Nagios plugin to monitor ZFS appliance share usage

• Intro
• User
• Plugin
• Graph
• Add new share for monitoring

Intro

User

(zfs app): configuration users nagios> show

Properties:
                      logname = nagios
                          type = local
                           uid = 2000000000
                      fullname = nagios
              initial_password = *************
            require_annotation = false
                         roles = ro
                    kiosk_mode = false
                  kiosk_screen = status/dashboard
 

(zfs app):configuration users nagios preferences keys> show

Keys:
NAME     MODIFIED              TYPE   COMMENT
key-000  2015-3-2 17:51:31     RSA    nagios_rsa_public_key

Plugin

#!/bin/ksh
#set -x
# -----------------------------------------------------------------
# Zarko, nagios plugin to check disk space on zfs storage appliance
# 4-7-2017 : zd: Add support for a pool
# 4-11-2017 : zd: Add support for a snapshot
# -----------------------------------------------------------------

PROGNAME=`/bin/basename $0`

# ------ Nagios plugin return values
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4

# usage function
usage() {
echo " \
Usage
        ${PROGNAME} <zfsapp> <pool> <project> <share> <war_free_%> <crit_free_%>
"
}

# End script with output, with performance data for pnp4nagios
endscript () {
        echo ${RESULT}
        exit ${EXIT_STATUS}
}

# check if there are 6 arguments
if [ $# != 6 ]; then
        usage ; exit 3
fi

# temp file and cleanup of same
tmp_file=/tmp/${PROGNAME}.tmp

# --- CLEANING SUBROUTINE
tmp_file_cleaning () {
        [ -f ${tmp_file}.$$ ] && rm ${tmp_file}.$$
}

# --- cleaning in case of script termination and regular exit
trap tmp_file_cleaning HUP INT QUIT ABRT EXIT

ssh -i /var/log/nagios/.ssh/id_rsa nagios@$1 <<EOF > ${tmp_file}.$$
shares
set pool=$2
select $3
select $4
show
EOF

# Calculation is done in Mega

# Get quota
share_quota=`grep "quota =" ${tmp_file}.$$ | awk '{print $3}'`
if [ ${share_quota: -1} = G ]; then
  share_quota=`(echo "scale=2; ${share_quota%?}*1024" | bc -l)`
elif [ ${share_quota: -1} = T ]; then
  share_quota=`(echo "scale=2; ${share_quota%?}*1024*1024" | bc -l)`
else
  share_quota=`echo ${share_quota%?}`
fi

# Get available space
share_space_available=`grep "space_available" ${tmp_file}.$$ | awk '{print $3}'`
if [ ${share_space_available: -1} = G ]; then
  share_space_available=`(echo "scale=2; ${share_space_available%?}*1024" | bc -l)`
elif [ ${share_space_available: -1} = T ]; then
  share_space_available=`(echo "scale=2; ${share_space_available%?}*1024*1024" | bc -l)`
else
  share_space_available=`echo ${share_space_available%?}`
fi

# Get space used by data
share_space_data=`grep "space_data" ${tmp_file}.$$ | awk '{print $3}'`
if [ ${share_space_data: -1} = G ]; then
  share_space_data=`(echo "scale=2; ${share_space_data%?}*1024" | bc -l)`
elif [ ${share_space_data: -1} = T ]; then
  share_space_data=`(echo "scale=2; ${share_space_data%?}*1024*1024" | bc -l)`
else
  share_space_data=`echo ${share_space_data%?}`
fi

# Get space used by snapshots
share_space_snapshots=`grep "space_snapshots" ${tmp_file}.$$ | awk '{print $3}'`
if [ ${share_space_snapshots} = 0 ]; then
  share_space_snapshots=0
elif [ ${share_space_snapshots: -1} = G ]; then
  share_space_snapshots=`(echo "scale=2; ${share_space_snapshots%?}*1024" | bc -l)`
elif [ ${share_space_snapshots: -1} = T ]; then
  share_space_snapshots=`(echo "scale=2; ${share_space_snapshots%?}*1024*1024" | bc -l)`
else
  share_space_snapshots=`echo ${share_space_snapshots%?}`
fi

# Total space is space used by data and snapshots
share_space_total=`(echo "scale=2; ${share_space_data}+${share_space_snapshots}" | bc -l)`

