shell web monitoring script, the whole set is in it, decision rules changed several times, multiple combinations can see the old version, follow-up lazy to write new.
Old version links: https://blog.51cto.com/junhai/2407485
This time there are three scripts, write down the process of implementation.
The effect of calculating failure time is as follows: https://blog.51cto.com/junhai/2430313
Using Wechat Robot: https://blog.51cto.com/junhai/2424374
Because the useless database can only use 3 txt like this, in fact, I will not... --! Multiple sed can be written in one sentence, and they are too lazy to take the time to optimize.
url.txt > url. del (store unreachable address, second script analysis after) > url. add (third script analysis whether the address is restored)
The monitoring effect is as follows:
For external use, alarms are sent by mail, because too many web s monitored by the government cloud need to send mail. Mailboxes are often considered to be for sending spam, and are blocked many times.
One alarm and one recovery, which prevents multiple alarms, only two emails will be sent out in one failure, and the third script control will be used for WEB monitoring of government cloud platform.
Three unreachable stop alarms, one recovery alarm, alarm @ related system personnel, using the second script control, for Changwei internal system monitoring alarm. Mail is optional, internal use of enterprise Wechat is faster.
#!/bin/bash
#20190914
#QQ450433231
. /root/weixin.sh
cur_time(){
date "+%Y/%m/%d %H:%M:%S"
}
systemname(){
name=`cat url2.txt|grep $url|wc -l`
if [ $name -eq 1 ];then
cat url2.txt|grep -w $url
else
echo "$url"
fi
}
[ ! -f /root/url.txt ] && echo "url.txt file does not exist" && exit 1
sed -i '/^$/d' url.txt
while read url
do
[ -z $url ] && echo "url.txt Existence of spaces Check file format" && exit 1
for ((i=1;i<7;i++))
do
rule=`curl -i -s -k -L -m 10 $url|grep -w "HTTP/1.1 200"|wc -l`
if [ $rule -eq 1 ];then
echo "$(cur_time) The first $i Secondary inspection $url Successful Web Access" >> check.log
break
elif [ $i = 5 ];then
echo $url >> url.del
info=`echo -e "$(cur_time) \n$(systemname) Web page can not be accessed, please check!"`
wx_web
echo -e "$(cur_time) \n\n$(systemname) Web page can not be accessed, please check!"|mail -s "[Important Alert] Web page unavailable" 450433231@qq.com
echo "$(cur_time) $(systemname) Webpage $(expr $i \* 3)Seconds can not be accessed, please check!" >> checkfail.log
sh /root/checkdel.sh #The second script
else
echo "$(cur_time) The first $i Secondary inspection $url Web page access failure" >> checkfail.log
sleep 3
fi
done
done < /root/url.txt
sh /root/checkadd.sh #The third script. I like to have the third script separately set at intervals of 1 minute.#!/bin/bash
. /root/weixin.sh
cur_time(){
date "+%Y/%m/%d %H:%M:%S"
} #Display time
sed -i '/^$/d' url.del
sed -i '/^$/d' url.delout
cat url.del|sort|uniq -c >> url.delout
while read line
do
i=`echo $line|awk '{print$1}'`
newurl=`echo $line|awk '{print$2}'`
if [ -z $newurl ];then
continue
elif [ $i -eq 1 ];then
echo $newurl >> url.add
sed -i "s|$newurl||" url.txt
sed -i "s|$newurl||" url.del
sed -i '/^$/d' url.txt
sed -i '/^$/d' url.del
#info=`echo-e'$(cur_time)\n $i alarm site $newurl not restored, suspend sending alarms' ___________.`
#wx_web
#echo'$(cur_time) $i warning website $newurl not restored suspend sending alarm'| mail-s' [suspend alarm]"450433231@qq.com
#echo "$(cur_time) website $newurl not restored suspended alarm" > checkfail.log
sed -i "s|$newurl||" url.delout
sed -i '/^$/d' url.delout
else
echo > url.del
continue
fi
done< url.delout
echo > url.delout#!/bin/bash
. /root/weixin.sh
. /root/downtime.sh
cur_time(){
date "+%Y/%m/%d %H:%M:%S"
}
systemname(){
name=`cat url2.txt|grep $url|wc -l`
if [ $name -eq 1 ];then
cat url2.txt|grep $url|awk '{print$1,$2}'
else
echo "$url"
fi
}
sed -i '/^$/d' url.add
while read url
do
[ -z $url ] && echo "url.del Existence of spaces Check file format" && exit 1
rule=`curl -i -s -k -L -m 10 $url|grep -w "HTTP/1.1 200"|wc -l`
if [ $rule -eq 1 ];then
u=$(grep -w "$url" url.txt|wc -l)
if [ $u -eq 1 ];then
echo "$url Already exist" >> checkfail.log
else
echo "$url" >> url.txt
fi
sed -i "s|$url||" url.add
sed -i '/^$/d' url.add
echo "$(cur_time) $url Monitoring has been restored" >> checkfail.log
info=`echo -e "$(cur_time) \n$(systemname) The website has been restored \n$(fail_time)"`
wx_ah
echo -e "$(cur_time) $(systemname) The website has been restored \n$(fail_time)"|mail -s "[Website Restoration]" 450433231@qq.com
else
echo "$(cur_time) inspect $url Web page access failure" >> /dev/null #checkfail.log
fi
done< url.addLater, I wrote a silly interactive script, which will be used to add and delete the sh script adds to the monitoring website to add the monitoring website in batches.
#!/bin/bash
#20190831
case $1 in
add )
[ -z $2 ] && echo "Please be there. add Later input to increase the monitoring site address Usage: add baidu.com" && exit 1
u=$(grep -w "$2" url.txt|wc -l)
if [ $u -eq 1 ];then
echo "$2 Website already exists"
else
echo $2 >> url.txt
echo "$2 Website has been added to the monitoring list"
fi
;;
del )
[ -z $2 ] && echo "Please be there. del Later enter the site address to remove the monitoring Usage: del baidu.com" && exit 1
u=$(grep -w "$2" url.txt|wc -l)
if [ $u -eq 1 ];then
sed -i "s|$2||" url.txt
sed -i '/^$/d' url.txt
echo "$2 Website deleted"
else
echo "$2 Website not found"
fi
;;
update )
vi url2.txt
;;
dis )
cat url.txt|sort|uniq -c
;;
back )
cp url.txt url.bk
echo "Monitoring List Backup File url.bk"
;;
disuniq )
cp url.txt url.bk
cat url.txt|sort|uniq -c|awk '{print$2}' > url.new
cat url.new > url.txt
rm -rf url.new
sed -i '/^$/d' url.txt
echo "Monitoring List Re-Completion Backup file url.bk"
;;
* )
echo "-----------`date`-----------------"
echo "sh $0 add Increase monitoring websites"
echo "sh $0 del Delete monitoring websites"
echo "sh $0 update Modify the contact information of the website"
echo "sh $0 dis Display monitoring list"
echo "sh $0 back Backup monitoring list"
echo "sh $0 disuniq Monitoring list de-duplication"
echo "----------------------------------"
;;
esacIncidentally, the usual crontab call frequency is used to read the old version of the blog.
*/5 * * * * sh /root/check.sh */1 * * * * sh /root/checkadd.sh 0 0 * * * echo > url.del