aix:scripts_category:nagios_aix_check
This is an old revision of the document!
Table of Contents
Scripts for AIX monitoring with Nagios
Monitor filesystems
script 1
This script with an exception list will check the percent usage of each filesystem, and you can also use wildcards.
root@aixserver - /opt/freeware/lib/nagios/plugins > cat MY_checkfs.cfg #fs_name:pct_warning:pct_critical # You can add exception in this file, else the default is used #Example: #/db*:98:100 , 98 and 100 whill be used for all filesystems /db* #default:90:95 /tmp:80:95 /usr:96:98 /opt:95:98 /cdrom:100:100 /mnt:100:100
root@aixserver - /opt/freeware/lib/nagios/plugins > cat MY_checkfs.sh
#!/usr/bin/ksh93
#set -x
##################################################
#@(#) send an alert if filesystems usage is higher than threshold
##################################################
# associate file is checkfs.cfg (exceptions)
# please do not modify this script use checkfs.cfg
# checkfs.cfg format:
# fs_name:pctwarning_threshold:pctcritical_threshold
# version: 1.1 10-02-2014 Manu
##################################################
dir=`dirname $0`
#. $dir/.env
sn=`basename $0 | cut -d. -f1`
configfile=/opt/freeware/lib/nagios/plugins/$sn.cfg
logpath=/tmp
logname=$logpath/$sn.log
tmpcfg=$logpath/$sn.txt
tmpscript=$logpath/$sn.scr
nb_file=25
nb_dir=15
DATE=$(date +"%Y-%m-%d %H:%M")
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4
#echo "$sn: $DATE" > $logname
#------------------------------------------------
# generate_case_file ()
#------------------------------------------------
generate_case_file()
{
pctwarn_def=$(grep "^#default" $configfile | cut -d":" -f2)
pctcrit_def=$(grep "^#default" $configfile | cut -d":" -f3)
cat $configfile | grep -v "^ *#" | sed '/^$/d' > $tmpcfg
cat $tmpcfg | awk -F':' '{print $1") pctwarn="$2" pctcrit="$3}' | sed '/^$/d;$G' | sed '$s/^/\*\)\ pctwarn=pctwarn_def\ pctcrit=pctcrit_def /' | sed "s/pctwarn_def/$pctwarn_def/" | sed "s/pctcrit_def/$pctcrit_def/" | sed '2,$s/^/\\n/' | sed 's/$/;;/' > $tmpcfg
}
#------------------------------------------------
# check_fs ()
#------------------------------------------------
check_fs()
{
EXIT_CODE=0
MSG_WARN=""
MSG_CRIT=""
OUTPUT="All filesystems OK"
clauses=$(cat $tmpcfg)
for i in $(df -k | egrep -v "Filesystem|:|/proc" | grep -v ":" | grep "/dev/" | awk '{print $4 $7}')
do
fs=`echo $i | cut -f 2 -d%`
pct_fs=`echo $i | cut -f 1 -d%`
mail=""
cmd="case $fs in \n $clauses \n esac"
echo $cmd > $tmpscript
. $tmpscript > /dev/null
if [ $pct_fs -gt $pctwarn ]
then
if [ $pct_fs -gt $pctcrit ]
then
MSG_CRIT="$MSG_CRIT $fs used $pct_fs"
EXIT_CODE=$(echo $STATE_CRITICAL)
else
MSG_WARN="$MSG_WARN $fs used $pct_fs"
if [ $STATE_WARNING -gt $EXIT_CODE ]
then
EXIT_CODE=$(echo $STATE_WARNING)
fi
fi
#info_fs >> $logname 2>&1 &
fi
done
if [ "$MSG_CRIT" != "" ]
then
OUTPUT="DISK CRITICAL - $MSG_CRIT"
fi
if [ "$MSG_WARN" != "" ]
then
if [ "$MSG_CRIT" != "" ]
then
OUTPUT="$OUTPUT -- DISK WARNING - $MSG_WARN"
else
OUTPUT="DISK WARNING - $MSG_WARN"
fi
fi
rm $tmpcfg $tmpscript > /dev/null 2>&1
}
#------------------------------------------------
# info_fs ()
#------------------------------------------------
info_fs()
{
echo "####### Here are the $nb_file biggest files from filesystem $fs"
find $fs -xdev -type f -ls | sort +6nr | head -$nb_file
echo "\n####### Here are the $nb_dir biggest directories from filesystem $fs"
du -ms $fs/* | sort -nr | head -$nb_dir
}
generate_case_file #>> $logname 2>&1
check_fs #>> $logname 2>&1
echo $OUTPUT
exit $EXIT_CODE
script 2
root@aixserver - /opt/freeware/lib/nagios/plugins > cat check_fs.cfg #filesystem_name:alert_w(%):alert_c(%) /tmp:80:95 /usr:96:98 /cdrom:100:100 /mnt:100:100 /export/aix6100-09:99:100 /export/aix7100-03:99:100 /export/aix7200-02:99:100 /export/mksysb:95:99 /export/vios226:99:100 /export/vios310:99:100 /export/software:99:100
root@aixserver - /opt/freeware/lib/nagios/plugins > cat check_fs.sh
#!/bin/sh
#set -x
##################################################
#@(#) check filesystems usage is higher than threshold
##################################################
# associate file is check_fs.cfg (exceptions)
# please do not modify this script use check_fs.cfg
# check_fs.cfg format:
# fs_name:pct_warn:pct_crit
# check also GPFS quotas if used
# version: 1.0 10-2020 Manu
##################################################
STATUS=0
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
MSG=""
configfile=/usr/local/nagios/libexec/check_fs.cfg
os=$(uname -a | awk '{print $1}')
defaultwarn=92
defaultcrit=98
#------------------------------------------------
# check_gpfs ()
#------------------------------------------------
check_gpfs()
{
# test gpfs quota if applicable
lsfs -a | sed '1d' | awk '{print $4}' | grep "mmfs" > /dev/null 2>&1
if [ $? -eq 0 ]
then
quotaalert=2000000
for i in sybase kplus
do
used=$(mmlsquota -u $i |grep clkpfs|awk '{print $3}')
max=$(mmlsquota -u $i |grep clkpfs|awk '{print $4}')
diff=$(expr $max - $used)
if [ $diff -lt $quotaalert ]
then
echo "***** GPFS Quota-WARNING *****" >> $logname
echo " User $i has ounly $diff KB free quota " >> $logname
echo " Reduce used space or use mmedquota -u $i to modifie quota" >> $logname
fi
done
fi
}
#------------------------------------------------
# check_aix_fs ()
#------------------------------------------------
check_aix_fs()
{
df -k | egrep -v "Filesystem|:|/proc" | grep -v ":" | grep "/dev/" | awk '{print $4,$7}' | sed 's/\%//' | while read pct fs
do
if [ -e $configfile ]
then
line=$(grep "^$fs:" $configfile)
if [ "$line" != "" ]
then
warn=$(echo $line |cut -d':' -f2)
crit=$(echo $line |cut -d':' -f3)
else
warn=$defaultwarn
crit=$defaultcrit
fi
else
warn=$defaultwarn
crit=$defaultcrit
fi
if [ "$pct" -gt "$crit" ]
then
MSG=$(echo "$MSG $fs:${pct}%")
STATE=${STATE_CRITICAL}
else
if [ "$pct" -gt "$warn" ]
then
MSG=$(echo "$MSG $fs:${pct}%")
STATE=${STATE_WARNING}
fi
fi
if [ "$STATUS" -lt "$STATE" ]
then
STATUS=$STATE
fi
done
}
#------------------------------------------------
# check_lnx_fs ()
#------------------------------------------------
check_lnx_fs()
{
df -k | egrep -v "Filesystem|:|/proc" | grep -v ":" | grep "/dev/" | rev | awk '{print $1,$2}' | rev | sed 's/\%//' | while read pct fs
do
if [ -e $configfile ]
then
line=$(grep "^$fs:" $configfile)
if [ "$line" != "" ]
then
warn=$(echo $line |cut -d':' -f2)
crit=$(echo $line |cut -d':' -f3)
else
warn=$defaultwarn
crit=$defaultcrit
fi
else
warn=$defaultwarn
crit=$defaultcrit
fi
if [ "$pct" -gt "$crit" ]
then
MSG=$(echo "$MSG $fs:${pct}%")
STATE=${STATE_CRITICAL}
else
if [ "$pct" -gt "$warn" ]
then
MSG=$(echo "$MSG $fs:${pct}%")
STATE=${STATE_WARNING}
fi
fi
if [ "$STATUS" -lt "$STATE" ]
then
STATUS=$STATE
fi
done
}
###########################################
if [ "$os" = "Linux" ]
then
check_lnx_fs
else
if [ "$os" = "AIX" ]
then
check_aix_fs
else
echo "########## Unknown OS"
STATUS=$STATE_UNKNOWN
fi
fi
#check_gpfs
case "$STATUS" in
"$STATE_OK") MSG=$(echo "OK") ;;
"$STATE_WARNING") MSG=$(echo "WARNING: $MSG") ;;
"$STATE_CRITICAL") MSG=$(echo "CRITICAL: $MSG") ;;
esac
echo $MSG
exit $STATUS
check_ntp_aix.sh
#!/bin/ksh93
#
# AIX check NTP for Nagios v1.0 06/2023
#
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4
STATE=$(/usr/bin/lssrc -ls xntpd > /dev/null 2>&1; echo $?)
