Nagios系列文章：

Nagios监控平台之二：nrpe监控远程Linux主机

监控监控linux本地主机时，我们可以直接更改配置文件进行监控，如果需要监控的主机与nagios不在同一机器上，即监控远程linux主机时，我们需要借助NRPE插件实现。

nrpe工作原理图：

远程主机的操作

安装支持：

# yum -y install openssl openssl-devel

下载Nagios Plugins和NRPE

#cd /tmp

#wget http://iweb.dl.sourceforge.net/project/nagios/nrpe-2.x/nrpe-2.13/nrpe-2.13.tar.gz

#wget http://iweb.dl.sourceforge.net/project/nagiosplug/nagiosplug/1.4.16/nagios-plugins-1.4.16.tar.gz

创建nagios帐号

# useradd nagios

# passwd nagios

安装nagios-plugin

# cd /tmp

# tar xvfz nagios-plugins-1.4.16.tar.gz

# cd nagios-plugins-1.4.16

# export LDFLAGS=-ldl

# ./configure --with-nagios-user=nagios --with-nagios-group=nagios --enable-redhat-pthread-workaround

# make

# make install

#chown nagios.nagios /usr/local/nagios

#chown -R nagios.nagios /usr/local/nagios/libexec/

安装NRPE

# cd /tmp

# tar xvfz nrpe-2.13.tar.gz

# cd nrpe-2.13

# ./configure

# make all

# make install-plugin

# make install-daemon

# make install-daemon-config

# yum install xinetd

# make install-xinetd

配置NRPE以守护进程运行

1、更改/etc/xinetd.d/nrpe文件，设置允许nagios服务器连接,如nagios服务器的ip为192.168.1.2:

only_from       = 127.0.0.1 192.168.1.2

2、在/etc/services结尾增加：

nrpe 5666/tcp # NRPE

3、启动xinetd

# service xinetd restart

4、验证nrpe是否监听

# netstat -at | grep nrpe

5、测试nrpe是否正常运行

# /usr/local/nagios/libexec/check_nrpe -H localhost

NRPE v2.13

6、更改 /usr/local/nagios/etc/nrpe.cfg

nrpe.cfg文件里包含需要监控远程主机的命令，如下面是我的配置：

#用户登录数

command[check_users]=/usr/local/nagios/libexec/check_users -w 5 -c 10

#CPU负载

command[check_load]=/usr/local/nagios/libexec/check_load -w 15,10,5 -c 30,25,20

#磁盘空间

command[check_disk]=/usr/local/nagios/libexec/check_disk -w  20 -c 10 -p /dev/sda

#僵尸进程数

command[check_zombie_procs]=/usr/local/nagios/libexec/check_procs -w 5 -c 10 -s Z

#进程总数

command[check_total_procs]=/usr/local/nagios/libexec/check_procs -w 150 -c 200

#物理内存

command[check_mem]=/usr/local/nagios/libexec/check_mem -H $HOSTADDRESS$

#脚本内容见下文

物理内存检查脚本/usr/local/nagios/libexec/check_mem：

#!/bin/bash

# check memory script

# Total memory

# by Barlow

# 2014-06-13

help() {

echo "Usage: `basename $0` -w -c "

echo "-w is WARNING % of used mem;-c is CRITICAL % of used mem!"

exit 3

}

TOTAL=`free -m | head -2 |tail -1 |gawk '{print $2}'`

# check memory

FREE=`free -m | head -3 |tail -1 |gawk '{print $4}'`

USED=`free -m | head -3 |tail -1 |gawk '{print $3}'`

# to calculate free percent

# use the expression free * 100 / total

FREETMP=`expr $FREE \* 100`

USEDTMP=`expr $USED \* 100`

FREE_PERCENT=`expr $FREETMP / $TOTAL`

USED_PERCENT=`expr $USEDTMP / $TOTAL`

if [ $# -le 3 ];then

help

elif ! [ $1 == "-w" ]&>/dev/null;then

help

elif ! [ $3 == "-c" ]&>/dev/null;then

help

fi

WARNIFNUM() {

if ! [ "$WARN" == "$OPTARG" ];then

help

fi

}

CRITIFNUM() {

if ! [ "$CRIT" == "$OPTARG" ];then

help

fi

}

while getopts "w:c:h" OPT; do

case $OPT in

"w")

WARNTMP=$OPTARG

WARN=$(echo $WARNTMP |bc 2>/dev/null)

if ! [ "$WARN" == "$WARNTMP" ];then

help

fi

;;

"c")

CRITTMP=$OPTARG

CRIT=$(echo $CRITTMP |bc 2>/dev/null)

if ! [ "$CRIT" == "$CRITTMP" ];then

help

fi

;;

"h")

help;;

esac

done

CRIT_LEVEL=`expr $TOTAL \* $CRIT \/ 100`

WARN_LEVEL=`expr $TOTAL \* $WARN \/ 100`

if [ $USED_PERCENT -gt $CRIT ];then

echo "CRITICAL! Used Memory $USED MB ($USED_PERCENT%,Total=$TOTAL MB) | 'USED MEM'=${USED}MB;$WARN_LEVEL;$CRIT_LEVEL;0;$TOTAL"

exit 2

fi

if [ $USED_PERCENT -gt $WARN ];then

echo "WARNING! Used Memory $USED MB ($USED_PERCENT%,Total=$TOTAL MB) | 'USED MEM'=${USED}MB;$WARN_LEVEL;$CRIT_LEVEL;0;$TOTAL"

exit 1

else

echo "OK! Used Memory $USED MB ($USED_PERCENT%,Total=$TOTAL MB) | 'USED MEM'=${USED}MB;$WARN_LEVEL;$CRIT_LEVEL;0;$TOTAL"

exit 0

fi

nagios服务器的操作

下载安装NRPE

# cd /tmp

# wget http://iweb.dl.sourceforge.net/project/nagios/nrpe-2.x/nrpe-2.13/nrpe-2.13.tar.gz

# tar xvfz nrpe-2.13.tar.gz

# cd nrpe-2.13

# ./configure

# make all

# make install-plugin

测试是否正常:

# /usr/local/nagios/libexec/check_nrpe -H 192.168.1.3

NRPE v2.13

为监控远程主机定义host和service

1、定义check_nrpe命令

在文件/usr/local/nagios/etc/objects/commands.cfg后面增加：

# 'check_nrpe' command definition

define command{

command_name check_nrpe

command_line $USER1$/check_nrpe -H $HOSTADDRESS$ -t 30 -c $ARG1$

}

2、创建/usr/local/nagios/etc/objects/host.cfg    (需提前在nagios.cfg中定义)

host定义示例：

define host{

use linux-server

host_name remotehost

address 192.168.1.3

}

3、创建服务：vi /usr/local/nagios/etc/objects/services.cfg  (需提前在nagios.cfg中定义)

如定义监控远程主机磁盘空间示例(其他服务语法相同)：

define service{

use generic-service

host_name  remotehost

service_description sda磁盘空间

check_command check_nrpe!check_disk

}

之后重载nagios配置文件使其生效

# service nagios reload