Lucene search
K

MODx 1.0.6 XSS / Abuse Functionality / Denial Of Service

🗓️ 29 Nov 2012 00:00:00Reported by MustLiveType 
packetstorm
 packetstorm
🔗 packetstormsecurity.com👁 45 Views

MODx 1.0.6 XSS / Abuse Functionality / Denial Of Service - Security vulnerabilities in MODx CMS, including CSRF, login enumeration, and DoS attack

Code
`Hello list!  
  
I want to warn you about new security vulnerabilities in MODx. This is the   
second part of the vulnerabilities in this CMS (6 vulnerabilities to   
previous 19 vulnerabilities).  
  
These are Cross-Site Request Forgery, Abuse of Functionality, Denial of   
Service and Insufficient Anti-automation vulnerabilities in MODx. It's about   
0.x and 1.x (Evolution) versions of MODx CMS. In 2.x (Revolution) versions   
of MODx there are part of these holes and part of new holes - I've wrote   
separate advisory concerning it.  
  
-------------------------  
Affected products:  
-------------------------  
  
Vulnerable are MODx 1.0.6 and previous versions.  
  
----------  
Details:  
----------  
  
Cross-Site Request Forgery (WASC-09):  
  
Lack of captcha in login form (http://site/manager/) can be used for   
different attacks - for CSRF-attack to login into account (remote login - to   
conduct attacks on vulnerabilities inside of account), for automated   
entering into account, for phishing and other automated attacks. Which you   
can read about in the article "Attacks on unprotected login forms"   
(http://lists.webappsec.org/pipermail/websecurity_lists.webappsec.org/2011-April/007773.html).   
Bellow mentioned DoS attack is conducting via this CSRF vulnerability.  
  
Abuse of Functionality (Login Enumeration) (WASC-42):  
  
In login form (http://site/manager/) Login Enumeration is possible. If   
blocking isn't triggering after three requests (as can be set at the site),   
then there is no such login in the system, i.e. the blocking works only for   
working logins. The attack is possible, when blocking is turned on.  
  
So by sending three (by default) POST requests per verifiable login, it'll   
possible to pick up working logins. To use for attacks on earlier mentioned   
Brute Force vulnerability.  
  
Exploit:  
  
<body onLoad="document.hack.submit()">  
<form name="hack"   
action="http://site/manager/processors/login.processor.php" method="post">  
<input type="hidden" name="ajax" value="1">  
<input type="hidden" name="username" value="test">  
<input type="hidden" name="password" value="test">  
</form>  
</body>  
  
Abuse of Functionality (WASC-42):  
  
After finding of login with above-mentioned vulnerability it's possible to   
abuse blocking of accounts. After three unsuccessful attempts (as can be set   
at the site) the account is blocking (including account of the   
administrator). By persistent sending of requests to this functionality (by   
three incorrect requests), it's possible to persistently put the account in   
blocked state (including account of the administrator).  
  
Exploit:  
  
<body onLoad="document.hack.submit()">  
<form name="hack"   
action="http://site/manager/processors/login.processor.php" method="post">  
<input type="hidden" name="ajax" value="1">  
<input type="hidden" name="username" value="admin">  
</form>  
</body>  
  
Denial of Service (WASC-10):  
  
At sending of POST request to script   
http://site/manager/processors/login.processor.php, he returns to previous   
page, which again sends to this script. By this way it creates Looped DoS,   
which can overload the server.  
  
About Looped DoS vulnerabilities I've wrote in 2008's articles Looped DoS   
and Classification of DoS vulnerabilities in web applications   
(http://websecurity.com.ua/2663/) and in 2009's article Redirectors: the   
phantom menace   
(http://lists.webappsec.org/pipermail/websecurity_lists.webappsec.org/2009-September/005722.html).   
According to my articles, for conducting of such DoS attack it's needed to   
give users a link to such looped redirect, which will be infinitely sending   
requests and overloading a server (in case of MODx it's possible to create   
unidirectional and bidirectional Looped DoS). In case of POST request it can   
be html form placed on popular site, e.g. in hidden iframe to conduct attack   
invisibly and as long as the page with iframe will be opened by users.  
  
Exploit:  
  
<body onLoad="document.hack.submit()">  
<form name="hack"   
action="http://site/manager/processors/login.processor.php" method="post">  
</form>  
</body>  
  
Insufficient Anti-automation (WASC-21):  
  
In login form (http://site/manager/) there is no protection against   
automated request, which allows to picking up logins (via Abuse of   
Functionality vulnerability) in automated way. As to pick up passwords (via   
Brute Force vulnerability) for picked up logins in automated way. And also   
to conduct automated blocking of revealed accounts.  
  
In password recovery form (http://site/manager/index.php?action=show_form)   
there is no protection against automated request, which allows to picking up   
e-mails of users in automated way.  
  
------------  
Timeline:  
------------   
  
2012.06.28 - announced at my site.  
2012.06.28 - informed developers about the first part of vulnerabilities.  
2012.06.30 - informed developers about the second part of vulnerabilities.  
2012.07.28 - informed developers about vulnerabilities in MODx Revolution   
and reminded about previous two letters.  
2012.07.28-2012.10.31 - during conversation with developers about MODx   
Revolution, I was constantly reminding them, that I've sent them info about   
holes in Evolution and I can resent them, because it was clear that they   
missed it (they only were answering concerning Revolution).  
2012.11.02 - after developers said they want to see this information (missed   
by them in June), I've resent the first two letters to the developers.  
2012.11.24 - disclosed at my site (http://websecurity.com.ua/5929/).  
  
Best wishes & regards,  
MustLive  
Administrator of Websecurity web site  
http://websecurity.com.ua   
  
`

Data

Build on a solid foundation with Vulners data

We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data

Api

Power your application with Vulners API

The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access

App

Assess and manage vulnerabilities with Vulners tools

Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation