ELK相关常用配置解析

一、filebeat配置采集多个目录的日志

采集多个目录日志，自己的配置：

- type: log

  enabled: true
  
  # Paths that should be crawled and fetched. Glob based paths.
  paths:
    - /opt/nginx-1.14.0/logs/stars/star.access.log  #指明读取的日志文件的位置
  tags: ["nginx-access"]    #使用tag来区分不同的日志

- type: log

  enabled: true
  
  # Paths that should be crawled and fetched. Glob based paths.
  paths:
    - /srv/www/runtime/logs/app.log
  tags: ["app"]    #使用tag来区分不同的日志

二、filebeat解析多行日志合并成一行

收集日志的时候，如果是像自己的项目日志，往往是多行trace日志，这个时候，就需要配置多行匹配，filebeat提供了multiline ooptions用来解析多行日志合并成一行。

multiline options 主要是三个主要参数:

multiline.pattern: '^[0-9]{4}-[0-9]{2}-[0-9]{2}'
multiline.negate: true
multiline.match: after

//上面配置的意思是：不以时间格式开头的行都合并到上一行的末尾（正则写的不好，忽略忽略）

所以，最终，想要配置收集不同目录的日志，并且多行日志匹配，具体配置为：

- type: log

  enabled: true
  
  # Paths that should be crawled and fetched. Glob based paths.
  paths:
    - /opt/nginx-1.14.0/logs/stars/star.access.log
  tags: ["nginx-access"]

- type: log

  enabled: true
  
  # Paths that should be crawled and fetched. Glob based paths.
  paths:
    - /srv/www/runtime/logs/app.log
  tags: ["app"] 
  multiline.pattern: '^[0-9]{4}-[0-9]{2}-[0-9]{2}'
  multiline.negate: true
  multiline.match: after

关键在于使用filelds中的type进行不同日志的区分

filebeat服务重启：systemctl restart filebeat

参考链接：

https://www.cnblogs.com/zhjh2...

https://blog.csdn.net/m0_3788...

三、logstash配置grok的match，output等

logstash配置文件，对不同的日志进行不同的filter匹配

#Logstash通过type字段进行判断
input {
        beats {
                host => '0.0.0.0'
                port => 5401 
        }
}

filter {
  if "nginx-access" in [tags]{   #对nginx-access进行匹配
    grok {  #grok插件对日志进行匹配，主要是正则表达式
          match => { "message" => "%{IPORHOST:remote_ip} - %{IPORHOST:host} - \[%{HTTPDATE:access_time}\] \"%{WORD:http_method} %{DATA:url} HTTP/%{NUMBER:http_version}\" - %{DATA:request_body} - %{INT:http_status} %{INT:body_bytes_sent} \"%{DATA:refer}\" \"%{DATA:user_agnet}\" \"%{DATA:x_forwarded_for}\" \"%{DATA:upstream_addr}\" \"response_location:%{DATA:response_location}\"" }
       }
  }
  else if "app" in [tags]{ 
    grok {
      match => {
        "message" => "%{DATESTAMP:log_time} \[%{IP:remote_ip}\]\[%{INT:uid}\]\[%{DATA:session_id}\]\[%{WORD:log_level}\]\[%{DATA:category}\] %{GREEDYDATA:message_text}"
      }
    }
}

output{
  if "nginx-access" in [tags]{
    elasticsearch {
      hosts => ["http://xxx.xxx.xxx.xx:9200"]
      index => "star_nginx_access_index_pattern-%{+YYYY.MM.dd}"
      user => "elastic"
      password => "!@#j3C"
    }
  }
  else if "app" in [tags]{
    elasticsearch {
      hosts => ["http://xxx.xxx.xxx.xx:9200"]
      index => "star_app_index_pattern-%{+YYYY.MM.dd}"
      user => "elastic"
      password => "!@#j3C"
    }
  }
}

主要是对grok的正则表达式进行匹配，为了将日志一条一条的匹配出来然后在Kibana中展示

重启logstash：systemctl restart logstash