Riccardo Riva

The following is a simple script to monitor a Software Raid configuration on a Linux System.
It execute a /proc/mdstat check to search a ‘blocks_’ occurency which indicates problems on the Raid system and in case of match it notify the system administrator with a mail message.

It should be very useful for non-presidiated system, even if I always prefer Hardware Raid.

#!/bin/bash
#
# Script created by Riccardo Riva
# http://www.riccardoriva.com
#
# It check a Software Raid subsystem and notify by mail any occurency problem

# Define variable
LOG_FILE=/tmp/raid-check.log
SYSTEM=`uname --nodename`
MAILTO='systemadmin@mail.exp'

# Checking /proc/mdstat
cat /proc/mdstat | grep 'blocks.*_' > $LOG_FILE

# Define function in case of problems detected
if [ $? -eq 0 ]
then
echo "The $SYSTEM system has RAID failures on it." >> $LOG_FILE
echo "Below is the output from /proc/mdstat" >> $LOG_FILE
echo "===========================================" >> $LOG_FILE
cat /proc/mdstat >> $LOG_FILE
echo "===========================================" >> $LOG_FILE
cat $LOG_FILE | mail -s 'URGENT: RAID disk failure detected' $MAILTO
fi

# Deleting log file
rm -f >> $LOG_FILE

# Exit
exit 0

Save the file above as /usr/local/bin/raidcheck.sh  and assign to it correct permission and ownership with the folowing :

chmod 700 /usr/local/bin/raidcheck.sh
chown root:nobody /usr/local/bin/raidcheck.sh

Change the email address variable with a real email address who will receive the notification.

You should now schedule the script execution at every time interval you want (i.e. twice a day, or hourly if you are paranoic).

Remember that is better to be paranoic that have a degraded system with unrecoverable data.

Hope this help

Bye
Riccardo

Bookmark It