Blogs & Stories

SpiderLabs Blog

Attracting more than a half-million annual readers, this is the security community's go-to destination for technical breakdowns of the latest threats, critical vulnerability disclosures and cutting-edge research.

[Honeypot Alert] JCE Joomla Extension Attacks

Our web honeypots picked up some increased exploit attempts for an old Joomla Content Editor (JCE) Extension vulnerability.

Screen Shot 2014-03-25 at 3.13.46 PM

Although this vulnerability is a few years old, botnet owners are heavily scanning for sites that are vulnerable and attempting to exploit them.

Exploit Details

Here are the exploit details from the vulnerability write-up on Exploit-DB:

Screen Shot 2014-03-25 at 3.15.35 PM

Web Honeypot Logs - Attack Sequence

Step 1: Attempt to Upload a PHP Webshell

The first step in this attack is to attempt to upload a webshell/backdoor file to the JCE ImageManager. Here is how the attack looked in the default Apache access_log file:

Screen Shot 2014-03-25 at 3.19.32 PM

This entry shows a POST request to the ImageManager Joomla Plugin using the "com_jce" option. We can also see a tell-tale sign of a malicious program in the User-Agent string value "BOT for JCE". Unfortunately, the default Apache access_log does not actually log the critical POST payload for this request so we can not see what was sent. Fortunately, we also have our ModSecurity WAF installed so we can go to the detailed audit log file to see the complete transaction. Here is the same transaction as logged by ModSecurity:

Screen Shot 2014-03-25 at 3.29.40 PM

We notice a few things here:

  • Under Section "B" - We can see the POST request is "MultiPart" meaning that it is an attempt to upload a file attachment.
  • Under Section "J" - we can see the meta-data about the file attachment. It was a file called "food.gif".
  • Under Section "H" - we see two different rules triggered on this request. One for a missing Accept request header and one from our commercial ModSecurity rules package for the suspicious User-Agent value.

File Attachment Analysis

ModSecurity has the ability to capture and store file attachments. When we inspect the "food.gif" file, we find the following:

Screen Shot 2014-03-26 at 10.06.25 AM

This is an obfuscated PHP file. After decoding, we find this section of code:

Screen Shot 2014-03-26 at 10.09.56 AM

This is a typical webshell/backdoor that allows the attacker to submit OS commands and upload files:

image from blog.spiderlabs.com

Step 2: Rename File Extension

After uploading the webshell file, the attacker next needs to rename the file and change the file extension from ".gif" to ".php" so that it will be executed as code by the application. There are known exploit tools to achieve this task. Here is an example screenshot of a PHP exploit page:

Screen Shot 2014-03-26 at 10.24.19 AM

The source code of the page shows the attack details:

Screen Shot 2014-03-26 at 10.26.38 AM

The highlighted section show the JSON request body content that uses the "folderRename" action to change the uploaded file extension. This is how the actual attack attempt looked in the ModSecurity audit log file:

Screen Shot 2014-03-26 at 10.20.12 AM

Note under Section H that our commercial rule captured this attack attempt.

Step 3: Access the Webshell

The last step in the attack sequence was for the attacker to try and access the webshell file:

Screen Shot 2014-03-26 at 10.35.09 AM
Since the exploit attempt did not work, the honeypot returned a 404.


Attackers are constantly scanning the internet for web sites that have out-dated software. It is critical that web site owners stay vigilant in updating their software and applying patches. In addition, using a Web Application Firewall (WAF) such as ModSecurity can also aid operational security personnel with incident response by providing essential transactional data.