if [ "$STATE" -ne 0 ]
then
EXIT_CODE=$STATE_CRITICAL
OUTPUT="Process: xntpd not running"
else
CODE=$(/usr/bin/lssrc -ls xntpd | tr -s ' ' | tr ' ' ';' | grep -i 'Leap;indicator:' | cut -d':' -f2- | cut -d'(' -f1 | sed 's/;//g')
if [ "$CODE" == "00" ]
then
EXIT_CODE=$STATE_OK
OUTPUT="Process: xntpd synchronized OK"
else
EXIT_CODE=$STATE_CRITICAL
OUTPUT="Process: xntpd Not synchronized"
fi
fi
echo $OUTPUT
exit $EXIT_CODE
check_mailq_aix
#!/usr/bin/ksh
# set -x
##################################################
#@(#) check mail server and mailq
# best practice is to stop sendmail service on AIX
# as it's not required to send email
# but if you leave it running, it happens that
# the service hangs, then no mails 'll be sent
##################################################
# v1.1 12-2020
##################################################
# Return codes:
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
DS='smtp.local.lu' # Add here the SMTP server to check
MAILDIR=/var/spool/mail
TMPfile=/tmp/check_mail.tmp
# no warning before 3 days
mailwarning=3 # max days mail pending
mailcritical=10 # max days mail pending
if [ $(grep "^DS${DS}" /etc/sendmail.cf > /dev/nul 2>&1; echo $?) -ne "0" ]
then
echo "CRITICAL: config file error /etc/sendmail.cf"
exit $STATE_CRITICAL
else
if [ $(ls $MAILDIR | grep -v lost | wc -l) != "0" ]
then
cat /dev/null > $TMPfile
for file1 in $(find $MAILDIR -type f)
do
grep '^Date:' $file1| cut -d' ' -f2-5 | head -5 | tr ' ' ';' >> $TMPfile
done
for line in $(cat $TMPfile)
do
val=$(echo $line | sed 's/;/\ /g')
/opt/freeware/bin/date -d"$val" "+%s" >> $TMPfile.1
done
val=$(cat $TMPfile.1 | sort | head -1)
val1=$(/opt/freeware/bin/date "+%s")
days=$(echo "scale=0;($val1-$val)/3600/24" | bc)
rm $TMPfile* 2>/dev/null
if [ "$days" -gt "$mailcritical" ]
then
echo "CRITICAL: mail pending $days days"
exit $STATE_CRITICAL
else
if [ "$days" -gt "$mailwarning" ]
then
echo "WARNING: mail pending $days days"
exit $STATE_WARNING
fi
fi
else
echo "OK"
exit $STATE_OK
fi
fi
check_aix_cpu.sh
#!/usr/bin/ksh
# check VCPU usage in percent
# according to a definition set by the app . owner
# the virtual cpu usage > 99% CPUs left on the system will result
# in an errorcode 2
# for nagios to be critical and alerted
SAR="/usr/sbin/sar"
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4
LIST_WARNING_THRESHOLD=${LIST_WARNING_THRESHOLD:-"95"}
LIST_CRITICAL_THRESHOLD=${LIST_CRITICAL_THRESHOLD:-"99"}
INTERVAL_SEC=${INTERVAL_SEC:="1"}
NUM_REPORT=${NUM_REPORT:="3"}
print_usage() {
echo ""
echo "$PROGNAME $RELEASE - CPU Utilization check script for Nagios"
echo ""
echo "Usage: check_cpu_stats.sh -w -c (-i -n)"
echo ""
echo " -w Warning threshold in % for warn_user,warn_system,warn_iowait CPU (default : ${LIST_WARNING_THRESHOLD})"
echo " Exit with WARNING status if cpu exceeds warn_n"
