nagios监控windows主机 && linux主机

一、NSclient++与nrpe

nagios对windows的监控主要有两种方法,一种是NSclient++,另一种是通过nrpe的方式来达到监控目的

NSclient++与nrpe最大的区别就是:

1、被监控机上安装有nrpe,并且还有插件,最终的监控是由这些插件来进行的.当监控主机将监控请求发给nrpe后,nrpe调用插件来完成监控.

2、NSclient++则不同,被监控机上只安装NSclient++,没有任何的插件.当监控主机将监控请求发给NSclient++后,NSclient++直接完成监控,所有的监控是由NSclient++完成的。

这也说明了NSclient++的一个很大的问题,不灵活,没有可扩展性.它只能完成自己本身包含的监控操作,不能由一些插件来扩展.好在NSclient++已经做的不错了,基本上可以完全满足我们的监控需要。

NSclient++的原理图

wKiom1PvKYqT3zN_AACJ0zqers8519.jpg

二、部署过程

1、在windows上安装NSclient++

(1)一直下一步

wKiom1PvKuiCrk9AAAAjAmNM0Rg841.png

(2)设置nagios服务器IP地址

wKiom1PvKv2ATA65AACCPbwcXss954.png

(3)检查NSclient++的端口是否成功开启

wKioL1PvLIChR2F7AAAZ7iDZyZk706.png

    如果服务没有开启,就:win+r --> services.msc --> nsclient++ 开启服务即可

(4)防火墙打开tcp 12489端口

2、设置nagios服务器

(1)检测nagios命令是否可以正常监测windows主机

[root@cacti libexec]#  ./check_nt -H 192.168.200.15 -p 12489 -s dianyi123 -v UPTIME
System Uptime - 3 day(s) 12 hour(s) 32 minute(s)
[root@cacti libexec]# 
[root@cacti libexec]# ./check_nt -H 192.168.200.15 -p 12489 -s dianyi123 -v CPULOAD -w 80 -c 90 -l 5,80,90
CPU Load 0% (5 min average) |   '5 min avg Load'=0%;80;90;0;100                     #-w 警告比例 -c 紧急比例  -l(小写L) 表示过去5分钟的平均值,80%为警告,90%为紧急
[root@cacti libexec]# 
[root@cacti libexec]# ./check_nt -H 192.168.200.15 -p 12489 -s dianyi123 -v USEDDISKSPACE -w 80 -c 90 -l C 
C:\ - total: 100.83 Gb - used: 13.71 Gb (14%) - free 87.12 Gb (86%) | 'C:\ Used Space'=13.71Gb;80.66;90.74;0.00;100.83

(2)定义命令、主机、服务

    ①、定义命令

[root@cacti ~]# vim /usr/local/nagios/etc/objects/commands.cfg
define command{
        command_name    check_win
        command_line    $USER1$/check_nt -H "$HOSTADDRESS$" -p 12489 -s dianyi123 -v $ARG1$ $ARG2$
}

    ②、定义主机和服务

    为了方便,主机和监控服务都定义在一个配置文件里面

    首先创在/usr/local/nagios/etc建一个文件夹servers专门保存各服务器的配置文件,然后以服务器IP命名各服务器配置文件

    这样的话,nagios.cfg里面就需要开启对servers目录的支持

[root@cacti etc]# pwd
/usr/local/nagios/etc
[root@cacti etc]# 
[root@cacti etc]# ls
cgi.cfg  htpasswd.users  nagios.cfg  objects  resource.cfg  servers
[root@cacti etc]# 
[root@cacti etc]# vim nagios.cfg
cfg_dir=/usr/local/nagios/etc/servers
[root@cacti etc]# 
[root@cacti etc]# vim servers/192.168.200.15.cfg
define host{
        use                     windows-server            ; Name of host template to use
        host_name               192.168.200.15
        alias                   my computer
        address                 192.168.200.15
        }

#define hostgroup{
#       hostgroup_name  windows-servers ; The name of the hostgroup
#       alias           Windows Servers ; Long name of the group
#       }

define service{
        use                             generic-service
        host_name                       192.168.200.15
        service_description             NSClient++ Version
        check_command                   check_win!CLIENTVERSION
         }

define service{
        use                             generic-service
        host_name                       192.168.200.15
        service_description             Uptime
        check_command                   check_win!UPTIME
        }

define service{
        use                             generic-service
        host_name                       192.168.200.15
        service_description             CPU Load
        check_command                   check_win!CPULOAD!-l 5,80,90
}

define service{
       use                     generic-service
       host_name               192.168.200.15
       service_description     Memory Usage
       check_command           check_win!MEMUSE!-w 80 -c 90
       }

define service{
       use                     generic-service
       host_name               192.168.200.15
       service_description     C:\ Drive Space
       check_command           check_win!USEDDISKSPACE!-l c -w 80 -c 90
       }

define service{
       use                     generic-service
       host_name               192.168.200.15
       service_description     D:\ Drive Space
       check_command           check_win!USEDDISKSPACE!-l d -w 80 -c 90
       }

define service{
       use                     generic-service
       host_name               192.168.200.15
       service_description     E:\ Drive Space
       check_command           check_win!USEDDISKSPACE!-l e -w 80 -c 90
       }

#define service{
#       use                     generic-service
#       host_name               192.168.200.15
#       service_description     W3SVC
#       check_command           check_win!SERVICESTATE!-d SHOWALL -l W3SVC
#       }

define service{
       use                     generic-service
       host_name               192.168.200.15
       service_description     Explorer
       check_command           check_win!PROCSTATE!-d SHOWALL -l Explorer.exe
       }

(3)检查配置文件有无错误

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

如果没有消息,那就是最好的消息,下一步,就可以重启nagios服务了

(4)重启nagios服务

[root@cacti ~]# service nagios restart
Stopping nagios:                                           [  OK  ]
Starting nagios:                                           [  OK  ]

三、nagios监控页面查看主机与服务

1、主机状态

wKioL1PvOGOCMIo0AACZgKI45rU930.png

2、服务状态

wKiom1PvN2nQ0ruDAABa77H1RVc898.png

四、排错阶段

本次部署nagios监控windows主机主要碰到两个问题

1、主机状态(status)是down,而不是正常的up

  原因:这种情况下,一般都是服务器禁ping了,监控服务器是通过ping服务来检查被监控服务器是否在线,当把windows服务器ping的回显请求开启后,监控成功

  解决:win2008:服务器管理器――设置――高级安全windows防火墙――入站规则――找到“文件和打印机共享(回显请求-ICMPv4-in)”右击……选择“启用规则”

2、could not fetch information from server

  当把第1个问题解决掉后,Status是UP起来了,可是所有的服务全部都是could not fetch information from server

wKioL1PvOneR3f1lAABUd5w7O6k288.png

  原因:出现这种状况的原因是因为nagios服务器没有从被监控端服务器上获得相关数据,直接原因就是NSclient++的配置文件中Allowed hosts的IP没有设置正确

  解决:NSclient++的配置文件中 Allowed hosts = nagios服务器IP

wKiom1PvOveg0Z7bAABSyeAbrf0246.png

  当时在安装NSclient++时,我的 Allowed hosts = 192.168.200.105 ,我的设置是正确的,但是为什么会变成15我也不知道为什么


五、nagios监控linux主机

1、服务端定义主机

define host{
        use                     linux-server
        host_name               192.168.200.111
        alias                   linux
        address                 192.168.200.111
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             root_/
        check_command                   check_nrpe!check_xvda!5%!10%
         }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             /dev/xvdb2
        check_command                   check_nrpe!check_xvdb2!5%!10%
         }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             Check Swap
        check_command                   check_nrpe!check_swap
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             total
        check_command                   check_nrpe!check_total_procs
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_load
        check_command                   check_nrpe!check_load
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_tcp_3306
        check_command                   check_tcp!3306
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_users
        check_command                   check_nrpe!check_users
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_mem
        check_command                   check_nrpe!check_mem
        }

