ELK log analysis system (example !!!)

ELK log analysis system overview

ELK是Elasticsearch、Logstash、Kibana的简称
Elasticsearch是实时全文搜索和分析引擎
Logstash是一个用来搜集、分析、过滤日志的工具
Kibana是一个基于Web的图形界面，用于搜索、分析和可视化存储在 Elasticsearch指标中的日志数据

Log Server

提高安全性
集中存放日志
缺陷：对日志的分析困难

---

ELK log analysis system

收集数据：LogstashAgent
建立索引：ElasticSearchCluster
数据可视化：KilbanaServer

Log Processing Step

 1.将日志进行集中化管理
 2.将日志格式化( Logstash )并输出到Elasticsearch
 3.对格式化后的数据进行索引和存储( Elasticsearch )
 4.前端数据的展示( Kibana )

Elasticsearch Introduction

Elasticsearch Overview

 提供了一个分布式多用户能力的全文搜索引擎

Elasticsearch concept

 接近实时
 集群
 节点
 索引：索引(库)-->类型(表)-->文档(记录)
 分片和副本

Logstash Introduction

 一款强大的数据处理工具，可以实现数据传输、格式处理、格式化输出
 数据输入、数据加工(如过滤，改写等)以及数据输出

LogStash major components

 Shipper
 Indexer
 Broker
 Search and Storage
 Web Interface

Kibana Kai绍

 一个针对Elasticsearch的开源分析及可视化平台
 搜索、查看存储在Elasticsearch索引中的数据
 通过各种图表进行高级数据分析及展示

The main function of Kibana

 Elasticsearch无缝之集成
 整合数据，复杂数据分析
 让更多团队成员受益
 接口灵活，分享更容易
 配置简单，可视化多数据源
 简单数据导出

ELK log analysis system deployment

Description of Requirement

 配置ELK日志分析群集
 使用Logstash收集日志
 使用Kibana查看分析日志

Installation Environment

 全部关闭防火墙
 Node1、Node2节点内存分配4G，Apache节点分配1G内存
 通过VMware虛拟网络Vmnet8连接

Host Name	IP addresses	The main software
Node1 server	192.168.142.152	Elasticsearch, Kibana
Node2 server	192.168.142.153	Elasticsearch
Apache server	192.168.142.155	Logstash

---

Step 1: Configure ES server node 1

1. Turn off the firewall and security features

systemctl stop firewalld.service  
setenforce 0

2. Modify the hostname configuration file

vim /etc/hosts
192.168.142.152 node1
192.168.142.153 node2

3. Remote Mount Resource Kit

mount.cifs //192.168.142.1/elk /mnt

4. Install Package

cd /mnt
rpm -ivh elasticsearch-5.5.0.rpm

5. Load System Services

systemctl daemon-reload

6. boot from the start Services

systemctl enable elasticsearch.service

7. Back up configuration files

cd /etc/elasticsearch/
cp /etc/elasticsearch/elasticsearch.yml /etc/elasticsearch/elasticsearch.yml.bak

8. Modify the main configuration file elasticsearch

vim /etc/elasticsearch/elasticsearch.yml

#第17行,取消注释并修改集群名称
cluster.name: my-elk-cluster

#第23行,取消注释修改节点名字
node.name: node1

#第33行,取消注释修改数据存放路径
path.data: /data/elk_data

#第37行,取消注释修改日志存放路径
path.logs: /var/log/elasticsearch/

#第43行,取消注释并修改,取消在不启动时锁定内存
bootstrap.memory_lock: false

#第55行,取消注释并修改地址,放入所有的地址(0.0.0.0代表所有地址)
network.host: 0.0.0.0

#第59行,取消注释,放开服务端口
http.port: 9200

#第68行,取消注释修改节点名称
discovery.zen.ping.unicast.hosts: ["node1", "node2"]

9. Create the database storage path

mkdir -p /data/elk_data

10. The authorization database storage path

chown elasticsearch:elasticsearch /data/elk_data/

11. Turn elasticsearch Service

systemctl start elasticsearch.service 

12. Review the status of port services

netstat -ntap | grep 9200
tcp6       0      0 :::9200                 :::*                    LISTEN      96970/java     

13. Install the build environment

yum install gcc gcc-c++ make -y

14. The decompression node node packages

cd /mnt
tar zxvf node-v8.2.1.tar.gz -C /opt

15. placement node

cd /opt/node-v8.2.1/
./configure

16. compile and install

make && make install

Step two: the front end of the frame mounting phantomjs

1. Unzip the package phantomjs

cd /mnt
tar jxvf phantomjs-2.1.1-linux-x86_64.tar.bz2 -C /usr/local/src 

2. Change directory to see phantomjs command

cd /usr/local/src/phantomjs-2.1.1-linux-x86_64//bin
ls
phantomjs

3. Copy the directory to the system directory

cp phantomjs /usr/local/bin/

Step 3: Install elasticsearch-head data visualization tools

1. Extract elasticsearch-head package

cd /mnt
tar zxvf elasticsearch-head.tar.gz -C /usr/local/src

2. Installation elasticsearch-head data visualization tools
CD / usr / local / the src / elasticsearch-head /
NPM the install

3. Modify the main configuration file elasticsearch

vim /etc/elasticsearch/elasticsearch.yml
#末行添加以下内容
http.cors.enabled: true
http.cors.allow-origin: "*"

4. Turn elasticsearch Service

systemctl restart elasticsearch.service

5. Start running in the background

npm run start &

6. Check the service port status

netstat -ntap | grep 9100

tcp        0      0 0.0.0.0:9100            0.0.0.0:*               LISTEN      50105/grunt         
[root@node1 elasticsearch-head]# netstat -ntap | grep 9200
tcp6       0      0 :::9200                 :::*                    LISTEN      96970/java   

---

Step Four: ES node configured as the server 2 of the node 1, the above-described operation is repeatedly performed to !!!

---

Step 5: Use your browser and enter the URL 192.168.142.152:9100, and connection address of another node, cluster check the health status

---

Step Six: Create Index

Back to the overview you can see the created index!

Step 7: Configure Apache server, install logstash collect log collection

#安装Apache服务
yum install -y httpd

#远程挂载资源包
mount.cifs //192.168.142.1/elk /mnt

#切换到挂载点
cd /mnt

#安装logstash
rpm -ivh logstash-5.5.1.rpm 

#开机自启动logstash服务
systemctl enable logstash.service

#启动logstash服务
systemctl start logstash.service

#建立命令软链接到系统
ln -s /usr/share/logstash/bin/logstash /usr/local/bin

#切换日志目录
cd /var/log

#授予他人读取权限
chmod o+r messages 

#查看权限
ll

#切入logstash配置目录
cd /etc/logstash/conf.d/

#编辑文件
vim system.conf

#写入以下内容,用以收集系统日志
input {
        file{
        path => "/var/log/messages"
        type => "system"
        start_position => "beginning"
        }
}
output {
        elasticsearch {
        #地址指向node1节点
        hosts => ["192.168.142.152:9200"]
        index => "system-%{+YYYY.MM.dd}"
        }
}

#重启服务
systemctl restart logstash.service

---

Step eight: View log information collected

---

Step 9: Back node1 node installation kibana

#切入挂载点
cd /mnt

#安装kibana
rpm -ivh kibana-5.5.1-x86_64.rpm 

#切入kibana目录
cd /etc/kibana/

#备份kibana.yml文件
cp kibana.yml kibana.yml.bak

#修改kibana.yml文件
vim kibana.yml

#取消第2行的注释,放开5601端口
server.port: 5601

#取消第7行的注释并修改地址,放入所有的地址(0.0.0.0代表所有地址)
server.host: "0.0.0.0"

#取消第21行的注释并指向node1节点的url
elasticsearch.url: "http://192.168.142.152:9200"

#取消第30行注释,放开kibana首页
kibana.index: ".kibana"

#启动kibana服务
systemctl start kibana.service

---

Step 10: Test kibana show log data, use the browser to access 192.168.142.152:5601

---

Step Eleven: All docking Apache Apache log file hosts (operating on the Apache server)

#编辑Apache日志配置文件
vim apache_log.conf

input {
        file{
        path => "/etc/httpd/logs/access_log"
        type => "access"
        start_position => "beginning"
        }
        file{
        path => "/etc/httpd/logs/error_log"
        type => "error"
        start_position => "beginning"
        }
}
output {
        if [type] == "access" {
        elasticsearch {
        hosts => ["192.168.142.152:9200"]
        index => "apache_access-%{+YYYY.MM.dd}"
        }
    }   
        if [type] == "error" {
        elasticsearch {
        hosts => ["192.168.142.152:9200"]
        index => "apache_error-%{+YYYY.MM.dd}"
        }
    }
}

#重启服务,稍等片刻!
logstash -f apache_log.conf

---

Step Twelve: Testing Apache logs show

That's all the ELK log analysis system, Thanks for reading !!!