因为业务需要要将线上elasticsearch-2.3升级到最新版本，官方称7.2.0版本比之前版本运行速度和效率有质的飞跃！
我们将安装elasticsearch-7.2.0配套的环境
一，首先去各自官网下载相应的文件：
elasticsearch-7.2.0
elasticsearch-head-master
kibana-7.2.0-linux-x86_64
logstash-7.2.0
elasticsearch-analysis-ik-7.2.0
node-v8.16.0-linux-x64.tar
在此我就不贴官网地址了，给大家提供个百度云盘我下载好的链接
链接：https://pan.baidu.com/s/1qUSDGHaIRHfjjyfyZ7A9Pg
提取码：e54v
二，安装前准备
1，因为elasticsearch不允许使用root账户启动，所以我们首先创建用户
useradd elk
2，更改系统资源限制
vim /etc/security/limits.conf
添加如下参数：
* soft nofile 65536
* hard nofile 65536
* soft nproc 2048
* hard nproc 4096
vim /etc/sysctl.conf
添加：
vm.max_map_count=655360
使用如下命令使参数生效
sysctl -p
三,安装部署
1，安装elasticsearch
因提供的都是二进制包，无需编译，可以直接使用
解压
tar -xvf elasticsearch-7.2.0
将文件放到合适的地方
mv elasticsearch-7.2.0 /home/elk/
将文件所有者更改为elk
chown -R elk.elk /home/elk/elasticsearch-7.2.0
编辑elasticsearch配置文件：
vim elasticsearch.yml
cluster.name: xavito
transport.tcp.compress: true
cluster.initial_master_nodes: ["vito248","vito203"]
discovery.seed_hosts: ["192.168.1.248", "192.168.1.203"]
node.name: vito203
http.port: 9200
bootstrap.memory_lock: false
bootstrap.system_call_filter: false
network.host: 0.0.0.0
http.cors.enabled: true
http.cors.allow-origin: "*"
其中我这是两个节点的集群，如果是单节点cluster参数就不用配置了，如果是集群要主要集群的写法，有别与之前的版本
写错了就会到导致服务起不来，或者后续elasticsearch-head-master在网页上连接不上。
同时也可以指定数据文件和日志文件路径，默认情况下在当前主文件下
#path.data: /path/to/data
#path.logs: /path/to/logs
直接用elk用户启动即可
su elk
/home/elk/elasticsearch-7.2.0/elasticsearch -d (-d为后台运行)
启动后可以用curl或者浏览其访问9200端口：
表示启动已经成功了
2，安装elasticsearch-head-master.zip
解压：
unzip elasticsearch-head-master.zip
mv elasticsearch-head-master /home/elk/
chown -R elk.elk /home/elk/elasticsearch-head-maste
安装：
首先必须有node.js，才可以安装，没有的话安装下，node只要在环境变量里配置好就行了
安装node如下:
将安装包上传到指定位置(我习惯放到：/usr/local/application/目录)，并解压
tar -xvf node-v10.6.0-linux-x64.tar.xz
重命名文件夹
1 mv node-v10.6.0-linux-x64 nodejs
通过建立软连接变为全局
ln -s /usr/local/application/nodejs/bin/npm /usr/local/bin/
ln -s /usr/local/application/nodejs/bin/node /usr/local/bin/
检查是否安装成功，命令：node-v
node -v
v10.6.0
已安装的上面步骤略过
cd /home/elk/elasticsearch-head-maste
执行 npm install
执行 npm run start (npm run start &)
netstat -nltp
看到9100端口开启，表示安装成功了
可以在浏览器登录
3、安装kibana-7.2.0-linux-x86_64.tar.gz，在7.2版本中官方直接支持中文，在配置文件中即可设置
解压
tar -xvf kibana-7.2.0-linux-x86_64.tar.gz
mv kibana-7.2.0-linux-x86_64 /home/elk/
chown -R elk.elk /home/elk/kibana-7.2.0-linux-x86_64
vim kibana.yml
在配置文件中添加如下参数：
server.host: "192.168.1.248"
i18n.locale: "zh-CN"
即可支持中文
4，安装elasticsearch-analysis-ik-7.2.0.zip中文分词器
unzip elasticsearch-analysis-ik-7.2.0.zip
chown -R elk.elk /home/elk/elasticsearch-analysis-ik-7.2.0
直接加载到elasticsearch的 plugins下
mv elasticsearch-analysis-ik-7.2.0 /home/elk/elasticsearch-7.2.0/plugins/ik
重新启动elasticsearch即可
5，安装logstash-7.2.0.tar.gz
解压：
tar -xvf logstash-7.2.0.tar.gz
mv logstash-7.2.0 /home/elk/
chown -R elk.elk /home/elk/ logstash-7.2.0
根据实际需要编辑配置文件：
例如：
vim format.cnf
input {
file {
