nagios系列(四)之nagios主动方式监控tcp常用的80/3306等端口监控web/syncd/mysql及url服务

nagios主动方式监控tcp服务web/syncd/mysql及url

cd /usr/local/nagios/libexec/

[root@node4 libexec]# ./check_tcp -H 192.168.8.40 -p 80

TCP OK - 0.010 second response time on port 80|time=0.010334s;;;0.000000;10.000000

[root@node4 libexec]# ./check_tcp -H 192.168.8.198 -p 8888

TCP OK - 0.002 second response time on port 8888|time=0.001964s;;;0.000000;10.000000

# ./check_tcp -H 192.168.8.198 -p 22

TCP OK - 0.002 second response time on port 22|time=0.001633s;;;0.000000;10.000000

1.添加关于url链接的监控

①添加自定义的servies目录

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

# mkdir services

[root@node4 objects]# chown -R nagios.nagios services

编辑/usr/local/nagios/etc/nagios.cfg文件

添加配置

cfg_dir=/usr/local/nagios/etc/objects/services



注：

作为别用增加一个services目录，优点很多，在目录下只要符合*.cfg就可以被nagios加载，使用脚本批量部署时非常方便的随机命名配置文件

②在services目录下添加：blog.cfg文件

文件内容可以参考/usr/local/nagios/etc/objects/templates.cfg文件的service部分

# cat services/blog.cfg

define service{

use generic-service

host_name centossz008

service_description blog_url

check_command check_weburl!-I 192.168.8.40

check_period 24x7

max_check_attempts 3

normal_check_interval 2

retry_check_interval 1

contact_groups admins

notification_period 24x7

notification_options w,u,c,r

notification_interval 30

}

③在commands.cfg添加关于checkweburl的命令

# 'check_weburl" command definition

define command{

command_name check_weburl

command_line $USER1$/check_http $ARG1$ -w 10 -c 30

}



检测语法service nagios checkconfig，重载nagios配置service nagios reload

2.对特殊url地址的监控

添加对blog.chinasoft.com的解析

/etc/hosts

192.168.8.40 blog.chinasoft.com

# curl -I http://blog.chinasoft.com
HTTP/1.1 200 OK

Date: Fri, 22 Jul 2016 01:55:44 GMT

Server: Apache/2.4.9 (Unix)

X-Powered-By: PHP/5.4.26

Content-Type: text/html

注意：类似/pma/index.php的这种uri需要添加双引号

define service{

use generic-service

host_name centossz008

service_description phpadmin_url

check_command check_weburl!-H blog.chinasoft.com -u "/pma/index.php"

check_period 24x7

max_check_attempts 3

normal_check_interval 2

retry_check_interval 1

contact_groups admins

notification_period 24x7

notification_options w,u,c,r

notification_interval 30

}

在客户机上执行破坏性操作，会提示错误

/web/a.com/htdocs/pma

# mv index.php index.php.bak

phpadmin_url

WARNING 07-22-2016 10:27:160d 0h 22m 2s3/3HTTP WARNING: HTTP/1.1 404 Not Found - 217 bytes in 0.006 second response
time



3.对tcp服务的监控，可以通过监控端口实现

①对rsyncd服务监控

在客户端node3.chinasoft.com执行：

# touch /etc/rsyncd.conf

[root@node3 ~]# rsync --daemon

[root@node3 ~]# lsof -i :873

COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME

rsync 59285 root 4u IPv4 237676 0t0 TCP *:rsync (LISTEN)

rsync 59285 root 5u IPv6 237677 0t0 TCP *:rsync (LISTEN)

vim /usr/local/nagios/etc/objects/services/blog.cfg

define service{

use generic-service

host_name node3.chinasoft.com

service_description sync_port

check_command check_tcp!837

check_period 24x7

max_check_attempts 3

normal_check_interval 2

retry_check_interval 1

contact_groups admins

notification_period 24x7

notification_options w,u,c,r

notification_interval 30

}



# ../../libexec/check_tcp -H 192.168.8.41 -p 873

TCP OK - 0.001 second response time on port 873|time=0.001093s;;;0.000000;10.000000

发现是定义的端口错误应该是873不是837：

check_command check_tcp!873

②对mysql的3306端口监控

vim /usr/local/nagios/etc/objects/services/blog.cfg

define service{

use generic-service

host_name centossz008

service_description mysql_port

check_command check_tcp!3306

check_period 24x7

max_check_attempts 3

normal_check_interval 2

retry_check_interval 1

contact_groups admins

notification_period 24x7

notification_options w,u,c,r

notification_interval 30

}

4.被动方式监控web的80端口

在被监控端即192.168.8.40的nrpe.cfg中添加命令定义：

command[check_port_80]=/usr/local/nagios/libexec/check_tcp -H 192.168.8.40 -p 80 -w 6 -c 10



编辑主服务器的文件

vim /usr/local/nagios/etc/objects/services/blog.cfg

define service{

use generic-service

host_name centossz008

service_description blog_port_80_beidong

check_command check_nrpe!80

check_period 24x7

max_check_attempts 3

normal_check_interval 2

retry_check_interval 1

contact_groups admins

notification_period 24x7

notification_options w,u,c,r

notification_interval 30

}

发现报错：



排错：

服务端执行

# /usr/local/nagios/libexec/check_tcp -H 192.168.8.40 -p 80 -w 6 -c 10

TCP OK - 0.002 second response time on port 80|time=0.001845s;6.000000;10.000000;0.000000;10.000000

# /usr/local/nagios/libexec/check_nrpe -H 192.168.8.40 -c check_port_80

NRPE: Command 'check_port_80' not defined

原来是定义反了，需要在客户端的nrpe.cfg文件中定义check_port_80命令才OK

5.定义监控组，如监控所有机器的交换分区swap状态

cat /usr/local/nagios/etc/objects/services/servergroup.cfg

define servicegroup{

servicegroup_name Swap Useage

alias Linux Servers

members node3.chinasoft.com,Swap Useage,node4.chinasoft.com,Swap Useage,centossz008,Swap Useage

}



报错：

Error: Could not find a service matching host name 'node3.chinasoft.com' and description 'Swap Useage' (config file '/usr/local/nagios/etc/objects/services/servergroup.cfg', starting on line 1)

/usr/local/nagios/etc/objects/servie.cfg中没有定义check_swap或者描述description 'Swap Useage'不一致，如下图：





备注：

可以通过# tail /usr/local/nagios/var/nagios.log来定位错误