Bellow I will present script checking Load Average on a Linux server which will send report if it becomes to high.
It is using Linux command ‘uptime’ which is pulling out server’s uptime as well as its Load Average:
-
# uptime
-
21:53:51 up 43 days, 2:19, 3 users, load average: 1.63, 0.70, 8.29
In my script bellow I am using not the last minute load average, but the last five minutes (the second) one.
If you are not aware you can interpret a load average of “1.63, 0.70, 7.89″ on a single-CPU system as:
– during the last minute, the system was overloaded by 63% on average (1.63 runnable processes, so that 0.73 processes had to wait for a turn for a single CPU system on average).
– during the last 5 minutes, the CPU was idling 30% of the time on average.
– during the last 15 minutes, the system was overloaded 698% on average (7.89 runnable processes, so that 6.98 processes had to wait for a turn for a single CPU system on average).
I have chosen the five minute interval as sending mails every minute is too aggressive in case of a server load. Also it could be something too short and handled by the server without notification.
-
-
#! /bin/bash
-
# Here you set the maximum Load Average you want to be notified on.
-
# It depends on the number of processes and the usual server load.
-
max_loadavge=3
-
-
# In case of load average: 3.63, 5.70, 3.29 – will extract 5.70
-
loadavg=$(uptime | awk -F'load average:' '{print $2}'|awk '{gsub(",",""); print $2}'
-
-
# Prints the load average with current time stamp into log file.
-
# It will save according to the
-
# You may comment this line on case you do not need it
-
echo "Load Average:$loadavg on `date`" >> /var/log/load.log
-
-
# Set the e-mail you will be notified here.
-
EMAIL1="email@domain.com"
-
-
Optional secondary e-mail
-
#EMAIL2="email2@domain.com"
-
-
#Checks if the maximum Load Average is smaller then the one from the 'uptime' command
-
if [[ "$loadavg" > "$max_loadavge" ]]
-
then
-
-
# Creates notification message subject
-
SUBJECT="`hostname` LOAD AVERAGE ALERT $loadavg(>$max_loadavge)"
-
-
#Both lines bellow will be the body of the message. It will contain the current load average with time stamp as well as list of
-
#the current processes running on the server
-
ETEXT1="Load Average:$loadavg on `date`\n\n"
-
ETEXT2="USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND\n`ps aux | sort -rgk3`"
-
-
echo -e "$ETEXT1$ETEXT2" | mail -s "$SUBJECT" "$EMAIL1"
-
-
# if you are going to send to more than one email the line will be:
-
# echo -e "$ETEXT1$ETEXT2" | mail -s "$SUBJECT" "$EMAIL1" "$EMAIL2"
-
-
##Here also you can execute scripts for restarting particular services e.g. Apache, MySQL
-
-
fi
-
-
exit
* Note that some settings may need tuning because of changed/different command output
Once the script is ready you can set it as Cron. job Mine is set to check every 5th minute:
-
# .—————- minute (0 – 59)
-
# | .————- hour (0 – 23)
-
# | | .———- day of month (1 – 31)
-
# | | | .——- month (1 – 12) OR jan,feb,mar,apr …
-
# | | | | .—- day of week (0 – 6) (Sunday=0 or 7) OR sun,mon,tue,wed,thu,fri,sat
-
# | | | | |
-
# * * * * * user-name command to be executed
-
*/5 * * * * /root/scrpts/health.check.sh &> /dev/null