echo " -c Critical threshold in % for crit_user,crit_system,crit_iowait CPU (default : ${LIST_CRITICAL_THRESHOLD})"
echo " Exit with CRITICAL status if cpu exceeds crit_n"
echo " -h Show this page"
echo ""
echo "Usage: $PROGNAME -w 95 -c 99"
echo ""
exit 0
}
# Parse parameters
while [ $# -gt 0 ]; do
case "$1" in
-h | --help)
print_usage
exit $STATE_OK
;;
-w | --warning)
shift
LIST_WARNING_THRESHOLD=$1
;;
-c | --critical)
shift
LIST_CRITICAL_THRESHOLD=$1
;;
esac
shift
done
NUM_VCPUS=$(lsdev -Cc processor|wc -l|awk '{print $1}')
PHYSC=$($SAR ${INTERVAL_SEC} ${NUM_REPORT} | tail -1 | awk '{print $6}')
PCTUSED=$(echo "scale=2\n ${PHYSC}*100/${NUM_VCPUS}" | bc)
if [ $PCTUSED -gt $LIST_WARNING_THRESHOLD ]
then
if [ $PCTUSED -gt $LIST_CRITICAL_THRESHOLD ]
then
echo "CRITICAL - CPU usage at $PCTUSED%"
exit $STATE_CRITICAL
else
echo "WARNING - CPU usage at $PCTUSED%"
exit $STATE_WARNING
fi
else
echo "OK - CPU usage at $PCTUSED%"
exit $STATE_OK
fi
Check path disks
# cat check_paths.sh
#!/bin/sh
#@(#) v1.0 Count number of paths per disk
# v1.1 add sudo linux
# v1.2 add alerting for sudo
# Add these lines in sudoers files
# [root@lnx log]# cat /etc/sudoers.d/99_nagios
# nagios ALL=(ALL) NOPASSWD: /usr/sbin/multipath
# nrpe ALL=(ALL) NOPASSWD: /usr/sbin/multipath
nbpathlpar=8
nbpathvios=4
nbpath=$1
STATUS=0
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
MSG=""
# check if server is VIOS
if [ -e "/usr/ios/cli/ioscli" ]
then
nbpath1=$nbpathvios
else
nbpath1=$nbpathlpar
fi
# specified value for nb paths
if [ "$npath" == "" ]
then
nbpath=$nbpath1
fi
os=$(uname -a | awk '{print $1}')
grepp() { [ $# -eq 1 ] && perl -00ne "print if /$1/i" || perl -00ne "print if /$1/i" < "$2";}
#---------------------
count_linux_path()
{
if [ ! -x /usr/sbin/multipath ] || [ $(lsscsi -s | grep -q VMware; echo $?) -eq 0 ]
then
MSG="OK: no multipathing"
STATUS=$STATE_OK
else
if [ $(timeout 10 sudo /usr/sbin/multipath ; echo $?) -ne "0" ]
then
MSG="$MSG Maybe error on sudo config"
STATUS=$STATE_UNKNOWN
else
for i in $(sudo /usr/sbin/multipath -ll | grep mpath | awk '{print $1}')
do
pathok=$(sudo /usr/sbin/multipath -ll $i | grep -v "policy=" | grep -v "size=" | grep -v '^mpath' | grep active | wc -l | awk '{print $1}')
pathok_pct=$(echo "scale=1;100*$pathok/$nbpath" | bc | cut -d '.' -f1)
if [ "$pathok_pct" -le "50" ]
then
MSG="$MSG $i;$pathok/$nbpath"
if [ $STATUS -lt $STATE_CRITICAL ]
then
STATUS=$STATE_CRITICAL
fi
else
if [ "$pathok_pct" -ne "100" ]
then
MSG="$MSG $i;$pathok/$nbpath"
if [ $STATUS -lt $STATE_CRITICAL ]
then
STATUS=$STATE_WARNING
fi
fi
fi
done
fi
fi
}
#---------------------
count_aix_path()
{
if [ $(lsdev -Cc disk | grep -v Available | wc -l | awk '{print $1}') -ne "0" ]
then
MSG="$MSG WARNING: disks defined"
STATUS=$STATE_WARNING
else
STATUS=$STATE_OK
fi
for i in $(lsdev -Cc disk | grep -i mpio | awk '{print $1}')
do
pathok=$(lspath -l $i | grep Enabled | wc -l | awk '{print $1}')
