ZFS JBOD Monitoring

If using ZFS software raid (RAIDZ2 for example) to provide Lustre OST's, monitoring disk and enclosure health can be a challenge. This is because typically vendor disk array monitoring is included as part of a package with RAID controllers.

If you are aware of any vendor-supported monitoring solutions for this or have your own solution, please add to this page.

UW SSEC Solution

This information is used for linux systems and monitoring Dell MD1200 disk arrays directly attached via SAS, with no RAID card.

Disk Failure: zpool status

To detect disk failure, simply check the zpool status. There are various scripts to do this for nagios/check_mk.

To perform checks outside of nagios scripts can be written to use the zpool status command. An example is as follows, which uses the ldev.conf file to know pool names associated with lustre, and mutt to send the email:

#!/bin/bash
#
# zfs monitoring script for lustre with zfs backend
# uses /etc/ldev.conf to locate zpools, then zpool status to find degraded pools.

HELP="
 This script uses /etc/ldev.conf and zpool status to identify mounted pools, then sends an email if 
 a pool returns a status other then ONLINE
"

LDEV_FILE="/etc/ldev.conf" 
EMAIL="[email protected]"

send_email ()
{
/usr/bin/mutt -s "zpool status warning on $HOSTNAME" $EMAIL<< EOF
"$1"
EOF
}

if [ ! -f $LDEV_FILE ]
then
	/usr/bin/mutt -s "WARNING, no ldev file found on $HOSTNAME" $EMAIL
	exit
fi

for POOL in `cat $LDEV_FILE`
do
	if [[ `echo $POOL | grep ost` ]]
	then
		POOL_NAME=`echo $POOL | cut -f2 -d":" | cut -f1 -d"/"`
		POOL_STATUS=`/sbin/zpool status $POOL_NAME`

		#check for errors running zpool
		if [ ! $? ]
		then
			send_email "$POOL_STATUS"	
		fi

		#get pool state		
		POOL_STATE=`echo "$POOL_STATUS" | grep state`
		if [[ $POOL_STATE != *ONLINE* ]]
		then
			send_email "$POOL_STATUS"
		fi
	fi

done

The script above can then be run as a cron job, with the below as an example of that cron job.

#zpool monitoring crontab for oss systems
SHELL=/bin/bash
PATH=/sbin:/bin:/usr/sbin:/usr/bin
HOME=/tmp
50 06 * * * root /usr/local/zfs_monitor/zfs_mon.sh 2>&1 /dev/null; exit 0

Predictive Failure: smartctl

To monitor predictive drive failure, we use 'smartctl' provided by the 'smartmontools' package for centos.

Example check_mk script:

#!/bin/bash
#

DISKS="$(/bin/ls /dev/disk/by-vdev| /bin/grep -v part)"
UNHEALTHY_COUNT=0

for DISK in ${DISKS}
do
HEALTH=`smartctl -H /dev/disk/by-vdev/${DISK} | grep SMART`
HEALTHSTATUS=`echo ${HEALTH} | cut -d ' ' -f 4`
if [[ $HEALTHSTATUS != "OK" ]]; then
status=2
else
status=0
fi
echo "$status SMART_Status_${DISK} - ${DISK} ${HEALTH}"

done

Enclosure Monitoring

While the above techniques tell you if you have a disk problem, you still need to monitor the status of the arrays themselves. For our particular problem this is MD1200 disk arrays via SAS. For us, sg3_utils and SCSI Enclosure Services sg_ses is the best answer so far.

To monitor our enclosures we use this script: http://www.ssec.wisc.edu/~scottn/Lustre_ZFS_notes/script/check_md1200.pl