News

  1. Home
  2. All posts
All posts, Linux by Team Ajara

script to find load average on the OS Level

script to find load average on the OS Level.

Whenever there is any performance issue in a  database  or server level ,it is very common to verify what is the load on the server level.

Load average in Linux OS server level can be retrieved by using uptime command.

By doing this we will get the process details of  any other database which  is consuming/eating resources like CPU and memory and it was not noticed by respective application team or the DBAs.

Based the process details ,one can take necessary action .

In this article ,we will see how to get load average and process details automatically to our mail  and how to run manually to get the output instantly.

Below commands and options have been used for preparing the script .
uptime
grep -o
awk ‘{ print $3 }’
cut -f1 -d .
ps -eo
head
coulmn -t
mailx

Script Details

Complete code is given below.

# Purpose:Script is used for checking load average and send mail 

#!/bin/bash

#script directory :/u01/app/oracle/scripts/OS

#Script name : load_average.sh

#Log file: /u01/app/oracle/scripts/OS/loadaverage.log

###

SCRIPT_DIR=/u01/app/oracle/scripts/OS

LOGFILE=loadaverage.log

LOG=${SCRIPT_DIR}/$LOGFILE

HOST=`hostname`

echo “+++ Load Average ( If load is more than 10 ) +++” >> $LOG

a=`uptime | grep -o ‘load average[s:][: ].*’ | awk ‘{ print $3 }’ | cut -f1 -d .`

if [ ${a} -ge 10 ]
then
echo “Host ${HOST}: Consuming High Load – ${a}” >> $LOG
printf ‘\n’
echo “Getting High CPU Processes——-” >> $LOG
print ‘\n’
ps -eo pid,ppid,cmd,%mem,%cpu –sort=-%cpu | head | column -t >> $LOG
mailx -s “RED Alert: ${HOST} High Load” dba@ajara.tech < $LOG
else
echo “Load Under Threshold. No Action Needed”
fi

rm $LOG

Parameters to be passed

None

How to run this script manually ?

sh /u01/app/oracle/scripts/OS/load_average.sh

Sample output

+++ Load Average ( If load is more than 10 ) +++
Host orasrv01: Consuming High Load – 19 Getting High CPU Processes——-
PID PPID CMD %MEM %CPU
121632 1 ora_j004_ORCLP 0.0 98.8
65562 31686 oracleORCLP (DESCRIPTION= 0.0 84.9
55061 31686 oracleORCLP (DESCRIPTION= 0.0 84.1
60944 31686 oracleORCLP (DESCRIPTION= 0.0 83.8
59012 31686 oracleORCLP (DESCRIPTION= 0.0 83.7
63094 31686 oracleORCLP (DESCRIPTION= 0.0 83.7
57027 31686 oracleORCLP (DESCRIPTION= 0.0 83.6
46888 31686 oracleORCLP (DESCRIPTION= 0.0 83.5
42365 31686 oracleORCLP (DESCRIPTION= 0.0 83.3

 

crontab entry

*/5 * * * * /u00/app/oracle/scripts/OS/load_average.sh > /u00/app/oracle/scripts/OS/load_average.log

Note : Or we can also write as below . Meaning of these two options is to run a Cron Job Every 5 Minutes. But  typing the whole list can be tedious and prone to errors.

0,5,10,15,20,25,30,35,40,45,50,55 * * * * /u00/app/oracle/scripts/OS/load_average.sh > /u00/app/oracle/scripts/OS/load_average.log
Note : Do not forget to give execute permission before we put the enty in crontab.

 

See also: