搭建 nagios 监控 - 经验分享 - 技术分享 | Hito = Hito 的公告栏 = 天下事有难易乎？为之，则难者亦易矣

# 监控端服务安装与配置

nagios 需要安装主程序和 nrpe（nagios 和各被监控主机都必须安装）。如需使用自研前端可通过安装 ndoutils (用于把 nagios 监控信息写入数据库) 和 mysql 实现。具体安装见官网
nagios 默认监控命令脚本放置在 libexec 中，自定义脚本也放到此处
etc/objects/commands.cfg 用于保存 nagios 默认监控命令
etc/nrpe.cfg 文件中需要添加用于被监控执行命令项（各被监控主机中都需要添加）
etc/objects/hosts 目录下配置需要被监控的主机信息
etc/objects/servers 目录下配置需要在被监控主机上执行的监控命令（第 4 项中的命令）

# 被监控端服务安装与配置

被监控端需要安装 nrpe
被监控端需要把监控端的命令写入到 nrpe 的配置文件中
被监控端需要把命令执行脚本放入 libexec 目录中
被监控端自动安装脚本（可借鉴）

1、请先修改脚本中的服务端IP。
2、如被监控端不支持let计算命令请执行sudo dpkg-reconfigure dash命令，弹出选择窗口后选择no。
3、需要在脚本同目录下创建CentOS、Ubuntu、sh和conf目录，目录下分别放置nrpe的tar安装包、监控脚本和nrpe配置文件。

#!/bin/bash
#服务端ip
SERVER_IP=10.10.10.121
#安装目录
INSTALL_HOME=`pwd`
#安装ubuntu版本
INSTALL_UBUNTU()
{
	sudo apt-get update
	sudo apt-get install -y autoconf automake gcc libc6 libmcrypt-dev make libssl-dev wget openssl
	
	cd /tmp
	#wget --no-check-certificate -O nrpe.tar.gz https://github.com/NagiosEnterprises/nrpe/archive/nrpe-4.0.3.tar.gz
	cp $INSTALL_HOME/Ubuntu/nrpe.tar.gz ./
	tar xzf nrpe.tar.gz

	cd /tmp/nrpe-nrpe-4.0.3/
	sudo ./configure --enable-command-args --with-ssl-lib=/usr/lib/x86_64-linux-gnu/
	sudo make all

	sudo make install-groups-users

	sudo make install

	sudo make install-config

	sudo sh -c "echo >> /etc/services"
	sudo sh -c "sudo echo '# Nagios services' >> /etc/services"
	sudo sh -c "sudo echo 'nrpe    5666/tcp' >> /etc/services"
	
	#判断系统是高版本还是低版本
	VERSION=`lsb_release -r --short`
	IS_LOW_VERSION=`echo "$VERSION < 15" | bc`

	if [ $IS_LOW_VERSION = 1 ]; then
		#低版本
		sudo make install-init
	else
		#高版本
		sudo make install-init
		sudo systemctl enable nrpe.service
	fi
	
	sudo mkdir -p /etc/ufw/applications.d
	sudo sh -c "echo '[NRPE]' > /etc/ufw/applications.d/nagios"
	sudo sh -c "echo 'title=Nagios Remote Plugin Executor' >> /etc/ufw/applications.d/nagios"
	sudo sh -c "echo 'description=Allows remote execution of Nagios plugins' >> /etc/ufw/applications.d/nagios"
	sudo sh -c "echo 'ports=5666/tcp' >> /etc/ufw/applications.d/nagios"
	sudo ufw allow NRPE
	sudo ufw reload
	
	sudo sh -c "sed -i '/^allowed_hosts=/s/$/,$SERVER_IP/' /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "sed -i 's/^dont_blame_nrpe=.*/dont_blame_nrpe=1/g' /usr/local/nagios/etc/nrpe.cfg"
	
	cd $INSTALL_HOME
	sudo cp ./sh/* /usr/local/nagios/libexec/
	sudo chmod +xr /usr/local/nagios/libexec/*
	
	sudo sh -c "echo 'command[check_ping]=/usr/local/nagios/libexec/check_ping -H 127.0.0.1 -w 3000.0,80% -c 5000.0,100% -p 5' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_cpu]=/usr/local/nagios/libexec/check_cpu.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_mem]=/usr/local/nagios/libexec/check_mem.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_disk]=/usr/local/nagios/libexec/check_disk.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_my_service]=/usr/local/nagios/libexec/check_my_service.sh \$ARG1\$' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_system_info]=/usr/local/nagios/libexec/check_system_info.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	NETWORK_INDEX=1
	for NETWORK_NAME in `cat /proc/net/dev | awk '{i++; if(i>2){print $1}}' | sed 's/^[\t]*//g' | sed 's/[:]*$//g'`;do
			if [ $NETWORK_NAME != 'lo' ]; then
					sudo sh -c "echo 'command[check_network$NETWORK_INDEX]=/usr/local/nagios/libexec/check_network.sh $NETWORK_NAME' >> /usr/local/nagios/etc/nrpe.cfg"
					NETWORK_INDEX=`expr $NETWORK_INDEX + 1 `
			fi
	done
	
	if [ $IS_LOW_VERSION = 1 ]; then
		#低版本
		sudo start nrpe
	else
		#高版本
		sudo systemctl start nrpe.service
	fi
}

