elk分析nginx、dns日誌
部署環境
192.168.122.187 | Logstash-1.5.1 elasticsearch-1.6.0 kibana-4.1.1 | Centos6.4 |
192.168.122.1 | Redis-2.8 | Centos7.1 |
192.168.122.2 | Centos6.4 | |
192.168.122.247 | Bind9 logstash-1.5.2 supervisor-2.1-9 java-1.7 | Centos6.2 |
安裝過程就不復述了，參考http://kibana.logstash.es/content/logstash/get_start/install.html
安裝時注意的幾個地方
1、java最好是1.7
2、server上的logstash我直接用rpm裝的就能用，但是agent端的就不好使，沒有深究
3、elasticsearch、kibana還有agent端的logstash我都是用supervisor運行的
4、supervisor直接就是epel的yum裝的
貼下配置
192.168.122.187上：
Logstash的配置
server端的logstash是rpm安裝的
[root@c6test ~]# cat /etc/logstash/conf.d/central.conf
input { redis {host => "192.168.122.1"
port => 6379
type => "redis-input"
data_type => "list"
key => "logstash"
codec => 'json'
}
}
output { elasticsearch {host => "127.0.0.1"
}
}
elasticsearch
/usr/local/elasticsearch-1.6.0/config/elasticsearch.yml保持默認
Kibana
/usr/local/kibana-4.1.1-linux-x64/config/kibana.yml 保持默認
192.168.122.1上
Redis的配置也沒動。。。
192.168.122.2上
Nginx的
#nginx這裏的區別就是log這塊的配置，配成json格式
log_format json '{"@timestamp":"$time_iso8601",''"host":"$server_addr",'
'"clientip":"$remote_addr",'
'"size":$body_bytes_sent,'
'"responsetime":$request_time,'
'"upstreamtime":"$upstream_response_time",'
'"upstreamhost":"$upstream_addr",'
'"http_host":"$host",'
'"url":"$uri",'
'"xff":"$http_x_forwarded_for",'
'"referer":"$http_referer",'
'"agent":"$http_user_agent",'
'"status":"$status"}';
-----------------------------
access_log /var/log/nginx/zabbix_access.log json;
logstash的
[root@zabbixproxy-005002 ~]# cat /usr/local/logstash-1.5.2/conf/shipper.conf
input { file {type => "test-nginx"
path => ["/var/log/nginx/zabbix_access.log"]
codec => "json"
}
}
output { stdout {} redis {host => "192.168.122.1"
port => 6379
data_type => "list"
key => "logstash"
}
}
Supervisor的
[root@zabbixproxy-005002 ~]# cat /etc/supervisord.conf |grep -v \;
[supervisord]
[program:logstash]
command=/usr/local/logstash-1.5.2/bin/logstash agent --verbose --config /usr/local/logstash-1.5.2/conf/shipper.conf --log /usr/local/logstash-1.5.2/logs/stdout.log
process_name=%(program_name)s
numprocs=1
autostart=true
autorestart=true
startretries=5
exitcodes=0
stopsignal=KILL
stopwaitsecs=5
redirect_stderr=true
[supervisorctl]
192.168.122.247上
Bind的配置用默認的即可
Logstash的
[root@sys-247245 ~]# cat /usr/local/logstash/conf/shipper.conf
input { file {type => "dnslog"
path => ["/home/dnslog/*.log"]
}
}
filter {#由於dns日誌沒辦法定義成json，我又不會grok，所以這裏用mutate來切割
mutate {gsub => ["message","#"," "]
split => ["message"," "]
}
mutate { add_field => { "client" => "%{[message][5]}" "domain_name" => "%{[message][10]}" "server" => "%{[message][14]}"}
}
}
output { stdout {} redis {host => "192.168.122.1"
port => 6379
data_type => "list"
key => "logstash"
}
}
Supervisor的
[root@sys-247245 ~]# cat /etc/supervisord.conf |grep -v \;|grep -v ^$
[supervisord]
[supervisorctl]
[program:logstash]
command=/usr/local/logstash/bin/logstash agent --verbose --config /usr/local/logstash/conf/shipper.conf --log /usr/local/logstash/logs/stdout.log
process_name=%(program_name)s
numprocs=1
autostart=true
autorestart=true
startretries=5
exitcodes=0
stopsignal=KILL
stopwaitsecs=5
redirect_stderr=true
配置kibana
Nginx
1、在discover搜索nginx相關的日誌，之後保存
2、在visualize部署單個的圖表，之後保存
3、在dashboard將幾個nginx的visualize的圖表連起來
Dns
1、在discover搜索dns相關的日誌，之後保存
2、在visualize部署單個的圖表，之後保存
3、在dashboard將幾個dns的visualize的圖表連起來
遇到的問題
自定義的field在discover上能看到，但是在製作visualize時看不到
這種情況是由於沒有刷新索引的field導致的，默認的索引用的是logstash-*，在“Settings”—Indices中看到，點擊logstash-*進去之後，點擊刷新按鈕