path => "/mnt/new/tomcat/LOGS/WebLogs/user/*.log"
type => "userlog"
start_position => "beginning"
sincedb_path => "/root/logstash-7.2.0/sincedb"
}
file {
path => "/mnt/new/tomcat/LOGS/*/output/*.log"
type => "syslog"
codec => multiline {
pattern => "^\d"
negate => true
what => "previous"
}
start_position => "beginning"
sincedb_path => "/root/logstash-7.2.0/sincedb"
}
file {
path => "/mnt/new/tomcat/LOGS/*/pay/*.log"
type => "pay"
codec => multiline {
pattern => "^\d"
negate => true
what => "previous"
}
start_position => "beginning"
sincedb_path => "/root/logstash-7.2.0/sincedb"
}
}
filter{
if [type] == "userlog" {
grok {
patterns_dir => "/root/logstash-7.2.0/reg"
match => {
"message" => "%{logdatetime:times} %{MSG_:msg}"
}
}
json {
source => "msg"
target => "msg"
}
if [msg][uri]=="/base/msg/tips/queryMsgTipsByUser.htm"{
drop {}
}
if [msg][ip] !~ "^127\.|^192\.168\.|^172\.1[6-9]\.|^172\.2[0-9]\.|^172\.3[01]\.|^10\." {
geoip {
source => "[msg][ip]"
database => "/root/logstash-7.2.0/vendor/bundle/jruby/2.5.0/gems/logstash-filter-geoip-6.0.1-java/vendor/GeoLite2-City.mmdb"
target => "geoip"
fields => ["region_name","city_name","location"]
}
}
useragent {
source => "[msg][ua]"
target => "ua"
}
date {
locale => "cn"
match => ["times", "yyyy-MM-dd HH:mm:ss.SSS","UNIX"]
target => "@timestamp"
}
mutate {
remove_field => "@version"
remove_field => "message"
remove_field => "[msg][ua]"
remove_field => "times"
}
}
if [type] == "syslog" {
grok {
patterns_dir => "/root/logstash-7.2.0/reg"
match=>["message","%{logdatetime:times} .*%{LOGLEVEL:loglevel } {0,4}%{CLASSNAME:classname} {0,4}- {0,4}%{INFOMSG:infomsg}"]
}
if [classname] =~"^com.alibaba.dubbo.config.AbstractConfig" {
drop {}
}
if [classname] =~"^org.springframework.web.servlet.mvc.annotation.DefaultAnnotationHandlerMapping" {
drop {}
}
if [classname] =~"^o.springframework.web.servlet.mvc.method.annotation.RequestMappingHandlerMapping" {
drop {}
}
if [loglevel] == "DEBUG" {
drop {}
}
date {
locale => "cn"
match => ["times", "yyyy-MM-dd HH:mm:ss.SSS","UNIX"]
target => "@timestamp"
}
mutate {
remove_field => "@version"
remove_field => "times"
remove_field => "message"
remove_field => "tags"
}
}
if [type] == "pay" {
grok {
patterns_dir => "/root/logstash-7.2.0/reg"
match=>["message","%{logdatetime:times} .*%{LOGLEVEL:loglevel } {0,4}%{CLASSNAME:classname} {0,4}- {0,4}%{INFOMSG:infomsg}"]
}
date {
locale => "cn"
match => ["times", "yyyy-MM-dd HH:mm:ss.SSS","UNIX"]
target => "@timestamp"
}
mutate {
remove_field => "@version"
remove_field => "times"
remove_field => "message"
remove_field => "tags"
}
}
}
output {
elasticsearch {
hosts => ["192.168.1.248:9200","192.168.1.203:9200"]
codec => "json"
index => "log-%{type}-%{+YYYYMM}"
manage_template => true
template_overwrite => true
template_name => "my_logstash"
template => "/root/logstash-7.2.0/bin/logstash.json"
}
}
启动：
/home/elk/ logstash-7.2.0/bin/logstash -f /path/format.cnf
到此安装完成！