| 注册
请输入搜索内容

热门搜索

Java Linux MySQL PHP JavaScript Hibernate jQuery Nginx
jopen
11年前发布

搭建自己的nagios监控集群.

本文目的: 即使之前未触过nagios,也能按照文中步骤搭建自己的nagios监控集群.

1. Nagios简介

Nagios是一个可运行在Linux/Unix平台之上的开源监视系统,可以用来监视系统运行状态和网络信息。Nagios可以监视所指定的本地或远程主机以及服务,同时提供异常通知功能。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。

搭建自己的nagios监控集群.

2. 相关环境

Host Name IP OS
Arch
duangr-1 192.168.56.10 CentOS 6.4 x86_64
duangr-2 192.168.56.11
CentOS 6.4
x86_64
duangr-3 192.168.56.12
CentOS 6.4
x86_64

3. 部署规划

监控服务主节点(Master)
duangr-1
被监控从节点(Slave) duangr-2, duangr-3

Nagios主节点需要安装:

  • nagios

  • nagios-plugin

  • nrpe

  • php

  • apache

Nagios从节点需要安装:

  • nagios-plugin

  • nrpe

安装路径规划

nagios安装路径
/usr/local/nagios
php安装路径 /usr/local/php
apache安装路径 /usr/local/apache2

4. 代码获取

5. 前提依赖

5.1 主机环境检查(全部节点)

# rpm -q gcc glibc glibc-common gd gd-devel xinetd openssl-devel  gcc-4.4.7-3.el6.x86_64  glibc-2.14.1-6.x86_64  glibc-common-2.14.1-6.x86_64  gd-2.0.35-11.el6.x86_64  package gd-devel is not installed  package xinetd is not installed  openssl-devel-1.0.0-27.el6.x86_64

若有缺失,请先安装. 可通过如下几个镜像网站下载相关安装包:

  • http://rpm.pbone.net/

  • http://mirrors.163.com/centos/6.4/os/x86_64/Packages/

  • http://mirrors.sohu.com/centos/6.4/os/x86_64/Packages/

安装后再次检查如下:

# rpm -q gcc glibc glibc-common gd gd-devel xinetd openssl-devel  gcc-4.4.7-3.el6.x86_64  glibc-2.14.1-6.x86_64  glibc-common-2.14.1-6.x86_64  gd-2.0.35-11.el6.x86_64  gd-devel-2.0.35-11.el6.x86_64  xinetd-2.3.14-38.el6.x86_64  openssl-devel-1.0.0-27.el6.x86_64

6. 编译安装

6.1 创建用户nagios(全部节点)

useradd nagios -d /usr/local/nagios  passwd nagios   (密码自定义)

6.2 安装nagios主程序(主节点安装)

tar -zxf nagios-4.0.2.tar.gz  cd nagios-4.0.2  ./configure --prefix=/usr/local/nagios       make all  make install && make install-init && make install-commandmode && make install-config

将nagios添加为服务

chkconfig --add nagios   chkconfig nagios off  chkconfig --level 35 nagios on  chkconfig --list nagios      nagios          0:关闭  1:关闭  2:关闭  3:启用  4:关闭  5:启用  6:关闭

6.3 安装nagios插件(全部节点安装)

tar -zxf nagios-plugins-1.5.tar.gz  cd nagios-plugins-1.5  ./configure --prefix=/usr/local/nagios --with-nagios-user=nagios --with-nagios-group=nagios        make && make install

如果出现mysql相关的编译错误,是mysql的默认安装路径被修改导致的,调整with-mysql后重新make

./configure --prefix=/usr/local/nagios  --with-mysql=/usr/local/mysql  make && make install

6.4 安装NRPE(全部节点安装)

tar -zxf nrpe-2.15.tar.gz  cd nrpe-2.15  ./configure --enable-command-args  make all  make install-plugin

下面步骤只需要在被监控节点执行

make install-daemon && make install-daemon-config && make install-xinetd

6.4.1 被监控节点配置

如果是被监控节点,需要配置NRPE已守护进程运行(通过xinetd来运行)