define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_mysql
        check_command                   check_nrpe!check_mysql
        }
define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_mysql_slave
        check_command                   check_nrpe!check_mysql_slave
        }
   
define service{
        use                             generic-service
        host_name                       192.168.200.111
        service_description             check_http  192.168.200.111/test.html
        check_command                   check_http!'-u /test.html'     #nagios监控网页状态(如 200),在commands.cfg中有自带check_http命令,也可监控域名!
        }

2、客户端修改:vim /usr/local/nagios/etc/nrpe.cfg

command[check_users]=/usr/local/nagios/libexec/check_users -w 3 -c 5
command[check_load]=/usr/local/nagios/libexec/check_load -w 15,10,5 -c 30,25,20
command[check_xvda]=/usr/local/nagios/libexec/check_disk -w 10% -c 5% -p /dev/xvda
command[check_zombie_procs]=/usr/local/nagios/libexec/check_procs -w 5 -c 10 -s Z
command[check_total_procs]=/usr/local/nagios/libexec/check_procs -w 150 -c 200 
command[check_xvdb2]=/usr/local/nagios/libexec/check_disk -w 10% -c 5% -p /dev/xvdb2   #阿里云
command[check_swap]=/usr/local/nagios/libexec/check_swap -w 20% -c 10%                 #/dev/xvdb1 分区做了swap
command[check_mem]=/usr/bin/sudo /usr/local/nagios/libexec/check_mem -w 20 -c 10 
command[check_mysql]=/usr/local/nagios/libexec/check_mysql -H 192.168.200.111 -unagios -dnagios_monitor -p dianyi123
command[check_mysql_slave]=/usr/local/nagios/libexec/check_mysql_slave

3、在nrpe.cfg配置文件中允许nagios服务器IP

[root@localhost ~]# vim /usr/local/nagios/etc/nrpe.cfg 
allowed_hosts=127.0.0.1,192.168.200.105

4、客户端以独立进程方式启动 nrpe

/usr/local/nagios/bin/nrpe -c /usr/local/nagios/etc/nrpe.cfg -d

5、修改nagios的命令模板

[root@monitor ~]# vim /usr/local/nagios/etc/objects/commands.cfg #添加以下一行
define command {
      command_name  check_nrpe
      command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$
}

否则重启nagios会报错:

Error: Service check command 'check_nrpe!check_total_procs' specified in service 'total' for host '192.168.200.105' not defined anywhere!

6、服务端检测 :

/usr/local/nagios/libexec/check_nrpe -H 192.168.200.111 -c check_sda

六、补充

1、nagios监控windows端口

    基本上socket(收发通信协议)写的程序都会对应一个tcp端口出来,我们只要监控此端口就相当于监控了此程序;如FTP 21,pop 110,smtp 25 这些是常见的tcp端口,常见的端口一般nagios内都有定义的check_nt!,如果不是常见的端口,就需自定义程序的tcp端口。

    在监控之前,要确认端口是打开的,可以在CMD中telnet一下端口

C:\Users\Administrator>telnet 192.168.200.15 3389

(1)定义命令

[root@cacti objects]# vim /usr/local/nagios/etc/objects/commands.cfg
define command{
       command_name    tcp3389
       command_line    $USER1$/check_tcp -H $HOSTADDRESS$ -p 3389 -v CLIENTVERSION
}

(2)定义服务

    主机已定义,主机和服务在一个配置文件里

[root@cacti servers]# vim /usr/local/nagios/etc/servers/192.168.200.15.cfg
define service{
        use                     generic-service
        host_name               192.168.200.15
        service_description     port3389
        check_command           tcp3389
}

(3)重启nagios服务

(4)查看验证

wKiom1P0jpHhVLokAABsKpw7J9w674.png

2、nagios监控linux端口

[root@cacti servers]# pwd
/usr/local/nagios/etc/servers
[root@cacti servers]# 
[root@cacti servers]# vim 192.168.200.18.cfg
define service{
        use                             generic-service
        host_name                       192.168.200.18
        service_description             check_tcp_3306
        check_command                   check_tcp!3306
        }

define service{
        use                             generic-service
        host_name                       192.168.200.18
        service_description             check_tcp_873
        check_command                   check_tcp!873
        }
#
[root@cacti ~]# service nagios restart

############# 如果监听的端口是这样的,而不是 *:5666 这样 ###############
tcp    LISTEN     0      50                             61.138.78.59:7003                                     *:*     
tcp    LISTEN     0      5                                         *:5666                                     *:*  
则需要修改commands中的 $HOSTADDRESS$ 为61.138.78.59,然后修改command_name,再定义服务即可

3、nagios监控mysql主从同步

   判断mysql的主从同步主要还是看那两个线程:Slave_IO线程和Slave_SQL线程,两个都是YES的话,就证明是没有问题的

MariaDB [(none)]> show slave status\G
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.200.17
                  Master_User: doteyplay
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: master-bin.000008
          Read_Master_Log_Pos: 1277
               Relay_Log_File: relay-bin.000025
                Relay_Log_Pos: 1486
        Relay_Master_Log_File: master-bin.000008
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes

第一部分:客户端配置

(1)在被监控的从服务器增加一个用户

MariaDB [(none)]> grant Replication client on *.* to nagios@localhost identified by 'nagios';
Query OK, 0 rows affected (0.00 sec)
MariaDB [(none)]> 
MariaDB [(none)]> flush privileges;
Query OK, 0 rows affected (0.00 sec)

(2)验证命名执行状态

[root@DBSlave ~]# mysql -unagios -pnagios -e "show slave status\G;"  
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.200.17
                  Master_User: doteyplay
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: master-bin.000008
          Read_Master_Log_Pos: 1277
               Relay_Log_File: relay-bin.000025
                Relay_Log_Pos: 1486
        Relay_Master_Log_File: master-bin.000008
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes

(3)编写脚本/usr/local/nagios/libexec/check_mysql_slave(这是监控其作用的核心)

#!/bin/sh 
declare -a    slave_is 
slave_is=($(/usr/local/mysql/bin/mysql -unagios -pnagios    -e "show slave status\G"|grep Running |awk '{print $2}')) 
if [ "${slave_is[0]}" = "Yes" -a "${slave_is[1]}" = "Yes" ] 
     then 
     echo "OK C2-slave is running" 
     exit 0 
else 
     echo "Critical C2-slave is error" 
     exit 2 
fi 
#
[root@DBSlave libexec]# chmod +x check_mysql_slave   #赋予执行权限
[root@DBSlave libexec]# chown  nagios.nagios check_mysql_slave

(4)在从服务器安装 nrpe,然后在配置文件nrpe.cfg加入一行

[root@DBSlave ~]# vim /usr/local/nagios/etc/nrpe.cfg
command[check_mysql_slave]=/usr/local/nagios/libexec/check_mysql_slave

(5)手动执行脚本,观察输出状态

[root@DBSlave libexec]# sh check_mysql_slave 
OK C2-slave is running

(6)检查被监控端的5666端口

[root@DBSlave libexec]# ss -antulp | grep 5666
tcp    LISTEN     0      5                     :::5666                 :::*      users:(("nrpe",26512,5))
tcp    LISTEN     0      5                      *:5666                  *:*      users:(("nrpe",26512,4))
[root@DBSlave libexec]# /usr/local/nagios/bin/nrpe -c /usr/local/nagios/etc/nrpe.cfg -d

