【nagios 二】nagios配置篇

上篇“自动化安装nagios”完成了nagios的服务器端的安装，现需要对nagios进行配置，实现基本的监控功能。

参考文章：老男孩的博客

          南非蚂蚁的博客

实验环境：

• CentOS 6.3 X86_64, VirtualBox 4.1.14
• 关闭防火墙和SELinux
• nagios版本nagios-3.4.1
• 插件版本nagios-plugins-1.4.16
• nrpe版本nrpe-2.13

实现步骤：

1. 添加被监控服务器的机器名到配置文件hosts.cfg
2. 添加被监控服务器的机器名到所属组的配置文件hostgrps.cfg
3. 添加被监控服务器的服务类型到配置文件services.cfg
4. 将nrpe添加到命令配置文件commands.cfg
5. 将上述配置文件的路径放在nagios的主配置文件nagios.cfg里声明

开始配置：

声明一个变量为nagios安装路径

nagios_dir="/usr/local/nagios"

一，添加被监控的主机到hosts.cfg

     cd $nagios_dir/etc/objects                  #进入nagios配置文件所在目录
     [ -f new.host ] && rm -f new.host           #判断new.host文件是否存在，若在则删除旧主机文件
     exec < hosts.list                           #hosts.list为手动创建的文件，用于指定主机名和IP
     while read line                             #从hosts.list提取主机名和IP并添加到new.host文件
     do
          echo "define host {" >> new.host
          echo "     use           linux-server" >> new.host
          echo "     host_name     `echo $line|awk '{print $1}'`" >> new.host
          echo "     alias         `echo $line|awk '{print $1}'`" >> new.host
          echo "     address       `echo $line|awk '{print $2}'`" >> new.host
          echo "}" >> new.host
          echo >> new.host
     done
     cat new.host >> hosts.cfg                  #将做好的主机模板添加到hosts.cfg文件

hosts.list文件需手动创建，格式为“主机名    IP地址”，可以一次添加多个,

若以后有新主机需要增加，则直接在hosts.list里清空旧内容，将新主机信息添加进去

test-machine1    192.168.0.100
test-machine2    192.168.0.101
test-machine3    192.168.0.102

new.host为生成的主机模板

define host {
     use           linux-server
     host_name     test-machine1
     alias         test-machine1
     address       192.168.0.100
}

最后将生成好的主机模板添加到nagios所需要的主机文件hosts.cfg，里面都是生成好的所定义的被监控主机模板。

二，添加主机所属组

将被监控主机归到一个监控组里，这里仅生成一个组，其他以此类推

    cd $nagios_dir/etc/objects
    newhosts=`cat hosts.list|awk '{print $1}'|tr '\n' ','`        #提取host.list文件里的主机名
    if [ ! -f hostgrps.cfg ]; then                                #判断hostgrps.cfg文件是否存在
         echo "define hostgroup {" >> hostgrps.cfg                #添加hostgrps.cfg文件正文
         echo "     hostgroup_name     lnx-srvs" >> hostgrps.cfg
         echo "     alias     lnx-srvs" >> hostgrps.cfg
         echo "     members     $newhosts" >> hostgrps.cfg
         echo "}" >> hostgrps.cfg
         sed -i.bak '/members/ s/,$//' hostgrps.cfg               #去掉主机名结尾的逗号
    else
         sed -i.bak '/members/ s/^.*$/&,'$newhosts'/' hostgrps.cfg #若hostgrps文件已存在，则仅将新主机名追加到members那一列最后
         sed -i.bak '/members/ s/,$//' hostgrps.cfg
    fi

最终所生成的hostgrps.cfg文件名格式为

define hostgroup{
        hostgroup_name  lnx_srvs
        alias           lnx_srvs
        members         test-machine1,test-machine2
}

三，将nrpe添加到命令配置文件

echo"
define command{
   command_name   nrpe
   command_line   $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$
   }" >> $natios_dir/etc/objects/commands.cfg

