linux用shell脚本分析nginx访问日志，统计访问量失败次数及失败url

原文链接https://blog.csdn.net/enjoyphp/article/details/88636738
我这是先从总日志内把昨日日志切分到一个新文件，进行分析。可以设置定时任务

#!/bin/sh
year=`date +%Y`
month=`date +%b`
day=`date +'%d' -d "-1 days"`
date=`date +%Y%m%d -d "-1 days"`
echo $date
sdate=$day/$month/$year
echo $sdate
#cat  access.log  | awk '$4 >="[20/Aug/2019:00:00:00" && $4 <="[20/Aug/2019:23:59:59"'  > "$date"-access.log
cat  access.log | awk  '$4 >="'[$sdate:00:00:00'"  && $4 <="'[$sdate:23:59:59'"'  > "$date"-access.log
pv=`awk '{print $7}' "$date"-access.log | wc -l`
ip=`awk '{print $1}' "$date"-access.log | sort -n | uniq -c | wc -l`
success=`awk '{print $9}'  "$date"-access.log  | grep -E  "20\d?" | wc -l`
error=$((pv-success))
error_rate=$(printf "%.2f" `echo "scale=2;$error/$pv*100"|bc`)
status200=`awk '{print $9}'  "$date"-access.log  | grep -E  "200" | wc -l`
status204=`awk '{print $9}'  "$date"-access.log  | grep -E  "204" | wc -l`
status404=`awk '{print $9}'  "$date"-access.log  | grep -E  "404" | wc -l`
status422=`awk '{print $9}'  "$date"-access.log  | grep -E  "422" | wc -l`
status500=`awk '{print $9}'  "$date"-access.log  | grep -E  "500" | wc -l`
echo $error_rate
status404top10=`awk '{print $7,$9}' "$date"-access.log | grep -E  "404"  | sort |uniq -c | sort -rn | head -n 10`
status422top10=`awk '{print $7,$9}' "$date"-access.log | grep -E  "422"  | sort |uniq -c | sort -rn | head -n 10`
status500top10=`awk '{print $7,$9}' "$date"-access.log | grep -E  "500"  | sort |uniq -c | sort -rn | head -n 10`
echo $status404top10
echo -e "\n$date\n PV is $pv , IP is $ip , success is $success , error is $error , error rate is $error_rate% , status200 is $status200 , status204 is $status204 , status404 is $status404 , status422 is $status422 , status500 is $status500 \nstatus404 top10 page:\n$status404top10\nstatus422 top10 page:$status422top10\nstatus500 top10 page:$status500top10" >> analyse_access.log