DEV Community

Cover image for All You should know about HTTP Host Header Injection
Nathan
Nathan

Posted on

All You should know about HTTP Host Header Injection

You probably did not hear about this injection because it is not too popular, despite the fact that attacks can be very impactful. In this story I will explain briefly what the host header is, how vulnerabilities can arise, then give several examples. Finally, we will see how to protect against them.

What is a Host header?

A Host header is a standard HTTP header field that is used to specify the domain name of the website that a user is trying to access. The Host header is used by the web server to determine which website to show the user.

For example, when a user types in www.example.com, the web server will look at the Host header to figure out which website to show the user. If the Host header is set to www.example.com, the web server will show the user the website that is associated with that domain name.

The Host header is also used by the web server to determine which website to show the user when the user types in an IP address. For example, if the user types in 192.168.1.1, the web server will look at the Host header to figure out which website to show the user.

How Host Header is relevant?

Today, with the growth of cloud computing, it is common that one IP address does not point to a single domain. In other words, multiple websites or resources can be accessed by the IP address.
The host header comes here to help reach the right component the client wants.

Host header can be vulnerable?

An HTTP Host header attack is a type of attack where the attacker sends a request to a server with a fake Host header. This can be used to trick the server into thinking the request is coming from a different domain, or to redirect the request to a different website.
An attacker can even inject a malicious payload that manipulates server-side behavior.
This can potentially lead to several others attacks:

  • SSRF

  • Web cache poisoning

  • SQL , code execution,..

Image description

In order to test for this type of vulnerability, the first step is to supply an arbitrary domain via the host header. To make this attack possible, you need to use a proxy that differentiates between the host header and the target IP. Then you can send an arbitrary host header while sending the request to the relevant website.

The behavior of the target website can vary. You should inspect it attentively. For example, some servers are configured with a fallback if the domain name is not recognized. This case is pretty good for us.

In other scenarios, if there is a firewall CDN or even a load-balancer, they may not know where to forward the request and that can result in an error. Your request will be unable to reach the website's server.

Even though the host header is handled more securely, depending on the configuration of the servers that process incoming requests, the host can theoretically be overridden by injecting other headers.

Sometimes website owners are unaware that certain headers are supported by default, and as a result, they may not be viewed with the same amount of attention.

Keep in mind , many of these vulnerabilities are caused by unsecured configuration of one or more components in the linked infrastructure rather than by insecure code.

These setup concerns can arise when websites include third-party technology into their architecture without fully comprehending the setting options and their security consequences.

Image description

How to test your Application?

This section requires you to be more aware of server behavior.
You need to understand how the website filters the host header.
You should enter into hacker mode and try various techniques. For example, try to bypass by using a malicious sub-domain or different port. Your goal is to reach the target application.
When systems dispute which header is accurate, it might lead to inconsistencies that you may be able to exploit.

  1. Duplicate Host headers.

Add a second Host header can be a possible method, if the website return a 200 status code the case can be investigate further.
When systems dispute on which header is accurate, it might lead to inconsistencies that you may be able to exploit.

Image description

  1. Full URL By supply a absolute URL you can cause an unusual website behavior.


GET https://target-website.com/ HTTP/1.1
Host: malicious payload


Enter fullscreen mode Exit fullscreen mode
  1. Line wrapping Try to indent a line with the "malicioussite" in the Host header.


GET /example HTTP/1.1
    Host: malicioussite
Host: vulnerable-website.com


Enter fullscreen mode Exit fullscreen mode

Some website may block request with multiple host headers, but you can try to bypass this by indenting.
Looking for a 200 response.

  1. Inject host override headers. The "X-Forwarded-Host" is used to identify the original request made by the client. You can sometimes use X-Forwarded-Host to inject your malicious input while circumventing any validation on the Host header itself.

How to exploit this misconfiguration?

HTTP Host header vulnerabilities are often caused by the incorrect belief that the header is not controlled by the user. This provides implicit trust in the Host header, resulting in insufficient validation or escaping of its value.

** Reset password functionality **
If the application includes the host header while creating a new password reset links, an attacker can modify the host header with a domain that behind his control.

** Web cache poisoning **
If the Host header is reflected in the response markup without HTML-encoding, or even used directly in script imports.

GET / HTTP/1.1
Host: attacker.com

The following will be served from the web cache when a victim visits the vulnerable application.

** SSRF **
Sometimes vulnerable host header can lead to Server-side-Forgery. Look for a 200 status code and investigate further.

How to prevent ?

There are a few different ways to remediate host header injection vulnerabilities:

Image description

  1. Use a web application firewall (WAF) to detect and block malicious requests.

  2. Validate user input before processing it. This can be done using a whitelist of allowed characters, or by using a regular expression to check the format of the input.
    This should include comparing it to a whitelist of allowed domains and rejecting or forwarding any requests for unknown sites.

  3. Use HTTPS to encrypt all communication between the user and the server. This will prevent attackers from being able to view or modify the data in transit.

  4. Don't support Host override headers.

If you appreciate reading my posts,don't hesitate to follow me :)
Join our discord server: https://discord.gg/ts7hBBxj

Top comments (1)

Collapse
 
priteshusadadiya profile image
Pritesh Usadadiya

[[..PingBack..]]
This article is curated as a part of #60th Issue of Software Testing Notes Newsletter.