四，添加服务文件

接下来是比较重要的nagios监控的服务文件services.cfg

     cd $nagios_dir/etc/objects
     newhosts=`cat hosts.list|awk '{print $1}'|tr '\n' ','`   #定义变量newhosts为提取的主机名
     if [ ! -f services.cfg ]; then                           #判断services.cfg文件是否存在
echo "                                                        #若不存在则新增模板正文
define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Current Users On Remote System
         check_command                    nrpe!check_users    #监控主机当前登录的用户数量
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Current System Loads
         check_command                    nrpe!check_load     #监控系统的负载
         }


define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Zombie Processes On Remote System
         check_command                    nrpe!check_zombie_procs  #监控主机僵尸进程数量
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Total Processes On Remote System
         check_command                    nrpe!check_total_procs   #监控主机总进程数量
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Disk Using Of Remote System
         check_command                    nrpe!check_disk  #监控磁盘空间
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Swap Using Of Remote System
         check_command                    nrpe!check_swap   #监控交换分区空间
         }" > serverces.cfg

          sed -i.bak '/host_name/ s/,$//' services.cfg      #去掉host_name最后的逗号
     else
          sed -i.bak '/host_name/ s/^.*$/&,'$newhosts'/' services.cfg  #若services.cfg已存在，则直接追加新主机名即可
          sed -i.bak '/host_name/ s/,$//' services.cfg
     fi

五，重启nagios服务

以上配置文件都需在nagios主配置文件nagios.cfg里声明

cfg_file=/usr/local/nagios/etc/objects/hosts.cfg
cfg_file=/usr/local/nagios/etc/objects/hostgrps.cfg
cfg_file=/usr/local/nagios/etc/objects/services.cfg

重启nagios服务

/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg
service nagios restart

至此，nagios服务器端配置基本完成，剩下的则需要在被监控主机做相关配置。

完整的配置脚本如下：

#!/bin/bash
#auto configure nagios settings
#2012-11-28

nagios_dir="/usr/local/nagios"

function Hosts_cfg()
{
     cd $nagios_dir/etc/objects
     [ -f new.host ] && rm -f new.host
     exec < hosts.list
     while read line
     do
          echo "define host {" >> new.host
          echo "     use           linux-server" >> new.host
          echo "     host_name     `echo $line|awk '{print $1}'`" >> new.host
          echo "     alias         `echo $line|awk '{print $1}'`" >> new.host
          echo "     address       `echo $line|awk '{print $2}'`" >> new.host
          echo "}" >> new.host
          echo >> new.host
     done
     cat new.host >> hosts.cfg
}

function Hostgrps_cfg()
{
     cd $nagios_dir/etc/objects
     newhosts=`cat hosts.list|awk '{print $1}'|tr '\n' ','`
     if [ ! -f hostgrps.cfg ]; then
          echo "define hostgroup {" >> hostgrps.cfg
          echo "     hostgroup_name     lnx-srvs" >> hostgrps.cfg
          echo "     alias     lnx-srvs" >> hostgrps.cfg
          echo "     members     $newhosts" >> hostgrps.cfg
          echo "}" >> hostgrps.cfg
          sed -i.bak '/members/ s/,$//' hostgrps.cfg
     else
          sed -i.bak '/members/ s/^.*$/&,'$newhosts'/' hostgrps.cfg
          sed -i.bak '/members/ s/,$//' hostgrps.cfg
     fi
}

function Add_command()
{
    echo"
define command{
   command_name   nrpe
   command_line   $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$
   }" >> $natios_dir/etc/objects/commands.cfg
}

function Services_cfg()
{
     cd $nagios_dir/etc/objects
     newhosts=`cat hosts.list|awk '{print $1}'|tr '\n' ','`
     if [ ! -f services.cfg ]; then
echo "
define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Current Users On Remote System
         check_command                    nrpe!check_users
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Current System Loads
         check_command                    nrpe!check_load
         }


define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Zombie Processes On Remote System
         check_command                    nrpe!check_zombie_procs
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Total Processes On Remote System
         check_command                    nrpe!check_total_procs
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Disk Using Of Remote System
         check_command                    nrpe!check_disk
         }

define service{
         use                              remote-service
         host_name                        '$newhosts'
         service_description              Swap Using Of Remote System
         check_command                    nrpe!check_swap
         }" > services.cfg

          sed -i.bak '/host_name/ s/,$//' services.cfg
     else
          sed -i.bak '/host_name/ s/^.*$/&,'$newhosts'/' services.cfg
          sed -i.bak '/host_name/ s/,$//' services.cfg
     fi

}

function Restart_service()
{
    /usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg
    service nagios restart
}

case $1 in
host)
    Hosts_cfg
    ;;
group)
    Hostgrps_cfg
    ;;
command)
    Add_command
    ;;
service)
    Services_cfg
    ;;
restart)
    Restart_service
    ;;
*)
    echo "Usage: ./setnagios.sh {host|group|command|service|restart}"
    ;;
esac

需要相应的功能时则运行脚本带上相应参数即可

./setnagios.sh host    #添加主机

./setnagios.sh group   #添加组或修改组成员

./setnagios.sh command  #将nrpe添加到命令配置文件

./setnagios.sh service #添加服务文件或修改主机成员

./setnagios.sh restart #重启nagios 服务