Note: There is a file embedded within this post, please visit this post to download the file.
机房没有温度报警装置，我用此方法实现对机房温度的掌控，如果只有一台报警，则可认为单机故障，如果几台同时报警，则可认为机房空调出现了问题。
具体实现方法如下：
环境：被监控机：CentOS 6.4
1、安装硬件传感器监控软件 sensors
#yum install lm_sensors*
2、运行sensors-detect进行传感器检测
#sensors-detect ##一路回车即可，此步我在虚拟机下报错，但在物理机上没有问题
3、运行sensors看是否能读取数据，如下像下面这样表示正常：
[root@rd02 ~]# sensors
coretemp-isa-0000
Adapter: ISA adapter
Core 0: +32.0°C (high = +76.0°C, crit = +100.0°C)
Core 1: +32.0°C (high = +76.0°C, crit = +100.0°C)
4、#vi /usr/local/nagios/libexec/check_cputemp ##粘贴如下#号之间的内容
##########################################################
#!/bin/sh
#########check_cputemp###########
#date : May 2011
#Licence GPLv2
#INSTALLATION
#the script need to install lm_sensors
#sensors’s output need like below format
#########################################
#coretemp-isa-0000#
#Adapter: ISA adapter#
#Core 0: +27°C (high = +85°C)#
#
#coretemp-isa-0001#
#Adapter: ISA adapter#
#Core 1: +25°C (high = +85°C) #
#########################################
#you can use NRPE to define service in nagios
#check_nrpe!check_cputemp.sh
# Plugin return statements
STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
print_help_msg(){
$Echo “Usage: $0 -h to get help.”
}
print_full_help_msg(){
$Echo “Usage:”
$Echo “$0 [ -v ] -m sensors -w cpuT -c cpuT”
$Echo “Sepicify the method to use the temperature data sensors.”
$Echo “And the corresponding Critical value must greater than Warning value.”
$Echo “Example:”
$Echo “${0} -m sensors -w 40 -c 50″
}
print_err_msg(){
$Echo “Error.”
print_full_help_msg
}
to_debug(){
if [ "$Debug" = "true" ]; then
$Echo “$*” >> /var/log/check_sys_temperature.log.$$ 2>&1
fi
}
unset LANG
Echo=”echo -e”
if [ $# -lt 1 ]; then
print_help_msg
exit 3
else
while getopts :vhm:w:c: OPTION
do
case $OPTION
in
v)
#$Echo “Verbose mode.”
Debug=true
;;
m)
method=$OPTARG
;;
w)
WARNING=$OPTARG
;;
c)
CRITICAL=$OPTARG ;;
h)
print_full_help_msg
exit 3
;;
?)
$Echo “Error: Illegal Option.”
print_help_msg
exit 3
;;
esac
done
if [ "$method" = "sensors" ]; then
use_sensors=”true”
to_debug use_sensors
else
$Echo “Error. Must to sepcify the method to use sensors.”
print_full_help_msg
exit 3
fi
to_debug All Values are \” Warning: “$WARNING” and Critical: “$CRITICAL” \”.
fi
#########lm_sensors##################
if [ "$use_sensors" = "true" ]; then
sensorsCheckOut=`which sensors 2>&1`
if [ $? -ne 0 ];then
echo $sensorsCheckOut
echo Maybe you need to check your sensors.
exit 3
fi
to_debug Use $sensorsCheckOut to check system temperature
TEMP1=`sensors | head -3 | tail -1 | gawk ‘{print $3}’ | grep -o [0-9][0-9]`
TEMP2=`sensors | head -4 | tail -1 | gawk ‘{print $3}’ | grep -o [0-9][0-9]`
SUM=$(( $TEMP1 + $TEMP2 ))
TEMP=$(($SUM/2))
if [ -z "$TEMP" ] ; then
$Echo “No Data been get here. Please confirm your ARGS and re-check it with Verbose mode, then to check the log.”
exit 3
fi
to_debug temperature data is $TEMP
else
$Echo “Error. Must to sepcify the method to use sensors”
print_full_help_msg
exit 3
fi
######### Comparaison with the warnings and criticals thresholds given by user############
CPU_TEMP=$TEMP
#if [ "$WARNING" != "0" ] || [ "$CRITICAL" != "0" ]; then
if [ "$CPU_TEMP" -gt "$CRITICAL" ] && [ "$CRITICAL" != "0" ]; then
STATE=”$STATE_CRITICAL”
STATE_MESSAGE=”CRITICAL”
to_debug $STATE , Message is $STATE_MESSAGE
elif [ "$CPU_TEMP" -gt "$WARNING" ] && [ "$WARNING" != "0" ]; then
STATE=”$STATE_WARNING”
STATE_MESSAGE=”WARNING”
to_debug $STATE , Message is $STATE_MESSAGE
else
STATE=”$STATE_OK”
STATE_MESSAGE=”OK”
to_debug $STATE , Message is $STATE_MESSAGE
fi
echo “The TEMPERATURE “$STATE_MESSAGE” “-” The CPU’s Temperature is “$CPU_TEMP” ℃ !”
exit $STATE
##########################################################
5、赋予上述脚本执行权限：
#chmod +x /usr/local/nagios/libexec/check_cputemp
6、配置nrpe.cfg，添加如下一行：
command[check_cputemp]=/usr/local/nagios/libexec/check_cputemp -m sensors -w 38 -c 45
注意：以上六步均在被监控机上完成。
7、在Nagios服务器配置服务：
define service{
use generic-service
host_name <hostname>
service_description CPU Temperature
check_command check_nrpe!check_cputemp
}
保存后重启nagios服务即可。