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
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)
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
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
