1. 文本命令數據分析

   假設線上倒出的接口訪問日誌有上百行，該日誌的記錄格式如下：

   /data1/www/logs/archives/170524/170524.v6.weibo.com_10.72.13.113.0.cn.gz:v6.weibo.com 123.125.104.20 0.016s - [24/May/2017:14:04:37 +0800] "POST /aj/video/playstatistics?ajwvr=6&cuid=2008282113&lang=zh-cn&ip=60.255.47.150&curl=http%3A%2F%2Fd.weibo.com%2F%3Ftopnav%3D1%26amp%3Bmod%3Dlogo%26amp%3Bwvr%3D6&ua=Mozilla%2F5.0%20%28Windows%20NT%205.1%29%20AppleWebKit%2F537.36%20%28KHTML%2C%20like%20Gecko%29%20Chrome%2F49.0.2623.221%20Safari%2F537.36%20SE%202.X%20MetaSr%201.0&wvr=v5 HTTP/1.1" 200 71 "http://zhaoren.weibo.com" - "SUP=- SUBP=-" "REQUEST_ID=1000659645207911167" "Weibo.com Swift framework HttpRequest class" "REQ_UID=2008282113"

   統計日誌中根據ip進行排重，並統計相同ip統計的次數，執行命令如下：

   cat play.log | awk -F ‘ ‘ ‘{print $2}‘ | sort -k 1 -n -r | uniq -c > rizhi.log

   說明：每行以空格分割，輸出第二個參數，並根據第一行排序，－n數字排序－r降序，並統計每行在文本中出現的次數，輸出結果如下：

   
   

   3 223.166.87.59

1 60.12.35.5

1 1.189.96.233

本文出自 “PHP程序猿” 博客，請務必保留此出處http://okowo.blog.51cto.com/4923464/1929992

linux下awk日誌分析