# warn and crit arguments are in %
WARNING=$5
CRITICAL=$6
# Warning size is $warn_size_free
# Critical size is $crit_size_free
warn_used=`(echo "scale=2; (100-${WARNING})*${share_quota}/100" | bc -l)`
warn_free=`(echo "scale=2; ${WARNING}*${share_quota}/100" | bc -l)`
crit_used=`(echo "scale=2; (100-${CRITICAL})*${share_quota}/100" | bc -l)`
crit_free=`(echo "scale=2; ${CRITICAL}*${share_quota}/100" | bc -l)`

if [ ${share_space_available} -gt ${warn_free} ]
then
  RESULT="$2/$3/$4 OK - Free space ${share_space_available}MB | $2/$3/$4=${share_space_total}MB;${warn_used};${crit_used};0;${share_quota}"
  EXIT_STATUS=${STATE_OK}
elif [ ${share_space_available} -le ${warn_free} ] && [ ${share_space_available} -gt ${crit_free} ]
then
  RESULT="$2/$3/$4 WARNING - Free space ${share_space_available}MB | $2/$3/$4=${share_space_total}MB;${warn_used};${crit_used};0;${share_quota}"
  EXIT_STATUS=${STATE_WARNING}
else
  RESULT="$2/$3/$4 CRITICAL - Free space ${share_space_available}MB | $2/$3/$4=${share_space_total}MB;${warn_used};${crit_used};0;${share_quota}"
  EXIT_STATUS=${STATE_CRITICAL}
fi

# ------- provide output and nagios return value
endscript

Graph

<?php
foreach ($this->DS as $KEY=>$VAL) {
# set initial values
        $fmt = '%7.3lf';
        $pct = '';
        $upper = "";
        $maximum = "";
        $divis = 1;
        $return = '\n';
        $unit = "B";
        $label = $unit;
        if ($VAL['UNIT'] != "") {
                $unit = $VAL['UNIT'];
                $label = $unit;
                if ($VAL['UNIT'] == "%%") {
                        $label = '%';
                        $fmt = '%5.1lf';
                        $pct = '%';
                }
        }
        if ($VAL['MAX'] != "") {
                # adjust value and unit, details in .../helpers/pnp.php
                $max = pnp::adjust_unit( $VAL['MAX'].$unit,1024,$fmt );
                $upper = "-u $max[1] ";
                $maximum = "of $max[1] $max[2]$pct used";
                $label = $max[2];
                $divis = $max[3];
                $return = '';
        }

        # dont wanna replace _ with / ????
        $ds_name[$KEY] = str_replace("_","/",$VAL['NAME']);

        # set graph labels
        $opt[$KEY]     = "--vertical-label $label -l 0 $upper --title \"Filesystem $ds_name[$KEY]\" ";

        # Graph Definitions
        $def[$KEY]     = rrd::def( "var1", $VAL['RRDFILE'], $VAL['DS'], "AVERAGE" );
        # "normalize" graph values
        $def[$KEY]    .= rrd::cdef( "v_n","var1,$divis,/");
        $def[$KEY]    .= rrd::area( "v_n", "#c6c6c6",  $ds_name[$KEY] );
        $def[$KEY]    .= rrd::line1( "v_n", "#003300" );
        # show values in legend
        $def[$KEY]    .= rrd::gprint( "v_n", "LAST", "$fmt $label$pct $maximum ");
        $def[$KEY]    .= rrd::gprint( "v_n", "AVERAGE", "$fmt $label$pct avg used $return");
        # create max line and legend
        if ($VAL['MAX'] != "") {
                $def[$KEY] .= rrd::gprint( "v_n", "MAX", "$fmt $label$pct max used \\n" );
                $def[$KEY] .= rrd::hrule( $max[1], "#003300", "Size of FS  $max[0] \\n");
        }
        # create warning line and legend
        if ($VAL['WARN'] != "") {
                $warn = pnp::adjust_unit( $VAL['WARN'].$unit,1024,$fmt );
                $def[$KEY] .= rrd::hrule( $warn[1], "#ffff00", "Warning  on $warn[0] \\n" );
        }
        # create critical line and legend
        if ($VAL['CRIT'] != "") {
                $crit = pnp::adjust_unit( $VAL['CRIT'].$unit,1024,$fmt );
                $def[$KEY] .= rrd::hrule( $crit[1], "#ff0000", "Critical on $crit[0]\\n" );
        }
}
?>

Add new share for monitoring

# check usage of ZFS Appliance shares
define command{
   command_name check_zfsappliance_share_usage
   command_line $USER1$/check_share_space_zfs_appliance.ksh $ARG1$ $ARG2$ $ARG3$ $ARG4$ $ARG5$ $ARG6$
   }

define service{
        use                     generic-service,srv-pnp
        host_name               zfs-appliance-hostname
        service_description     Disk usage pool-name/project/share-name
        check_command           check_zfsappliance_share_usage!$HOSTADDRESS$!pool-name!project!share-name!20!10
        normal_check_interval   15
        }