pathok_pct=$(echo "scale=1;100*$pathok/$nbpath" | bc | cut -d '.' -f1)
if [ "$pathok_pct" -le "50" ]
then
MSG="$MSG $i;$pathok/$nbpath"
if [ $STATUS -lt $STATE_CRITICAL ]
then
STATUS=$STATE_CRITICAL
fi
else
if [ "$pathok_pct" -ne "100" ]
then
MSG="$MSG $i;$pathok/$nbpath"
if [ $STATUS -lt $STATE_CRITICAL ]
then
STATUS=$STATE_WARNING
fi
fi
fi
done
}
if [ "$os" = "Linux" ]
then
count_linux_path
else
if [ "$os" = "AIX" ]
then
count_aix_path
else
echo "########## Unknown OS"
STATUS=$STATE_UNKNOWN
fi
fi
if [ $STATUS -eq $STATE_OK ]
then
echo "OK"
else
echo "$MSG"
fi
exit $STATUS
Modification TODO:
count_linux_path()
{
for disk in $(/usr/sbin/multipath -ll | awk '{print $1}' | grep '^mpath')
do
echo $disk $(/usr/sbin/multipath -ll | grep -v "policy=" | grep -v "size=" | sed '/^mpath/i \\n' | grepp $disk | grep -v '^mpath' | grep active | wc -l) | sed 's/\ /;/g' | sed 's/$/;/' | sed "s/$/$numberpath;/" >> $output
done
($verbose) && cat $output
}
Check SEA status
Check Shared Ethernet Adapter status on VIOS
# cat check_sea.sh
#!/bin/ksh
#valeurs retour Nagios
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4
EXITSTATUS=$STATE_UNKNOWN # Default Exit Code as UNKNOWN.
#Recuperation variable
hostname=$(hostname)
SEA=$(lsdev -Cc adapter | grep Shared|grep -v Defined | wc -l | awk '{print $1}')
now=$(date +"%d_%m_%Y_%H%M%S")
#echo $now
#Initialisation
echo "" > verif_eth_$now.txt
#Verification Carte
#echo $SEA
if [ $SEA = "0" ]
then
#echo ""
echo "No SEA available on $hostname"
#echo ""
EXITSTATUS=2 #Exit status critical
#echo $EXITSTATUS
exit $EXITSTATUS
fi
#Verification ensemble des cartes si disponible
for ent in $(lsdev -Cc adapter | grep Shared | awk '{print $1}')
do
entstat -d $ent 2>&1 | grep -i status| grep "Physical Port Link Status" > verif_eth_$now.txt
count_all=$(cat verif_eth_$now.txt |wc -l| awk '{print $1}')
count_up=$(cat verif_eth_$now.txt |grep Up |wc -l| awk '{print $1}')
#echo $count_up
#echo $count_all
if [ $count_up -ne $count_all ]
then
echo "One or more Ethernet port down"
#cat verif_eth_$now.txt
EXITSTATUS=1 #Exit status Warning
#echo $EXITSTATUS
else
echo "No ethernet port issue"
#cat verif_eth_$now.txt
EXITSTATUS=0 #Exit status OK
#echo $EXITSTATUS
fi
done
rm verif_eth_$now.txt
exit $EXITSTATUS
Modification TODO:
count_linux_path()
{
for disk in $(/usr/sbin/multipath -ll | awk '{print $1}' | grep '^mpath')
do
echo $disk $(/usr/sbin/multipath -ll | grep -v "policy=" | grep -v "size=" | sed '/^mpath/i \\n' | grepp $disk | grep -v '^mpath' | grep active | wc -l) | sed 's/\ /;/g' | sed 's/$/;/' | sed "s/$/$numberpath;/" >> $output
done
($verbose) && cat $output
}
aix/scripts_category/nagios_aix_check.1700660660.txt.gz · Last modified: 2023/11/22 14:44 by manu
Except where otherwise noted, content on this wiki is licensed under the following license: GNU Free Documentation License 1.3
