Today's Question:  What are bad parts of Windows 10?       GIVE A SHOUT

Technical Article => Operating System =>  Linux/Unix

Server monitoring shell scripts

 sonic0002     2013-08-30 21:56:14      6,615    0    8

There are many open source server monitoring software such as cacti and nagios. Besides these, can we write our own shell scripts to monitor them? The shell scripts written by ourselves can fulfill our special requirements better and have a more detailed coverage.

Below are some commonly used shell scripts used by Evangelist, an Oracle DBA.

1. Check network gateway traffic

#!/bin/bash
#network
#Mike.Xu
while : ; do
      time='date +%m"-"%d" "%k":"%M'
      day='date +%m"-"%d'
      rx_before='ifconfig eth0|sed -n "8"p|awk '{print $2}'|cut -c7-'
      tx_before='ifconfig eth0|sed -n "8"p|awk '{print $6}'|cut -c7-'
      sleep 2
      rx_after='ifconfig eth0|sed -n "8"p|awk '{print $2}'|cut -c7-'
      tx_after='ifconfig eth0|sed -n "8"p|awk '{print $6}'|cut -c7-'
      rx_result=$[(rx_after-rx_before)/256]
      tx_result=$[(tx_after-tx_before)/256]
      echo "$time Now_In_Speed: "$rx_result"kbps Now_OUt_Speed: "$tx_result"kbps"
      sleep 2
done

2. System monitoring

#!/bin/sh
#systemstat.sh
#Mike.Xu
IP=192.168.1.227
top -n 2| grep "Cpu" >>./temp/cpu.txt
free -m | grep "Mem" >> ./temp/mem.txt
df -k | grep "sda1" >> ./temp/drive_sda1.txt
#df -k | grep sda2 >> ./temp/drive_sda2.txt
df -k | grep "/mnt/storage_0" >> ./temp/mnt_storage_0.txt
df -k | grep "/mnt/storage_pic" >> ./temp/mnt_storage_pic.txt
time=`date +%m"."%d" "%k":"%M`
connect=`netstat -na | grep "219.238.148.30:80" | wc -l`
echo "$time  $connect" >> ./temp/connect_count.txt

3. Monitor disk space. mail will be sent when the usage is over 90%

#!/bin/bash
#monitor available disk space
SPACE='df | sed -n '/ \ / $ / p' | gawk '{print $5}' | sed  's/%//'
if [ $SPACE -ge 90 ]
then
fty89@163.com
fi

4. Monitor CPU and memory usage

#!/bin/bash
#script  to capture system statistics
OUTFILE=/home/xu/capstats.csv
DATE='date +%m/%d/%Y'
TIME='date +%k:%m:%s'
TIMEOUT='uptime'
VMOUT='vmstat 1 2'
 USERS='echo $TIMEOUT | gawk '{print $4}' '
LOAD='echo $TIMEOUT | gawk '{print $9}' | sed "s/,//' '
FREE='echo $VMOUT | sed -n '/[0-9]/p' | sed -n '2p' | gawk '{print $4} ' '
IDLE='echo  $VMOUT | sed -n '/[0-9]/p' | sed -n '2p' |gawk '{print $15}' '
echo "$DATE,$TIME,$USERS,$LOAD,$FREE,$IDLE" >> $OUTFILE

5. Full system check

#!/bin/bash
# check_xu.sh
# 0 * * * * /home/check_xu.sh
 
DAT="`date +%Y%m%d`"
HOUR="`date +%H`"
DIR="/home/oslog/host_${DAT}/${HOUR}"
DELAY=60
COUNT=60
# whether the responsible directory exist
if ! test -d ${DIR}
then
        /bin/mkdir -p ${DIR}
fi
# general check
export TERM=linux
/usr/bin/top -b -d ${DELAY} -n ${COUNT} > ${DIR}/top_${DAT}.log 2>&1 &
# cpu check
/usr/bin/sar -u ${DELAY} ${COUNT} > ${DIR}/cpu_${DAT}.log 2>&1 &
#/usr/bin/mpstat -P 0 ${DELAY} ${COUNT} > ${DIR}/cpu_0_${DAT}.log 2>&1 &
#/usr/bin/mpstat -P 1 ${DELAY} ${COUNT} > ${DIR}/cpu_1_${DAT}.log 2>&1 &
# memory check
/usr/bin/vmstat ${DELAY} ${COUNT} > ${DIR}/vmstat_${DAT}.log 2>&1 &
# I/O check
/usr/bin/iostat ${DELAY} ${COUNT} > ${DIR}/iostat_${DAT}.log 2>&1 &
# network check
/usr/bin/sar -n DEV ${DELAY} ${COUNT} > ${DIR}/net_${DAT}.log 2>&1 &
#/usr/bin/sar -n EDEV ${DELAY} ${COUNT} > ${DIR}/net_edev_${DAT}.log 2>&1 &

You can create a cron job to run automatically:

0 * * * * /home/check_xu.sh

There will be CPU,memory,network and IO statistic data in /home/oslog/host_yyyymmdd/hh each hour. So if thee are issues on the server at some time, you can check the logs directly for that time.

 Tags : Server monitor,CPU,memory,shell   Read(6,615) Comment(0) Version(8)

  SAVE AS PDF   MARK AS READ   MARK AS IMPORTANT

 RELATED


 COMMENT  (TOTAL 0 COMMENT)


No comment for this article.


Nickname  
Email 
Comment

Is this car fully tested?

By sonic0002
Is this car fully tested? Or is it doing the Ice Bucket Challenge for ALS?

 CONTRIBUTE


Share your geeky stuff now.
You make the world better, the world makes us better.

 WRITE ARTICLE

 VERSION


 VIEW ALL