Nagios Support Forum

Support for Nagios products and services
https://support.nagios.com/forum/

Problem with custom bash script for HPE StoreVirtual 3200

https://support.nagios.com/forum/viewtopic.php?t=59807

Page 1 of 1

Problem with custom bash script for HPE StoreVirtual 3200

Posted: Mon Aug 24, 2020 7:00 am
by Ulfandu
Hello,

I've been dabbling with a script that would check the RAM status for both storage controllers on a HPE StoreVirtual 3200 SAN.

The script seems to work so far that I'm able to fetch the amount of total ram, its usage, and calculate the amount of RAM available. However, I'm not understanding why my script is constantly returning CRITICAL.

Here's the script in question:

Code: Select all

#! /bin/sh
#
# MIBS required for the script to work:
#       LEFTHAND-NETWORKS-NSM-CLUSTERING-MIB.mib 
#       LEFTHAND-NETWORKS-GLOBAL-REG-MIB.mib
#       LEFTHAND-NETWORKS-NSM-MIB.mib
#
# Usage: ./check_lhc_mem.sh IP

# return states for nagios
STATE_OK=$(expr 0)
STATE_WARNING=$(expr 1)
STATE_CRITICAL=$(expr 2)
STATE_UNKNOWN=$(expr 3)

# check device name
Name=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -o SNMPv2-MIB::sysName.0|cut -d" " -f4)

RET=$?
if [[ $RET -ne 0 ]]
then
echo "query problem - No data received from host"
exit $STATE_UNKNOWN
fi

# read Physical Memory
RamTotal=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -o HOST-RESOURCES-MIB::hrStorageSize.1|cut -d" " -f4)

# read Used Memory
RamUsed=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -o HOST-RESOURCES-MIB::hrStorageUsed.1|cut -d" " -f4)


checkcount=0
iomodulestring=""

# convert KB --> GB
RamTotal=$(echo "$RamTotal / 1024" | bc)
RamUsed=$(echo "$RamUsed / 1024" | bc)
RamAvail=$(echo "$RamTotal - $RamUsed" |bc)

# convert MB --> GB
RamTotal=$(echo "$RamTotal / 1024" | bc)
RamUsed=$(echo "$RamUsed / 1024" | bc)
RamAvail=$(echo "$RamTotal - $RamUsed" |bc)

# calculate the percentage of available RAM
percentfree=$(echo "scale=2; $RamAvail/$RamTotal " |bc)
percentfree=$(echo "$percentfree*100" | bc)


# thresholds for critical and warning
critical=$(echo 10.0)
warning=$(echo 15.0)

# compare 
 if [[ "$(echo "$percentfree <= $critical" | bc -l)" ]]; then
  echo "CRITICAL - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
  exit $STATE_CRITICAL

 elif [[ "$(echo "$percentfree > $warning" | bc -l)" ]]; then
    echo "CRITICAL - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
   exit $STATE_OK

 elif [[ "$(echo "$percentfree <= $warning" | bc -l)" ]]; then
    echo "CRITICAL - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
  exit $STATE_WARNING

  else
 echo "problem - No data received from host"
   exit $STATE_UNKNOWN
 fi
Here's a debug run of the script on a certain SV3200 controller:

Code: Select all

++ expr 0
+ STATE_OK=0
++ expr 1
+ STATE_WARNING=1
++ expr 2
+ STATE_CRITICAL=2
++ expr 3
+ STATE_UNKNOWN=3
++ /usr/local/nagios/libexec/check_snmp -P 2c -H 10.88.255.22 -C corpsnmp -o SNMPv2-MIB::sysName.0
++ cut '-d ' -f4
+ Name=Company-SV3200-2-SC1
+ RET=0
+ [[ 0 -ne 0 ]]
++ /usr/local/nagios/libexec/check_snmp -P 2c -H 10.88.255.22 -C corpsnmp -o HOST-RESOURCES-MIB::hrStorageSize.1
++ cut '-d ' -f4
+ RamTotal=6649216
++ /usr/local/nagios/libexec/check_snmp -P 2c -H 10.88.255.22 -C corpsnmp -o HOST-RESOURCES-MIB::hrStorageUsed.1
++ cut '-d ' -f4
+ RamUsed=5991744
+ checkcount=0
+ iomodulestring=
++ echo '6649216 / 1024'
++ bc
+ RamTotal=6493
++ echo '5991744 / 1024'
++ bc
+ RamUsed=5851
++ bc
++ echo '6493 - 5851'
+ RamAvail=642
++ echo '6493 / 1024'
++ bc
+ RamTotal=6
++ echo '5851 / 1024'
++ bc
+ RamUsed=5
++ echo '6 - 5'
++ bc
+ RamAvail=1
++ bc
++ echo 'scale=2; 1/6 '
+ percentfree=.16
++ echo '.16*100'
++ bc
+ percentfree=16.00
++ echo 10.0
+ critical=10.0
++ echo 15.0
+ warning=15.0
++ echo '16.00 <= 10.0'
++ bc -l
+ [[ -n 0 ]]
+ echo 'CRITICAL - Company-SV3200-2-SC1 is 16.00 % free | used=5GB;;total=6GB;;available=1GB'
CRITICAL - Company-SV3200-2-SC1 is 16.00 % free | used=5GB;;total=6GB;;available=1GB
+ exit 2
I would greatly appreciate any help on this matter as I have ran into a brick wall :|

I can provide the MIBs used for the script if necessary.