1、更改/etc/xinetd.d/nrpe文件,设置允许nagios主节点服务器连接

vi /etc/xinetd.d/nrpe  only_from       = 127.0.0.1 192.168.56.10

2、在/etc/services结尾增加:

nrpe      5666/tcp       # NRPE

3、增加对参数的支持

vi /usr/local/nagios/etc/nrpe.cfg  dont_blame_nrpe=1

4、启动xinetd

service xinetd restart

5、验证nrpe是否监听

netstat -at | grep nrpe

6、测试nrpe是否正常运行

/usr/local/nagios/libexec/check_nrpe -H localhost  NRPE v2.15

6.4.2 主节点配置

如果是监控服务主节点,在全部被监控节点NRPE配置完成后,可以依次做下检测

/usr/local/nagios/libexec/check_nrpe -H 192.168.56.11  NRPE v2.15  /usr/local/nagios/libexec/check_nrpe -H 192.168.56.12  NRPE v2.15

6.5 安装Apache(主节点安装)

tar -zxf httpd-2.2.23.tar.gz  cd httpd-2.2.23  ./configure --prefix=/usr/local/apache2  make && make install

6.6 安装PHP(主节点安装)

cd /export/home/tools/soft/php  tar -zxf php-5.4.10.tar.gz  cd /php-5.4.10  ./configure --prefix=/usr/local/php  --with-apxs2=/usr/local/apache2/bin/apxs  make  && make install

6.7 使用apache 发布PHP的WEB

vi /usr/local/apache2/conf/httpd.conf

....  Listen 80  ....     DirectoryIndex index.html index.php      AddType application/x-httpd-php .php    ....  #setting for nagios  ScriptAlias /nagios/cgi-bin "/usr/local/nagios/sbin"         AuthType Basic       Options ExecCGI       AllowOverride None       Order allow,deny       Allow from all       AuthName "Nagios Access"       AuthUserFile /usr/local/nagios/etc/htpasswd       Require valid-user    Alias /nagios "/usr/local/nagios/share"         AuthType Basic       Options None       AllowOverride None       Order allow,deny       Allow from all       AuthName "nagios Access"       AuthUserFile /usr/local/nagios/etc/htpasswd       Require valid-user  

为web访问时添加用户名和密码(此处用户名为admin,可自定义)

/usr/local/apache2/bin/htpasswd -c /usr/local/nagios/etc/htpasswd admin

启动apache

/usr/local/apache2/bin/apachectl start

访问页面:

   http://192.168.56.10/nagios/

7. 配置Nagios

7.1 配置远程被监控节点

7.1.1 修改配置文件

# su - nagios  $ vi /usr/local/nagios/etc/nrpe.cfg

修改为如下配置内容:

command[check_users]=/usr/local/nagios/libexec/check_users -w $ARG1$ -c $ARG2$  command[check_load]=/usr/local/nagios/libexec/check_load -w $ARG1$ -c $ARG2$  command[check_disk]=/usr/local/nagios/libexec/check_disk -w $ARG1$ -c $ARG2$ -p $ARG3$  command[check_procs]=/usr/local/nagios/libexec/check_procs -w $ARG1$ -c $ARG2$ -s $ARG3$  command[check_procs_args]=/usr/local/nagios/libexec/check_procs  $ARG1$  command[check_swap]=/usr/local/nagios/libexec/check_swap -w $ARG1$ -c $ARG2$

以上监控命令功能:

  • check_users                    监控登陆用户数

  • check_load                     监控CPU负载

  • check_disk                      监控磁盘的使用

  • check_procs                   监控进程数量,状态包括 RSZDT

  • check_swap                    监控SWAP分区使用

7.1.2 重启xinetd服务

配置完上述命令后,重启 xinetd服务

service xinetd restart

7.1.3 校验配置

检查监控命令配置是否ok

/usr/local/nagios/libexec/check_nrpe -H localhost -c check_users  -a 5 10  /usr/local/nagios/libexec/check_nrpe -H localhost -c check_load   -a 15,10,5 30,25,20  /usr/local/nagios/libexec/check_nrpe -H localhost -c check_disk    -a 20% 10% /  /usr/local/nagios/libexec/check_nrpe -H localhost -c check_procs -a 200 400 RSZDT  /usr/local/nagios/libexec/check_nrpe -H localhost -c check_swap  -a 20% 10%

