The huge number of critical new vulnerabilities disclosed by Microsoft on Tuesday *do not* appear to reflect increased failures by their development process. Instead, they are a result of the recent upsurge in organized criminal hacker activity that has already shown up in 450% increases in bank losses due to cyber fraud (since the first half of 2005), broad penetration of US government (and other governments') computers as well as those of military contractor systems. The number of people engaged in cyber crime as a full-time "profession" in Eastern Europe and, especially, in Asia is skyrocketing.
The increasing threat level makes this a particularly important time to ensure your security people have current, hands-on technical skills so they can find the penetrations, clean them up, and build sensible defenses. By far the best way to do that is to encourage them to attend one of the "national" SANS training programs. The national programs have all SANS best instructors, so you learn from the top teachers in the world, and they have big expos, free evening classes, and more. This Friday is the early registration deadline for the next SANS national conference, SANS Network Security 2006 in Las Vegas, Oct. 1-9. The full 37 course matrix is at http://www.sans.org/ns2006/caag.php
A bonus section at the end of this issue from SPI Dynamics covers Port Scanning and Exploitation of Internal Networks that shows how simply visiting an infected web site, whether or not your system is patched, causes an exploitation tool to get on your system and take over systems inside your perimeter. Read the Bonus Section to see how it works.
Alan
@RISK is the SANS community's consensus bulletin summarizing the most important vulnerabilities and exploits identified during the past week and providing guidance on appropriate actions to protect your systems (PART I). It also includes a comprehensive list of all new vulnerabilities discovered in the past week (PART II).
****************** Sponsored By Shavlik Technologies ********************
Free software! Patch & Spyware Management!
Simplify your complex enterprise network security with Shavlik NetChk Protect, Patch & Spyware Management on one easy-to-use console. For a complete security solution, remediate spyware and install patches with Shavlik NetChk Protect.
Download Now: http://www.sans.org/info.php?id=1269
************************************************************************* How Good Are SANS Courses? Ask the alumni.
++ "I have attended courses by several of SANS rivals, and SANS blew them away." - Alton Thompson, US Marines
++ "This is the only conference/training I've ever attended at which I learned techniques and found tools I could apply immediately." - Dwight Leo, Defense Logistics Agency, DLA
++ "This conference provided the opportunity to learn from many of the people who are defining the future direction of information technology" - Larry Anderson, Computer Sciences Corp.
++ "The SANS classes have been uniformly excellent. To learn as much through traditional classes would have entailed weeks away from work." - David Ritch, Department of Defense
*************************************************************************
************************ Sponsored Links: *****************************
1) FREE WebcastFeaturing Penetration Testing with CORE IMPACT. Wednesday, August 16 at 1:00 PM EDT (1700 UTC/GMT) http://www.sans.org/info.php?id=1270
2) CDI 8570 Training Event 16-22 OCTOBER, Silver Spring, MD SEC309, To earn the certifications now required of all contractors and military personnel with security responsibility: Security Essentials (GSEC), SANS Training for CISSP, And SANS Security Leadership (GSLC) http://www.sans.org/info.php?id=1271
*************************************************************************
Part I is compiled by Rob King and Rohit Dhamankar at TippingPoint, a division of 3Com, as a by-product of that company's continuous effort to ensure that its intrusion prevention products effectively block exploits using known vulnerabilities. TippingPoint's analysis is complemented by input from a council of security managers from twelve large organizations who confidentially share with SANS the specific actions they have taken to protect their systems. A detailed description of the process may be found at http://www.sans.org/newsletters/cva/#process
Description: The Microsoft Windows Server Service, used to provide various operating system and networking services, suffers from a remotely-exploitable buffer overflow in its RPC (Remote Procedure Call) interface. By sending a specially-crafted "NetpwPathCanoncalize" RPC request, an attacker could exploit this buffer overflow to execute arbitrary code and take a complete control of the vulnerable system. Exploit code has been publicly posted. The exploit code is being actively used by a few bots to infect Windows 2000 and XP SP1 systems. Note that CERT and SANS have also seen this vulnerability being exploited prior to the release of this security bulletin.
Status: Microsoft has released a patch that is referenced in the security bulletin MS06-040. Users are advised to apply this patch immediately. eEye and other scanning tools can be used to locate vulnerable systems in a network. A workaround is to block TCP and UDP ports 445 and 139 at the network perimeter.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: Microsoft Internet Explorer contains multiple remotely-exploitable vulnerabilities. These vulnerabilities include arbitrary remote code execution with the privileges of the current user, information disclosure, and the ability to execute arbitrary FTP commands embedded in an FTP URL.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: Microsoft Windows fails to properly parse MHTML URLs in links. MHTML is an extension to HTML (the HyperText Markup Language; the language used to write most web pages) that allows embedded objects and metadata. A malicious web page containing a specially-crafted MHTML link could exploit this vulnerability and execute arbitrary code with the privileges of the current user. Because this flaw exists in a core library, applications other than web browsers and email clients may be affected.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: An ActiveX component used by the Microsoft HTML help system contains a remote code execution vulnerability. A malicious web page that instantiates the vulnerable component could trigger this vulnerability and execute arbitrary code with the privileges of the current user. Users can limit the impact of this vulnerability by preventing the vulnerable component from being instantiated inside Internet Explorer via the "killbit" mechanism for CLSID "{52a2aaae-085d-4187-97ea-8c30db990436}".