第二部分:服务端配置

(1)在监控机上检查是否可成功监控被监控机

[root@cacti ~]# /usr/local/nagios/libexec/check_nrpe -H 192.168.200.18 -c check_mysql_slave
NRPE: Command 'check_mysql_slave' not defined     #遇到问题

排错:NRPE: Command 'check_mysql_slave' not defined

[root@cacti ~]# /usr/local/nagios/libexec/check_nrpe -H 192.168.200.18 
NRPE v2.15

 证明在被监测主机上配置的NRPE已经正常工作,并且监测主机能够通过SSL与被监测主机上的NRPE正常通信。

[root@DBSlave libexec]# ps -ef | grep nrpe
root     10287  9703  0 12:01 pts/1    00:00:00 vim /usr/local/nagios/etc/nrpe.cfg
root     10522  9639  0 12:30 pts/0    00:00:00 grep nrpe
nagios   26512     1  0 Aug15 ?        00:01:09 /usr/local/nagios/bin/nrpe -c /usr/local/nagios/etc/nrpe.cfg -d   #这里的nrpe是以独立进程运行的,而非守护进程。先kill一下nrpe再说
[root@DBSlave libexec]# 
[root@DBSlave libexec]# kill -9 26512   #kill nrpe进程
[root@DBSlave libexec]# 
[root@DBSlave libexec]# ps -ef | grep nrpe
root     10287  9703  0 12:01 pts/1    00:00:00 vim /usr/local/nagios/etc/nrpe.cfg
root     10524  9639  0 12:31 pts/0    00:00:00 grep nrpe       #kill 成功
[root@DBSlave libexec]# 
[root@DBSlave libexec]# 
[root@DBSlave libexec]# /usr/local/nagios/bin/nrpe -c /usr/local/nagios/etc/nrpe.cfg -d   #重启nrpe
[root@DBSlave libexec]# 
[root@DBSlave libexec]# ps -ef | grep nrpe
root     10287  9703  0 12:01 pts/1    00:00:00 vim /usr/local/nagios/etc/nrpe.cfg
nagios   10526     1  0 12:31 ?        00:00:00 /usr/local/nagios/bin/nrpe -c /usr/local/nagios/etc/nrpe.cfg -d
root     10528  9639  0 12:31 pts/0    00:00:00 grep nrpe

再次在监控端测试

[root@cacti ~]# /usr/local/nagios/libexec/check_nrpe -H 192.168.200.18 -c check_mysql_slave
OK C2-slave is running   #终于顺利通过了,就是nrpe进程的事儿

(2)定义主机、服务

[root@cacti servers]# pwd
/usr/local/nagios/etc/servers
[root@cacti servers]# vim 192.168.200.18.cfg
define host{
        use                     linux-server
        host_name               192.168.200.18
        alias                   linux
        address                 192.168.200.18
        }
       
define service{
        use                             generic-service
        host_name                       192.168.200.18
        service_description             check_mysql_slave
        check_command                   check_nrpe!check_mysql_slave
        }

(3)重启nagios服务

(4)查看监控状态

wKioL1P1eVvjb_cnAAAzfT1y6aY028.png

4、nagios通过web界面修改某个服务时报错

例如对某个服务进行临时安排其执行时间,或者不让它发警告,web页面上都有这样的设置.但是常常会有错误信息如下:

Could not open command file '/usr/local/nagios/var/rw/nagiosNaNd' for update!The permissions on the external command file and/or directory may be incorrect. Read the FAQs on how to setup proper permissions.An error occurred while attempting to commit your command for processing.

 (1)修改属组

[root@monitor ~]# chown -R nagios.nagios /usr/local/nagios/var/rw/

(2)把apache用户加入到nagios组

[root@monitor ~]# usermod -G nagios apache

(3)重启服务

[root@monitor ~]# service nagios restart
[root@monitor ~]# service httpd restart




你可能感兴趣的:(nsclient++)