Re: Problem with custom bash script for HPE StoreVirtual 320

Posted: Mon Aug 24, 2020 3:51 pm
by jdunitz
There are a few problems I noticed:

One is, you might consider converting your calculations to whole numbers by dividing by 1 and using bc -l:

Code: Select all

     percentfree=$(echo "scale=2; $RamAvail/$RamTotal " |bc -l)
     percentfree=$(echo "$percentfree*100" | bc -l)
     percentfree=$(echo "scale=0; $percentfree / 1"|  bc -l)
Then, I rewrote your comparison section at the end to not use bc, but just do numeric comparisons in the shell:

Code: Select all

    # compare
    if [[ $percentfree -le $critical ]]; then
      echo "CRITICAL - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
      exit $STATE_CRITICAL

    elif [[ $percentfree -le $warning ]]; then
        echo "WARNING - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
       exit $STATE_OK

    elif [[ $percentfree  > $warning ]]; then
        echo "OK - "$Name" is "$percentfree" % free | used="$RamUsed"GB;;total="$RamTotal"GB;;available="$RamAvail"GB"
      exit $STATE_WARNING

      else
    echo "problem - No data received from host"
       exit $STATE_UNKNOWN
    fi

This isn't perfect, but I think it's closer to what you're after.

--Jeffrey

Re: Problem with custom bash script for HPE StoreVirtual 320

Posted: Wed Aug 26, 2020 1:28 am
by Ulfandu
Thanks! This helped a lot and I managed to figure out how to fix it. For those wondering, here's the full script now:

Code: Select all

#! /bin/sh
#
# Example output:
# WARNING - corp-SV3200-1-SC2 is 15 % free | used=5508MB;;total=6493MB;;available=985MB
#
#
# Usage: ./check_lhc_mem.sh IP

# return states for nagios
STATE_OK=$(expr 0)
STATE_WARNING=$(expr 1)
STATE_CRITICAL=$(expr 2)
STATE_UNKNOWN=$(expr 3)

# check device name
Name=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -o SNMPv2-MIB::sysName.0|cut -d" " -f4)

RET=$?
if [[ $RET -ne 0 ]]
then
echo "query problem - No data received from host"
exit $STATE_UNKNOWN
fi

# read Physical Memory
RamTotal=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -t 60 -o HOST-RESOURCES-MIB::hrStorageSize.1|cut -d" " -f4)

# read Used Memory
RamUsed=$(/usr/local/nagios/libexec/check_snmp -P 2c -H $1 -C corpsnmp -t 60 -o HOST-RESOURCES-MIB::hrStorageUsed.1|cut -d" " -f4)


checkcount=0
iomodulestring=""

# convert KB --> MB
RamTotal=$(echo "$RamTotal / 1024" | bc)
RamUsed=$(echo "$RamUsed / 1024" | bc)
RamAvail=$(echo "$RamTotal - $RamUsed" |bc)

# calculate the percentage of available RAM
percentfree=$(echo "scale=2; $RamAvail/$RamTotal " |bc -l)
percentfree=$(echo "$percentfree*100" | bc -l)
percentfree=$(echo "scale=0; $percentfree / 1"|  bc -l)

# thresholds for critical and warning
critical=$(echo 10)
warning=$(echo 15)

# compare free ram against thresholds
if [[ $percentfree -le $critical ]]; then
	echo "CRITICAL - "$Name" is "$percentfree" % free | used="$RamUsed"MB;;total="$RamTotal"MB;;available="$RamAvail"MB"
    exit $STATE_CRITICAL

elif [[ $percentfree -le $warning ]]; then
    echo "WARNING - "$Name" is "$percentfree" % free | used="$RamUsed"MB;;total="$RamTotal"MB;;available="$RamAvail"MB"
    exit $STATE_OK

elif [[ $percentfree -ge $warning ]]; then
    echo "OK - "$Name" is "$percentfree" % free | used="$RamUsed"MB;;total="$RamTotal"MB;;available="$RamAvail"MB"
    exit $STATE_WARNING

else
    echo "problem - No data received from host"
    exit $STATE_UNKNOWN
fi
All times are UTC-05:00
Page 1 of 1

Powered by phpBB® Forum Software © phpBB Limited