nagios php升级,DIY nagios API自动更新监控主机

nagios 虽然是一个比较不错的监控软件，不过因为其和zabbix相比，API功能几乎就是空白，而在很多企业中仍然以nagios为主要监控工具。在大批量需要自动化监控的环境中，没有API的nagios 监控自动化被严重制肘。不过东西是死的，人是灵活的。本篇就结合一个现网环境要求，通过php写一个简单的更新配置nagios配置文件并可以自动加载配置的API 功能来实现主机的基本监控块可以实现自动添加。在大批量部署时，可以再结合puppt、cfengine、saltstack等自动化工具完成。

一、updatecfg.php文件

nagios更新配置文件的API文件：

$spwd=$_POST["key"];

if ($spwd != '31d860f7-6f7f-48d3-97b3-8407d5083f34') {

exit('Wrong key!');

}

$iid=$_POST["iid"];

$pip=$_POST["pip"];

$hostname=$_POST["hostname"];

if ($iid==""||$pip==""||$hostname=="") {

echo "Some args is null!";

echo "iid: $iid, pip: $pip, hostname: $hostname";

exit;

}

$dir="/etc/nagios3/servers/";

$file="${dir}/${hostname}.cfg";

if (file_exists("${file}")) {

echo "Config file exist! Changing config file...";

$ofile=$file;

$wfile=$file;

}else {

echo "Config file not exist! Generating config file...";

$ofile="${dir}/TEMPLATE";

$wfile=$file;

}

$ofp=fopen($ofile,'r');

$ncont="";

while (!feof($ofp))

{

$buffer=fgets($ofp,4096);

$buffer=preg_replace("/(s+host_name)s+.*/","$1ttt$hostname",$buffer);

$buffer=preg_replace("/(s+alias)s+.*/","$1ttt$iid",$buffer);

$buffer=preg_replace("/(s+address)s+.*/","$1ttt$pip",$buffer);

# echo "$buffer
n";

$ncont.=$buffer;

}

fclose($ofp);

$wfp=fopen($wfile,'w');

fwrite($wfp,$ncont);

fclose($wfp);

echo exec("/etc/init.d/nagios3 reload");

echo "Updata config file success!";

?>

原理很简单，这里定义三个需要传入的变量－－－主机名、别名、IP ，并且在接受client post过来的参数值时，需要先比对一个KEY值，如果KEY值通过就通过模板文件，进行参数替换，并将新的文件保存的相应的路径。最后reload配置文件生效(reload的好处就是，既然配置文件中有错误，其照样能将解析正常的配置文件，不影响整个监控)。

二、client post脚本

这里使用的shell 脚本，内容为：

#!/bin/bash

key="31d860f7-6f7f-48d3-97b3-8407d5083f34"

hostname=$(curl -s http://169.254.169.254/latest/user-data | json.sh | grep -m1 '["hostname"]' | awk '{print $2}' | sed 's/"//g')

pip=$(curl -s http://169.254.169.254/latest/meta-data/public-ipv4)

iid=$(curl -s http://169.254.169.254/latest/meta-data/instance-id)

postdata="hostname=${hostname}&pip=${pip}&iid=${iid}&key=${key}"

curl -u "nagios:abc123" -sk -d ${postdata} http://nagios.361way.com/nagios3/cgi-bin/updatecfg.php

注：由于我这里是AWS主机，通过上面的URL可以取到本机的三个值(实际非AWS主机，可以通过shell 语句获取本机信息) 。同样，可以通过调用ec2metadata命令取值，如下：

# ec2metadata

ami-id: ami-3b879652

ami-launch-index: 0

ami-manifest-path: (unknown)

ancestor-ami-ids: unavailable

availability-zone: us-east-1a

block-device-mapping: ami

ebs1

ephemeral0

root

instance-action: none

instance-id: i-05b3e152

instance-type: c3.xlarge

local-hostname: ip-10-19-255-21.ec2.internal

local-ipv4: 10.19.255.21

kernel-id: aki-88aa75e1

mac: unavailable

profile: default-paravirtual

product-codes: unavailable

public-hostname: ec2-58-85-121-145.compute-1.amazonaws.com

public-ipv4: 58.85.121.145

//不要打我ssh 公钥的主意，IP是伪造的，key也是伪造的，嘿嘿

public-keys: ['ssh-rsa YDUNjq8WYsDh685BSofB4v4Kq+CjsXs3QF+5lERjVjet0PBwibk/Gs0tG3oBE1+HqMtiZaOTeifehnxMQUn4RsFCLqfGy US_E_VPC']

ramdisk-id: unavailable

reserveration-id: unavailable

security-groups: vpc_norules

user-data: {"hostname":"AMZ-IAD-Zabbix-255-21"}

三、TEMPLATE文件

在updatecfg.php文件中调用的TEMPLATE文件内容如下：

##### host define

define host{

use generic-host,host-pnp4nagios

host_name DY-HOSTNAME

alias INSTANCEID

address IPADDRESS

}

##### common service

define service{

use hour-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Disk Space

check_command check_nrpe!check_disk!check_disk -w 20% -c 10%

}

define service{

use generic-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Current Users

check_command check_nrpe!check_users!check_users -w 8 -c 10

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Total Processes

check_command check_nrpe!check_procs!check_perf check_procs -w 250 -c 400

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Zombie Processes

check_command check_nrpe!check_procs!check_perf check_procs -w 5 -c 10 -s Z

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description CF-Execd Processes

check_command check_nrpe!check_procs!check_perf check_procs -w 1: -c 1: -C cf-execd

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Current Load

check_command check_nrpe!check_load!check_load -w 5.0,5.0,5.0 -c 8.0,8.0,8.0

}

define service{

use minute-service

host_name DY-HOSTNAME

service_description SSH

check_command check_ssh

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description CPU Usage

check_command check_nrpe!check_cpu!check_cpu -w 80 -c 90

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Open FD

check_command check_nrpe!check_open_fds!check_open_fds -W 80 -C 90

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Context Switches

check_command check_nrpe!check_context_switches!check_context_switches -w 20000 -c 25000

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Eth0 Traffic

check_command check_nrpe!check_iftraffic!check_iftraffic -i eth0 -b 100 -u m

}

define service{

use minute-service,service-pnp4nagios

host_name DY-HOSTNAME

service_description Memory Usage

check_command check_nrpe!check_mem!check_mem -w 90 -c 95

}

##### unique service

以上代码已上传至本人github站点上。