Multiple vulnerabilities have been found in Fortinet's unified threat management (UTM) device that allow remote attackers to hijack sessions and inject malicious code.

Flaws were found in a string of Fortinet UTM devices ranging from FortiGate 5000 down to the 200 series. They have now been patched.
Vulnerability-Lab researcher Benjamin Kunz Mejri found multiple remote persistent web vulnerabilities in the FortiGate UTM appliance application of 13 devices which it graded as high severity with a score (CVSS-SIG) of five.
Remote attackers and low privileged users could inject malicious code to manipulate customer and administrator requests. The vulnerability was found in the “add or tags” module.
“Successful exploitation results in content module request manipulation, execution of persistent malicious script code, session hijacking, account steal and persistent phishing,” the notice said.
Fortinet was notified on 11 May, 2012 and issued patches 25 August, 2012 ahead of the public disclosure earlier this month.
Additional but less severe cross site scripting vulnerabilities were found by Mejri in the Web Application Firewall modules in the same Fortinet UTM devices.
The advisory reported that “remote attackers [could] hijack admin and customer sessions with required client side user interaction. Successful exploitation allows [attackers] to phish user accounts, hijacking sessions, redirect over client side requests or manipulate website context on client-side browser requests”.
Remote attackers could exploit the non-persistent vulnerability with “medium or high required user interaction”.
It was graded as a medium severity vulnerability with a score of 3.5.
Proof of concept code for the three vulnerabilities was available on the respective linked advisories.
This article originally stated that vulnerabilities were found in Dell SonicWALL’s UTM which could be exploited by remote and low-privileged attackers. This was incorrect.