不想寫python，而且我比較相信awk+uniq+sort的速度
|
1 |
awk -v date="$(date '+\%d/\%b/\%Y')" '($4 ~ date && $9 ~ /200/ ){print $1,$7}' /var/log/nginx/access.log | sort | uniq |awk '{print $2}'|sort|uniq -c|sort -r|head -n 50 |
- cron的
時區很奇怪，不會調，系統時間是+8，tail /var/log/syslog之後發現是西五區，手動+13小時可以通過修改localtime，修改完之後重啟service cron（還是重啟系統好了）參閱 -
12rm /etc/localtimecp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
- 第一步取出所有日期是今天且響應是200的請求，打印請求ip和請求uri
- 第二、三步對(ip uri)排序，去重，單個ip訪問單個頁面只會被記為一次
- 打印uri，去重，排序，顯示前50個
time計時
52M日誌，26w條
|
1 2 3 4 5 6 7 8 9 |
awk -v date="$(date '+%d/%b/%Y')" /var/log/nginx/access.log 0.14s user 0.01s system 92% cpu 0.159 total sort 0.00s user 0.00s system 0% cpu 0.156 total uniq -c 0.00s user 0.00s system 0% cpu 0.145 total sort 0.00s user 0.00s system 0% cpu 0.136 total awk '{print $3}' 0.00s user 0.00s system 0% cpu 0.137 total sort 0.00s user 0.00s system 0% cpu 0.125 total uniq -c 0.00s user 0.00s system 0% cpu 0.124 total sort -r 0.00s user 0.00s system 0% cpu 0.114 total head -n 50 0.00s user 0.00s system 0% cpu 0.113 total |
134M日誌，加上了兩個自定義條件，65w條
|
1 2 3 4 5 6 7 8 |
awk '($9 ~ /200/ && $7 !~ /#/){print $1,$7}' 0.70s user 0.09s system 53% cpu 1.457 total sort 0.03s user 0.01s system 2% cpu 1.530 total uniq -c 0.02s user 0.00s system 1% cpu 1.530 total awk '{print $3}' 0.02s user 0.00s system 1% cpu 1.541 total sort 0.02s user 0.00s system 1% cpu 1.563 total uniq -c 0.01s user 0.00s system 0% cpu 1.569 total sort -r 0.01s user 0.00s system 0% cpu 1.568 total head -n 50 0.00s user 0.00s system 0% cpu 1.567 total |
小站可以考慮