#安装centos版本
INSTALL_CENTOS()
{
	yum install -y gcc glibc glibc-common openssl openssl-devel perl wget

	cd /tmp
	#wget --no-check-certificate -O nrpe.tar.gz https://github.com/NagiosEnterprises/nrpe/archive/nrpe-4.0.3.tar.gz
	cp $INSTALL_HOME/CentOS/nrpe.tar.gz ./
	tar xzf nrpe.tar.gz

	cd /tmp/nrpe-nrpe-4.0.3/
	./configure --enable-command-args
	make all

	make install-groups-users

	make install

	make install-config

	echo >> /etc/services
	echo '# Nagios services' >> /etc/services
	echo 'nrpe    5666/tcp' >> /etc/services
	
	#判断系统是高版本还是低版本
	VERSION=`rpm -q centos-release|cut -d- -f3`
	#安装bc命令
	yum -y install bc
	IS_LOW_VERSION=`echo "$VERSION < 7" | bc`

	if [ $IS_LOW_VERSION = 1 ]; then
		#低版本
		make install-init
		
		iptables -I INPUT -p tcp --destination-port 5666 -j ACCEPT
		service iptables save
		ip6tables -I INPUT -p tcp --destination-port 5666 -j ACCEPT
		service ip6tables save
	else
		#高版本
		make install-init
		systemctl enable nrpe.service
		
		firewall-cmd --zone=public --add-port=5666/tcp
		firewall-cmd --zone=public --add-port=5666/tcp --permanent
	fi
	
	sudo sh -c "sed -i '/^allowed_hosts=/s/$/,$SERVER_IP/' /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "sed -i 's/^dont_blame_nrpe=.*/dont_blame_nrpe=1/g' /usr/local/nagios/etc/nrpe.cfg"
	
	cd $INSTALL_HOME
	sudo cp ./sh/* /usr/local/nagios/libexec/
	sudo chmod +xr /usr/local/nagios/libexec/*
	
	sudo sh -c "echo 'command[check_ping]=/usr/local/nagios/libexec/check_ping -H 127.0.0.1 -w 3000.0,80% -c 5000.0,100% -p 5' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_cpu]=/usr/local/nagios/libexec/check_cpu.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_mem]=/usr/local/nagios/libexec/check_mem.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_disk]=/usr/local/nagios/libexec/check_disk.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_my_service]=/usr/local/nagios/libexec/check_my_service.sh \$ARG1\$' >> /usr/local/nagios/etc/nrpe.cfg"
	sudo sh -c "echo 'command[check_system_info]=/usr/local/nagios/libexec/check_system_info.sh' >> /usr/local/nagios/etc/nrpe.cfg"
	NETWORK_INDEX=1
	for NETWORK_NAME in `cat /proc/net/dev | awk '{i++; if(i>2){print $1}}' | sed 's/^[\t]*//g' | sed 's/[:]*$//g'`;do
			if [ $NETWORK_NAME != 'lo' ]; then
					sudo sh -c "echo 'command[check_network$NETWORK_INDEX]=/usr/local/nagios/libexec/check_network.sh $NETWORK_NAME' >> /usr/local/nagios/etc/nrpe.cfg"
					NETWORK_INDEX=`expr $NETWORK_INDEX + 1 `
			fi
	done
	
	if [ $IS_LOW_VERSION = 1 ]; then
		#低版本
		if [ $(echo "$VERSION < 6" | bc) -eq 1 ]; then
			service nrpe start
		else
			start nrpe
		fi
	else
		#高版本
		systemctl start nrpe.service
	fi
}

#安装其它版本
INSTALL_OTHER()
{
	echo "Not supported at the moment."
}

#根据不同系统安装不同版本
INSTALL()
{
	if grep -Eqii "CentOS" /etc/issue || grep -Eq "CentOS" /etc/*-release; then
		DISTRO='CentOS'
        PM='yum'
		INSTALL_CENTOS
    elif grep -Eqi "Red Hat Enterprise Linux Server" /etc/issue || grep -Eq "Red Hat Enterprise Linux Server" /etc/*-release; then
        DISTRO='RHEL'
        PM='yum'
		INSTALL_OTHER
    elif grep -Eqi "Aliyun" /etc/issue || grep -Eq "Aliyun" /etc/*-release; then
        DISTRO='Aliyun'
        PM='yum'
		INSTALL_OTHER
    elif grep -Eqi "Fedora" /etc/issue || grep -Eq "Fedora" /etc/*-release; then
        DISTRO='Fedora'
        PM='yum'
		INSTALL_OTHER
    elif grep -Eqi "Debian" /etc/issue || grep -Eq "Debian" /etc/*-release; then
        DISTRO='Debian'
        PM='apt'
		INSTALL_OTHER
    elif grep -Eqi "Ubuntu" /etc/issue || grep -Eq "Ubuntu" /etc/*-release; then
        DISTRO='Ubuntu'
        PM='apt'
		INSTALL_UBUNTU
    elif grep -Eqi "Raspbian" /etc/issue || grep -Eq "Raspbian" /etc/*-release; then
        DISTRO='Raspbian'
        PM='apt'
		INSTALL_OTHER
    else
        echo "unknow linux."
        exit 1
    fi
    echo $DISTRO
}

INSTALL

exit 0

# 常见问题

监控日志未写入 mysql（可能是 ndoutils 服务不正常导致）。采用以下脚本命令解决

#!/bin/bash
sudo rm -f /usr/local/nagios/var/ndo2db.pid
sudo rm -f /usr/local/nagios/var/ndo.sock
sudo systemctl restart ndo2db.service
sudo systemctl status ndo2db.service

nagios 服务监控