7.2 配置监控服务主节点

7.2.1 cgi.cfg(控制CGI访问的配置文件)

(使用 nagios 用户)

vi /usr/local/nagios/etc/cgi.cfg

修改如下内容,为admin用户增加权限:

default_user_name=admin  authorized_for_system_information=nagiosadmin,admin  authorized_for_configuration_information=nagiosadmin,admin  authorized_for_system_commands=nagiosadmin,admin  authorized_for_all_services=nagiosadmin,admin  authorized_for_all_hosts=nagiosadmin,admin  authorized_for_all_service_commands=nagiosadmin,admin  authorized_for_all_host_commands=nagiosadmin,admin

7.2.2 nagios.cfg(nagios主配置文件)

(使用 nagios 用户)

vi /usr/local/nagios/etc/nagios.cfg

#cfg_file=/export/home/nagios/etc/objects/localhost.cfg      (注释掉)  cfg_dir=/export/home/nagios/etc/servers

主配置文件声明了监控脚本的存储路径为 ./servers, 默认没有此目录,需要手工创建

nagios 会读取 servers 目录下面后缀为.cfg的全部文件作为配置文件

cd /usr/local/nagios/etc  mkdir servers  cd servers

7.2.3 定义监控的主机组

声明一个监控的主机组,将主机环境中提到的三台主机全部加入监控

vi /export/home/nagios/etc/servers/group.cfg

新文件,内容如下:

define hostgroup{     hostgroup_name      duangr-server     alias               duangr Server     members             duangr-1,duangr-2,duangr-3  }

解释下上面的配置:

  • hostgroup_name:    主机组的名称,可随意指定

  • alias:                        主机组别名,可随意指定

  • members:                主机组成员,多个主机名称之前使用逗号分隔.另外主机名称必须与 define host 中host_name 一致.

主机的定义,后面会说到.

7.2.4 定义监控的主机

下面开始定义具体的主机

7.2.4.1 本地主机监控配置

先定义本地主机 duangr-1

vi /export/home/nagios/etc/servers/duangr-1.cfg

新文件,内容如下:

define host{         use                          linux-server         host_name                    duangr-1         alias                        duangr-1         address                      192.168.56.10         }    define service{         use                             local-service         host_name                       duangr-1         service_description             Host Alive         check_command                   check-host-alive         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Users         check_command                   check_local_users!20!50         }  define service{         use                             local-service         host_name                       duangr-1         service_description             CPU         check_command                   check_local_load!5.0,4.0,3.0!10.0,6.0,4.0         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Disk Root         check_command                   check_local_disk!20%!10%!/         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Disk Home         check_command                   check_local_disk!20%!10%!/export/home         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Zombie Procs         check_command                   check_local_procs!5!10!Z         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Total Procs         check_command                   check_local_procs!250!400!RSZDT         }  define service{         use                             local-service         host_name                       duangr-1         service_description             Swap Usage         check_command                   check_local_swap!20!10         }

说明下,由于是此主机也是监控服务主节点所在主机,因此可以使用check_local_* 的相关命令来进行监控.

这个文件中已经将常用的监控项配置进去.

7.2.4.2 远程主机监控配置

再定义远程主机duangr-2和duangr-3

定义远程主机的监控之前,需要先定义check_nrpe命令

vi /usr/local/nagios/etc/objects/commands.cfg

在文件的最后面添加如下内容:

# 'check_nrpe' command definition  define command{         command_name    check_nrpe         command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -t 30 -c $ARG1$         }  define command{         command_name    check_nrpe_args         command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -t 30 -c $ARG1$ -a $ARG2$         }

定义duangr-2主机的监控配置

$ vi /usr/local/nagios/etc/servers/duangr-2.cfg

新文件,内容如下:

define host{         use                     linux-server         host_name               duangr-2         alias                   duangr-2         address                 192.168.56.11         }    define service{         use                             local-service         host_name                       duangr-2         service_description             Host Alive         check_command                   check-host-alive         }  define service{         use                             local-service         host_name                       duangr-2         service_description             Users         check_command                   check_nrpe_args!check_users!5 10         }  define service{         use                             local-service         host_name                       duangr-2         service_description             CPU         check_command                   check_nrpe_args!check_load!15,10,5 30,25,20         }  define service{         use                             local-service         host_name                       duangr-2         service_description             Disk Root         check_command                   check_nrpe_args!check_disk!20% 10% /         }  define service{         use                             local-service         host_name                       duangr-2         service_description             Disk /export/home         check_command                   check_nrpe_args!check_disk!20% 10% /export/home         }  define service{        use                             local-service        host_name                       duangr-2        service_description             Procs Zombie        check_command                   check_nrpe_args!check_procs!5 10 Z        }  define service{        use                             local-service        host_name                       duangr-2        service_description             Procs Total        check_command                   check_nrpe_args!check_procs_args!"-w400 -c600"        }  define service{         use                             local-service         host_name                       duangr-2         service_description             Swap Usage         check_command                   check_nrpe_args!check_swap!20% 10%         }    ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;  ;;  下面是一些常用进程的监控,主要是云平台相关进程  ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;  ;; 监控crond进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: crond         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Ccrond"         }  ;; 监控zookeeper进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: QuorumPeerMain         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.quorum.QuorumPeerMain"         }  ;;监控storm的从节点进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: supervisor         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -adaemon.supervisor"         }  ;; 监控storm的主节点进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: nimbus         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -adaemon.nimbus"         }  ;; 监控MetaQ进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: MetaQ         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -ametamorphosis-server-w"         }  ;; 监控Redis进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: redis-server         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Credis-server"         }  ;; 监控hadoop主节点NameNode进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: NameNode          check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.namenode.NameNode"         }  ;; 监控hadoop主节点SecondaryNameNode进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: SecondaryNameNode         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.namenode.SecondaryNameNode"         }  ;; 监控hadoop主节点ResourceManager进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: ResourceManager         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.resourcemanager.ResourceManager"         }  ;; 监控hadoop从节点DataNode进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: DataNode         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.datanode.DataNode"         }  ;;监控hadoop从节点NodeManager进程  define service{         use                             local-service         host_name                       duangr-2         service_description             PS: NodeManager         check_command                   check_nrpe_args!check_procs_args!"-c1:1 -Cjava -aserver.nodemanager.NodeManager"         }

说明下,由于duangr-2是远程主机,因此使用check_nrpe_args命令来监控.

这个文件中已经将常用的监控项配置进去, 同时还包含了hadoop、storm、zookeeper、metaq、redis的相关进程监控,主要的监控思路是判断进程是否存在。


定义duangr-3主机的监控配置

vi duangr-3.cfg 

内容与duangr-2.cfg类似,只需要修改 host_name 、alias、 address即可.

7.2.4.3 邮件监控

定义监控人邮件地址

vi /usr/local/nagios/etc/objects/contacts.cfg

define contact{         contact_name                    nagiosadmin             ; Short name of user         use                             generic-contact         ; Inherit default values from generic-contact template (defined above)         alias                           Nagios Admin            ; Full name of user         email                           yourname@domain.com                                                                  ; <<***** CHANGE THIS TO YOUR EMAIL ADDRESS ******         }

除了配置监控邮件的接收人外,还要确保:

  • 本主机与邮件服务器互通

  • 本主机SendMail可以使用外部SMTP服务发送邮件

7.2.4.4 校验配置

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

7.2.4.5 启动

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

nagios已经是一个服务,也可以执行如下操作:

service nagios start/stop/restart/status

8. 监控页面

http://192.168.56.10/nagios

搭建自己的nagios监控集群.

搭建自己的nagios监控集群.

来自:http://my.oschina.net/duangr/blog/183160

 本文由用户 jopen 自行上传分享,仅供网友学习交流。所有权归原作者,若您的权利被侵害,请联系管理员。
 转载本站原创文章,请注明出处,并保留原始链接、图片水印。
 本站是一个以用户分享为主的开源技术平台,欢迎各类分享!
 本文地址:https://www.open-open.com/lib/view/open1386727031705.html
系统监控 Nagios