具有数据容错功能,在数据采集的时候,如果发生宕机或断开的情况,会断点续传(会记录读取的偏移量)

综上,该工具主要用途为采集日志数据

flume:

 高可用方面要比logstach强大

 flume一直在强调数据的安全性,flume在数据传输过程中是由事务控制的

 flume可以应用在多类型数据传输领域

数据对接

将logstach.gz文件上传解压即可

可以在logstach目录下创建conf文件,用来存储配置文件

一  命令启动

1.bin/logstash -e 'input { stdin {} } output { stdout{} }'  

stdin/stdout(标准输入输出流)

hello xixi

2018-09-12T21:58:58.649Z hadoop01 hello xixi

hello haha

2018-09-12T21:59:19.487Z hadoop01 hello haha

2.bin/logstash -e 'input { stdin {} } output { stdout{codec => rubydebug} }'

hello xixi

{

       "message" => "hello xixi",

      "@version" => "1",

    "@timestamp" => "2018-09-12T22:00:49.612Z",

          "host" => "hadoop01"

}

3.es集群中 ,需要启动es集群

bin/logstash -e 'input { stdin {} } output { elasticsearch {hosts => ["192.168.88.81:9200"]} stdout{} }'

输入命令后,es自动生成index,自动mapping.

hello haha

2018-09-12T22:13:05.361Z hadoop01 hehello haha

bin/logstash -e 'input { stdin {} } output { elasticsearch {hosts => ["192.168.88.81:9200", "192.168.88.82:9200"]} stdout{} }'

4.kafka集群中,启动kafka集群

bin/logstash -e 'input { stdin {} } output { elasticsearch {hosts => ["192.168.88.81:9200", "192.168.88.82:9200"]} stdout{} }'

二  配置文件启动

需要启动zookeeper集群,kafka集群,es集群

1.与kafka数据对接

vi logstash-kafka.conf

启动

bin/logstash -f logstash-kafka.conf  (-f:指定文件)

在另一节点上启动kafka消费命令

input {

  file {

    path => "/root/data/test.log"

    discover_interval => 5

    start_position => "beginning"

  }

}

 

output {

    kafka {

  topic_id => "test1"

  codec => plain {

        format => "%{message}"

charset => "UTF-8"

      }

  bootstrap_servers => "node01:9092,node02:9092,node03:9092"

    }

}

2.与kafka-es数据对接

vi logstash-es.conf

#启动logstash

bin/logstash -f logstash-es.conf

在另一节点上启动kafka消费命令

input {

file {

type => "gamelog"

path => "/log/*/*.log"

discover_interval => 10

start_position => "beginning"

}

}

 

output {

    elasticsearch {

index => "gamelog-%{+YYYY.MM.dd}"

        hosts => ["node01:9200", "node02:9200", "node03:9200"]

    }

}

数据对接过程

logstach节点存放: 哪个节点空闲资源多放入哪个节点 (灵活存放)

好程序员大数据学习路线之Logstach与flume对比

1.启动logstach监控logserver目录,把数据采集到kafka

2.启动另外一个logstach,监控kafka某个topic数据,把他采集到elasticsearch

数据对接案例

需要启动两个logstach,调用各个配置文件,进行对接

1.采集数据到kafka

cd conf

创建配置文件: vi gs-kafka.conf

input {

  file {

codec => plain {

      charset => "GB2312"

    }

    path => "/root/basedir/*/*.txt"

    discover_interval => 5

    start_position => "beginning"

  }

}

 

output {

    kafka {

  topic_id => "gamelogs"

  codec => plain {

        format => "%{message}"

charset => "GB2312"

      }

  bootstrap_servers => "node01:9092,node02:9092,node03:9092"

    }

}

创建kafka对应的topic

bin/kafka-topics.sh --create --zookeeper hadoop01:2181 --replication-factor 1 --partitions 1 --topic gamelogs

2.在hadoop01上启动logstach

bin/logstash -f conf/gs-kafka.conf

3.在hadoop02上启动另外一个logstach

cd logstach/conf

vi kafka-es.conf

input {

  kafka {

    type => "accesslogs"

    codec => "plain"

    auto_offset_reset => "smallest"

    group_id => "elas1"

    topic_id => "accesslogs"

    zk_connect => "node01:2181,node02:2181,node03:2181"

  }

 

  kafka {

    type => "gamelogs"

    auto_offset_reset => "smallest"

    codec => "plain"

    group_id => "elas2"

    topic_id => "gamelogs"

    zk_connect => "node01:2181,node02:2181,node03:2181"

  }

}

 

filter {

  if [type] == "accesslogs" {

    json {

      source => "message"

  remove_field => [ "message" ]

  target => "access"

    }

  }

 

  if [type] == "gamelogs" {

    mutate {

      split => { "message" => " " }

      add_field => {

        "event_type" => "%{message[3]}"

        "current_map" => "%{message[4]}"

        "current_X" => "%{message[5]}"

        "current_y" => "%{message[6]}"

        "user" => "%{message[7]}"

        "item" => "%{message[8]}"

        "item_id" => "%{message[9]}"

        "current_time" => "%{message[12]}"

     }

     remove_field => [ "message" ]

   }

  }

}

 

output {

 

  if [type] == "accesslogs" {

    elasticsearch {

      index => "accesslogs"

  codec => "json"

      hosts => ["node01:9200", "node02:9200", "node03:9200"]

    }

  }

 

  if [type] == "gamelogs" {

    elasticsearch {

      index => "gamelogs1"

      codec => plain {

        charset => "UTF-16BE"

      }

      hosts => ["node01:9200", "node02:9200", "node03:9200"]

    }

  }

}

 bin/logstash -f conf/kafka-es.conf

4.修改basedir文件中任意数据即可产生es的index文件

好程序员大数据学习路线之Logstach与flume对比

5.网页数据存储在设置的/data/esdata中

6.在网页中查找指定字段

默认分词器为term,只能查找单个汉字,query_string可以查找全汉字

好程序员大数据学习路线之Logstach与flume对比


本文名称：好程序员大数据学习路线之Logstach与flume对比
文章起源：http://ybzwz.com/article/gdjsde.html

其他资讯