nagios的配置

        nagios简介:　Nagios是一款开源的电脑系统和网络监视工具，能有效监控Windows、Linux和Unix的主机状态，交换机路由器等网络设置，打印机等。在系统或服务状态异常时发出邮件或短信报警第一时间通知网站运维人员，在状态恢复后发出正常的邮件或短信通知。
以下主机所用系统为红帽6.5

nagios的安装：

在server2.example.com这台主机中：

主机ip为172.25.254.2

关闭防火墙和selinux

nagios从官网下载所需要的源码包和插件包。

这里用的源码包是 nagios-cn-3.2.3.tar.bz2插件包是 nagios-plugins-2.1.1.tar.gz

yum install gcc gd-devel                                        #安装所需的编译环境

yum install httpd                                                     ＃要通过apache访问

tar jxf nagios-cn-3.2.3.tar.bz2                               #解压源码包

cd nagios-cn-3.2.3

useradd  -M -d /usr/local/nagios  nagios             #创建nagios用户

groupadd nagcmd                                                  #创建可以通过web提交外部命令的组

usermod -G nagcmd  nagios

usermod -G nagcmd  apache                               #将apache和nagios加入这个组

./configure --with-command-group=nagcmd     #配置编译环境指定组为nagcmd

make  all                                                                 #将源码包编译成可执行的二进制文件

make install                                                             #安装

makeinstall-init                                                       #安装启动脚本                             

make install-config                                                 #安装简单的配置文件

make install-webconf                                            #安装简单的web配置文件

make install-commandmode 

htpasswd/usr/local/nagios/etc/htpasswd.users nagiosadmin      ＃设定nagios登录的用户密码

/usr/local/nagios/bin/nagios-v /usr/local/nagios/etc/nagios.cfg   ＃检查配置正确性

/etc/init.d/nagios start                                            ＃启动nagios

/etc/init.d/httpdstart                                               ＃启动apache
我们可以用web登录测试一下如图表示成功：

接下来安装插件

 tar zxf nagios-plugins-2.1.1.tar.gz                     ＃解压插件包

cd nagios-plugins-2.1.1

./configure



--with-mysql和--with-openssl显示no表明没有之后所需的mysql-devle和openssl-devel开发包：

yum installmysql-devel openssl-devel -y                 

./configure             ＃再次配置

make all

make install

cd /usr/local/nagios/libexec/        ＃nagiox插件所在目录

chown nagios.nagios . -R            ＃将这个目录下的文件都改成nagios所有组和所有用户

/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg  ＃检查配置

/etc/init.d/nagios reload                                         #重新加载nagios
如图显示插件成功：



接着是配置属于自己的nagios
cd /usr/local/nagios/etc/                                                                   

在nagios.cfg中

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

＃指定host.cfg文件用来指定被监控的主机地址以及相关属性信息

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

＃指定services.cfg文件用于定义监控的服务和主机资源

    #cfg_file=/usr/local/nagios/etc/objects/localhost.cfg

＃取消示例

cd /usr/local/nagios/etc/objects/

       cp localhost.cfg hosts.cfg -p

       cp localhost.cfg services.cfg -p

通过示例来配置hosts.cfg和services.cfg文件

在hosts.cfg中因为此时只监控本机一台主机所以只保留：

                 

                definehost{
                    use                     linux-server
                    host_name               server2.example.com
                    alias                   Manager
                    address                 172.25.254.2
                    icon_p_w_picpath              server.gif
                    statusmap_p_w_picpath         server.gd2
                    2d_coords               500,200
                    3d_coords               500,200,100
                    }


            define hostgroup{
            hostgroup_name  linux-servers
            alias           Linux Servers
            members         *
              }
在services.cfg中因为只件监控一台主机所以只写一个服务组

                define servicegroup{
               servicegroup_name 系统负荷检查
                alias 负荷检查
                members server2.example.com,进程总数,server2.example.com用户登录                                          数,server2.example.com,根分区,server2.example.com,交换空间利用率
               }

    监控ping的情况

            defineservice{   
                    use                             local-service       
                    host_name                       *
                    service_description             PING
                    check_command                   check_ping!100.0,20%!500.0,60%
                    }

     监控硬盘情况

              defineservice{
                   use                             local-service       
                    host_name                       server2.example.com
                    service_description             根分区
                    check_command                   check_local_disk!20%!10%!/
                   }

     监控登录用户数

            defineservice{
                use                             local-service       
               host_name                      server2.example.com
                service_description             登录用户数
                check_command                   check_local_users!20!50
               }
      监控系统负荷

          define service{
                use                             local-service       
                host_name                       server2.example.com
               service_description             系统负荷
                check_command                   check_local_load!5.0,4.0,3.0!10.0,6.0,4.0
                }

        监控进程总数

            defineservice{
                use                             local-service       
                host_name                       server2.example.com
                service_description             进程总数
               check_command                  check_local_procs!250!400!RSZDT
                }
        监控交换空间利用率

                define service{
                use                             local-service       
                host_name                       server2.example.com
                service_description             交换空间利用率
                check_command                   check_local_swap!20!10
               }

        监控ssh情况

                define service{
                    use                             local-service       
                   host_name                      server2.example.com
                   service_description             SSH
                   check_command                  check_tcp!22!1.0!10.0
                    notifications_enabled           0
                    }
          监控apache

                    defineservice{
                   use                             local-service       
                   host_name                      server2.example.com
                   service_description            HTTP
                   check_command                  check_http
                  notifications_enabled           0
                    }

/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg  ＃检查配置

/etc/init.d/nagios reload                                                                     #重新加载nagios
如图显示本机的监控成功

在server3.example.com主机中：

ip为172.25.254.3

关闭防火墙和selinux

yum install mysql-server                                                                   #安装mysql服务

/etc/init.d/mysqld start                                                                       ＃启动mysql

mysql_secure_installation                                                               #设置mysql root用户密码

mysql -p                                                                                              ＃用root用户进入mysql

create database nagdb                                                                        #创建nagdb库

 grant select on nagdb.* tonagios@172.25.254.2 identified by 'nagios'

＃给远程用户nagios查看nagdb的权力并设定密码为nagios

在server2.example.com中

cd    /usr/local/nagios/libexec

./check_mysql -H 172.25.254.3 -unagios -pnagios

出现如图画面nagix可以监控在server3.example.com中创建的mysql库

cd /usr/local/nagios/etc/objects

在 commands.cfg 中添加监控mysql的命令

        # 'check_mysql' command definition   
            definecommand{
               command_name   check_mysql
                command_line   $USER1$/check_mysql -H $HOSTADDRESS$ -u $ARG1$ -p $ARG2$
                }
在services.cfg中添加监控mysql

     ##########################check_mysql
                   define service{
                      use                             local-service        
                        host_name                       server3.example.com
                       service_description             MYSQL
                        check_command                   check_mysql!nagios!nagios
                        notifications_enabled           0
                        }

在hosts.cfg中添加

         define host{
               use                    linux-server
                host_name               server3.example.com
                alias                   Manager
                parents                 server2.example.com
                address                 172.25.254.3
               icon_p_w_picpath              server.gif
                statusmap_p_w_picpath         server.gd2
                2d_coords               400,100
                3d_coords               400,100,100
            }

/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg  ＃检查配置

/etc/init.d/nagios reload                                                                     #重新加载nagios

如图表示成功检测server3主机上的mysql


nrpe安装使nagios监控多台主机的情况

tar zxfnrpe-2.15.tar.gz                                                                       ＃解压nrpe包

tar zxf  nagios-plugins-2.1.1.tar.gz                     ＃解压插件包

yum install mysql-devel openssl-devel -y

useradd  -M -d /usr/local/nagios  nagios             #创建nagios用户

yum install gcc

cd nagios-plugins-2.1.1

./configure

make all

make install

cd /usr/local/nagios/libexec/        ＃nagiox插件所在目录

chown nagios.nagios . -R            ＃将这个目录下的文件都改成nagios所有组和所有用户

yum install xinetd

cd /root/nrpe-2.15                            #nrpe-2.15.tar.gz解压目录

./configure

make all

make install

make install-plugin
make install-daemon

make install-daemon-config

make install-xinetd

vim /etc/xinetd.d/nrpe                     #设定监控主机为172.25.254.2

        only_from       = 172.25.254.2

vim /etc/services

     nrpe   5666/tcp                           #nrpe的5666端口

/etc/init.d/xinetd start                     ＃开启xinetd

cd /usr/local/nagios/etc/

vim nrpe.cfg          

将command[check_hda1]=/usr/local/nagios/libexec/check_disk-w 20% -c 10% -p /dev/hda1改为

command[check_disk]=/usr/local/nagios/libexec/check_disk -w 20% -c 10% -p/  ＃监控根分区

/etc/init.d/xinetdrestart                  ＃重启xinetd服务

cd /usr/local/nagios/libexec/    

 scp  check_nrpe 172.25.254.2:/usr/local/nagios/libexec/ 将check_nrpe传到server2.exampl.com主机中   在server2.example.com中

cd /usr/local/nagios/libexec/

chown nagios.nagios  check_nrpe     #将check_nrpe所有人和所有组改为nagios3

./chek_nrpe  -H  172.25.254.3              #检测nrpe是否可用成功显示nrpe版本号

cd /usr/local/nagios/etc/objcts/

在 commands.cfg 中添加check_nrpe的命令

        # 'check_nrpe' commanddefinition  
            define command{
                command_name    check_nrpe
                command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c$ARG1$
                }

在services.cfg中添加

        define service{
        use                             local-service        
        host_name                       server3.example.com
        service_description             根分区
        check_command                   check_nrpe!check_disk
        }
define service{
        use                             local-service        
        host_name                       server3.example.com
        service_description             登录用户数
        check_command                   check_nrpe!check_users
        }
监控根分区和登录人数

  /usr/local/nagios/bin/nagios -v/usr/local/nagios/etc/nagios.cfg  ＃检查配置

/etc/init.d/nagios reload                                                                     #重新加载nagios

