1、介绍
elk是实时日志分析平台,主要是为开发和运维人员提供实时的日志分析,方便人员更好的了解系统状态和代码问题。
2、elk中的e(elasticsearch):
(2.1)先安装依赖包,官方文档说明使用java1.8
yum -y install java-1.8.0-openjdk
安装elasticsearch:
tar zvxf elasticsearch-1.7.0.tar.gz
mv elasticsearch-1.7.0 /usr/local/elasticsearch
vim /usr/local/elasticsearch/config
cp elasticsearch.yml elasticsearch.yml.bak
vim elasticsearch.yml(修改)
cluster.name: elasticsearch node.name: syk node.master: true node.data: true index.number_of_shards: 5 index.number_of_replicas: 1(分片副本) path.data: /usr/local/elasticsearch/data path.conf: /usr/local/elasticsearch/conf path.work: /usr/local/elasticsearch/work path.plugins: /usr/local/elasticsearch/plugins path.logs: /usr/local/elasticsearch/logs bootstrap.mlockall: true (内存)
启动:/usr/local/elasticsearch/bin/elasticsearch -d
netstat -tlnp查看
会有9200与9300的java进程
curl http://192.168.137.50:9200
显示:
{ "status" : 200, "name" : "syk", "cluster_name" : "elasticsearch", "version" : { "number" : "1.7.0", "build_hash" : "929b9739cae115e73c346cb5f9a6f24ba735a743", "build_timestamp" : "2015-07-16T14:31:07Z", "build_snapshot" : false, "lucene_version" : "4.10.4" }, "tagline" : "You Know, for Search" }
(2.2)使用官方给的启动脚本:
https://codeload.github.com/elastic/elasticsearch-servicewrapper/zip/master
用rz命令传到服务器上
unzip elasticsearch-servicewrapper-master.zip
mv elasticsearch-servicewrapper-master/service/ /usr/local/elasticsearch/bin/
cd /usr/local/elasticsearch/bin/service
./elasticsearch install(在init.d下自动创建服务脚本)
/etc/init.d/elasticsearch restart
curl -XGET 'http://192.168.137.50:9200/_count?pretty' -d ' > { > "query":{ > "match_all":{} > } > } > '
会返回:
{ "count" : 0, "_shards" : { "total" : 0, "successful" : 0, "failed" : 0 } }
(2.3)基于rest api的界面(可以增删改差)
安装插件:/usr/local/elasticsearch/bin/plugin -i elasticsearch/marvel/latest (自动安装)
网页访问:http://192.168.137.50:9200/_plugin/marvel
安装集群管理插件
/usr/local/elasticsearch/bin/plugin -i mobz/elasticsearch-head
或者:https://github.com/mobz/elasticsearch-head/archive/master.zip下载下来,rz传到服务器
unzip elasticsearch-head-master.zip
mv elasticsearch-head-master plugins/head
网页访问:http://192.168.137.50:9200/_plugin/head
可以以网页的方式显示你的分片已分片副本。
3、elk中的l(logstash):
(3.1)安装logstash:
i)、官方提供了yum安装的安装方式:
1、rpm --import https://packages.elastic.co/GPG-KEY-elasticsearch
2、vim /etc/yum.repos.d/logstash.repo
添加:
[logstash-2.3] name=Logstash repository for 2.3.x packages baseurl=https://packages.elastic.co/logstash/2.3/centos gpgcheck=1 gpgkey=https://packages.elastic.co/GPG-KEY-elasticsearch enabled=1
3、yum --enablerepo=logstash-2.3 -y install logstash
ii)、下载tar包安装:
tar zvxf logstash-1.5.3.tar.gz
mv logstash-1.5.3 /usr/local/logstash
(3.2)测试
/usr/local/logstash/bin/logstash -e 'input { stdin{} } output { stdout{codec => rubydebug} }'
输入hehe
显示:
Logstash startup completed hehe { "message" => "hehe", "@version" => "1", "@timestamp" => "2016-08-07T17:46:10.836Z", "host" => "web10.syk.com" }
这表示正常。
(3.3)写logstash配置文件
注意:
必须input{}与output{}
写法:符号使用=>
vim /etc/logstash.conf
input{ file { path => "/var/log/syk.log" } } output{ file { path => "/tmp/%{+YYYY-MM-dd}.syk.gz" gzip => true } }
启动logstash:/usr/local/logstash/bin/logstash -f /etc/logstash.conf
cd /var/log
cat maillog >> syk.log(追加到syk.log里)
在/tmp下可以看到以日期命名的syk.gz压缩文件
(3.4)使用redis存储logstash:
yum -y install redis(redis放在另外一台服务器上)
vim /etc/redis.conf(修改)
bind 192.168.137.52
在192.168.137.52服务器上也安装logstash
编写配置文件:
vim /etc/logstash.conf
input{ file { path => "/var/log/syk.log" } } output{ redis { data_type => "list" key => "system-messages" host => "192.168.137.52" port => "6379" db => "1" } }
启动52服务器的logstash:
/usr/local/logstash/bin/logstash -f /etc/logstash.conf
cd /var/log
cat maillog >> syk.log(追加到syk.log里)
进去redis里查看:
redis-cli -h 192.168.137.52 -p 6379 select 1 keys *(可以看到system-messages这个key) llen system-messages(可以看大system-messages这个key的长度)
(3.4)将logstash收集的日志信息传到es上
在192.168.137.50的服务器上写logstash配置文件:
vim /etc/logstash.conf
input { redis { data_type => "list" key => "system-messages" host => "192.168.137.52" port => "6379" db => "1" } } output { elasticsearch { host => "192.168.137.50" protocol => "http" index => "system-messages-%{+YYYY.MM.dd}" } }
启动logstash:
/usr/local/logstash/bin/logstash -f /etc/logstash.conf
这时我们去看redis的LLEN system-messages,会发现已经变成了0,这说明数据已经传输到es上了。
网页访问:http://192.168.137.50:9200/_plugin/head/
会多出来一个system-messages-2016.08.07的分片副本
4、elk中的k(kibana):
(4.1)安装:
解压 mv就行
cd /usr/local/kiabna/config/
vim kibana.yml修改:
elastcsearch: "http://192.168.137.50:9200"
启动:
nohup ./bin/kiban &(默认端口5601)
网页访问:
http://192.168.137.50:5601
相关操作需要配合图片说明,这里暂时不说了。