Shaun's Wiki

/usr/openv/netbackup/bin/support/nbperfchk -i zero: -o ./nbperfchk.tmp -s 150g

drive_down_checker.ksh

#!/bin/ksh
#----------------------------------------------------------------------
# Check drives, bring drive up if down, email
# If drives are up, do not create a report, runs from cron
#----------------------------------------------------------------------
# Add interested parties below:
alert_mail=mail@mydomain.co.za
/usr/openv/volmgr/bin/vmoprcmd -d ds | grep "DOWN" >/dev/null
if [ $? -ne 1 ];then
        rm /tmp/drivelog
        /bin/date >> /tmp/drivelog
 
        echo "" >> /tmp/drivelog
        for i in `/usr/openv/volmgr/bin/vmoprcmd -d ds | tail -n +5 | grep "DOWN"|awk '{print $1}'`
        do
          echo -e "Drive $i is in DOWN status, bringing drive UP\n" >> /tmp/drivelog
          /usr/openv/volmgr/bin/vmoprcmd -up $i
        done
/usr/openv/volmgr/bin/vmoprcmd -d ds >> /tmp/drivelog
cat /tmp/drivelog | /bin/mail -s "Drive(s) outages on `hostname`" \
  $alert_mail
fi

In shared environments it is imperative that SCSI traffic passes through the network/SAN infrastructure unmolested and without corruption, otherwise issues such as SCSI reservations can arise due to devices not releasing their reservations correctly and/or being unable to acquire them.

Disabling SCSI reservations in Netbackup may be a workaround - Netbackup will then handle drive reservations and conflicts internally.

Release reservations manually with the following command:

vmoprcmd -crawlreleasebyname <DRIVE_NAME>

nbrbutil -dump
nbrbutil -releaseMDS <key>
nbrbutil -reportInconsistentAllocations
nbrbutil -releaseDrive <drive name>