在server3.example.com中重启xinet.d服务

如图可见监控成功

nagios的110云报警
到onealter官网下载所用的软件包这里用的是
alert-agent-4.1.3.1-linux-x64.tar.gz
在onealter网页添加nagios应用


获取应用key

在server2.example.com主机中：
tar zxf alert-agent-4.1.3.1-linux-x64.tar.gz  
cp -R alert-agent /usr/local/nagios/libexec/  #将解压好的包放入nagios插件目录中
cp alert-agent/plugin/nagios-plugin/nagios /usr/local/nagios/libexec/
chmod +x /usr/local/nagios/libexec/nagios
cp alert-agent/plugin/nagios-plugin/110monitor.cfg/usr/local/nagios/etc/objects/   #将110配置文件放入nagios配置文件目录
在110monitor.cfg文件中添加Key：



修改/usr/local/nagios/etc/objects/contacts.cfg，新增110monitor到默认联系组
    define contactgroup{
    contactgroup_name       admins
    alias                   Nagios Administrators
    members                 nagiosadmin,110monitor
    }  
修改/usr/local/nagios/etc/nagios.cfg，将110monitor.cfg新增到nagios.cfg中
cfg_file=/usr/local/nagios/etc/objects/110monitor.cfg
/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg  ＃检查配置

/etc/init.d/nagios   reload     #重新加载nagios


用ganglia监控系统性能：
   Ganglia是UCBerkeley发起的一个开源集群监视项目，设计用于测量数以千计的节点。Ganglia的核心包含gmond、gmetad以及一个Web前端。主要是用来监控系统性能，如：cpu 、mem、硬盘利用率， I/O负载、网络流量情况等，通过曲线很容易见到每个节点的工作状态，对合理调整、分配系统资源，提高系统整体性能起到重要作用。
ganlia的安装：
在server2.exampl.com中

yum instll rpm-build
yum install -y libart_lgpl_devel  autoconf automake  libtool pcre-develexpat-devel rrdtool-devel-1.3.8-6.e16.x86_64  
yum install ibconfuse-2.6-3.el6.x86_64.rpm libconfuse-devel-2.6-3.el6.x86_64.rpm
#安装软件包依赖性

rpmbuild -tb ganglia-3.3.7.tar.gz #将源码包编下译成二进制可执行文件
rpmbuild -tb ganglia-web-3.4.2.tar.gz
cd /root/rpmbuild/RPMS/x86_64 
rpm -ivh *                     #安装此目录下生成的所有rpm包
其中ganglia-gmond-3.4.0-1.x86_64.rpm ganglia-gmond-modules-python-3.4.0-1.x86_64.rpm  libganglia-3.4.0-1.x86_64.rpm三个包是客户端所需的
cd /root/rpmbuild/RPMS/noarch 
yum install php-gd php -y
rpm -ivh ganglia-web-3.4.2-1.noarch.rpm             #安装此目录下需要解决依赖性的包
vim /etc/ganglia/gmetad.conf      #进入gmetad.conf中更改cluster名
   cluster {
    name = "hello cluster"
    owner = "unspecified"
    latlong = "unspecified"
    url = "unspecified"解决
    }
进入gmond.conf中更改为相同的名字
/etc/init.d/gmetad start
/etc/init.d/gmond start          #启动ganglia服务端和客户端
此时可以登陆http:server2.example.com/gweb测试以下

集成nagios报告ganglia指标
 tar zxf ganglia-3.4.0.tar.gz
cp /root/ganglia-3.4.0/contrib/check_ganglia.py /usr/local/nagios/libexec/
vim vim /usr/local/nagios/libexec/check_ganglia.py


配置nagios
vim usr/local/nagios/etc/objects/commands.cfg
        define command {
          command_name check_ganglia
          command_line $USER1$/check_ganglia.py -h$HOSTNAME$ -m $ARG1$ -w $ARG2$ -c $ARG3$
        }
vim /usr/local/nagios/etc/objects/templates.cfg
    define service {
     use generic-service
     name ganglia-service
          hostgroup_name ganglia-servers
          service_groups ganglia-metrics
         }                     #配置ganglia模版
vim /usr/local/nagios/etc/objects/hosts.cfg   #添加用户和用户组
   define host {
    use linux-server
    host_name  server4.example.com
    address 172.25.254.4
    }
define hostgroup {
    hostgroup_name ganglia-servers
    alias ganglia-servers
    members server4.example.com
}
vim /usr/local/nagios/etc/objects/services.cfg #设置监控的服务和资源
     define servicegroup {
    servicegroup_name ganglia-metrics
    alias ganglia-metrics
    }
      define service{
    use ganglia-service
    service_description  根分区空闲百分比
    check_commandcheck_ganglia!disk_free_percent_rootfs!20!10
    }
define service{
    use           ganglia-service
    service_description  每分钟系统负载
    check_command          check_ganglia!load_one!4!5
}
define service{
    use                ganglia-service
    service_desc       内存空闲
    check_command      check_ganglia!mem_free!50000!30000   
    }


/usr/local/nagios/bin/nagios -v /usr/local/nagios/etc/nagios.cfg #检测配置
/etc/init.d/nagios reload               #重新加载nagios