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. All sites plan to deploy the patch during their next regularly scheduled system update process.
Description: Clam AntiVirus (ClamAV), a popular open source virus scanning engine, contains a remotely-exploitable buffer overflow. By sending a specially-crafted UPX-compressed executable as an attachment to an email message through a server running ClamAV, an attacker could execute arbitrary code with the privileges of the ClamAV process. No user interaction is necessary to exploit this vulnerability. The default configuration of ClamAV is believed to be vulnerable. Users can limit the impact of this vulnerability by disabling the scanning of UPX-compressed executables. Note that a proof-of-concept is publicly available.
Status: ClamAV confirmed, updates available.
Council Site Actions: The affected software and/or configuration are not in production or widespread use, or are not officially supported at any of the council sites. They reported that no action was necessary.
Description: The McAfee Subscription Manager ActiveX component, installed along with all McAfee's Home and Home Business products, contains a remotely-exploitable buffer overflow. A malicious web page that instantiates the vulnerable component could exploit a buffer overflow in the McSubMgr.dll component and execute arbitrary code with the privileges of the current user. Note that technical details for this vulnerability are publicly available, as well as a simple proof-of-concept. Users may be able to limit the impact of this vulnerability by disallowing the instantiation of the vulnerable component in Internet Explorer via Microsoft's "killbit" mechanism for the CLSID "{9BE8D7B2-329C-442A-A4AC-ABA9D7572602}".
Status: McAfee confirmed, updates available.
Council Site Actions: Only one of the responding council sites is using the affected software and only on a small scale. They are relying on the ability of the end user to take the update directly from the vendor.
Description: A malicious web site can exploit a cross-site scripting vulnerability in Microsoft Windows. This vulnerability allows access to local HTML resource files in the Microsoft Management Console library; access to these files allows remote users to execute arbitrary commands with the privileges of the current user. Users can limit the impact of the vulnerability by disabling Microsoft "Active Scripting" for the "My Computer" zone. Note that this may affect operating system functionality. Users are also advised to read email messages in plain text.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. All sites plan to deploy the patch during their next regularly scheduled system update process.
Description: Failure to properly handle specially-crafted filenames in SMB and WebDAV fileshares could allow an attacker to execute arbitrary code with the privileges of the current user. This flaw can be exploited by visiting a fileshare containing a file with a specially-crafted name and double-clicking somewhere in that window. If the filename contains the GUID (Globally Unique Identifier) of an application, that application will be executed. Note that users must first visit a malicious share and then double-click in the newly-opened window. Technical details and a proof-of-concept for this vulnerability have been publicly posted.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. All sites plan to deploy the patch during their next regularly scheduled system update process.
Description: Clicking on a specially crafted link in an email message or Office document could result in arbitrary code execution with the privileges of the current user. This is due to multiple flaws in the Hyperlink Object Library, used to parse and manipulate hyperlinks. Note that, for at least one of the vulnerabilities, the provided link must go to a live website, limiting the life span of any potential malware based on this vulnerability. Note that the technical details for one of the vulnerabilities has been publicly posted. Because this is a flaw in an operating system library, other applications may be affected.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: The Barracuda spam firewall appliance, a popular enterprise anti-spam appliance, contains a remote command injection vulnerability. By sending specially-crafted requests to the "/cgi-bin/preview_email.cgi" script on the appliance, an unauthenticated attacker could execute arbitrary commands with the administrative privileges. Technical details for this vulnerability are publicly available. Users are advised to block access to the barracuda administrative interface at the network perimeter.
Status: Barracuda confirmed, updates available.
Council Site Actions: Only one of the reporting council sites is using the affected software. They use it for incoming mail for a large user population. They noticed that the Matthew Hall bugtraq posting on August 4 says "It was noted that the original file input sanitation vulnerability seems to have been 'silently' fixed by Barracuda Networks (as of 11pm GMT 03/08/06), which mitigates the attacks above." Thus they plan no action at this time.
Description: Multiple flaws in Microsoft's DNS (Domain Name System) client implementation allow attackers to take complete control of the vulnerable system. The first flaw may be exercised by a specially-crafted web page that forces the vulnerable system to call the affected API, an attacker could execute arbitrary code with SYSTEM privileges. The second flaw can be triggered by forcing the vulnerable system to look up a specially-crafted record on a malicious nameserver; this also allows arbitrary code execution with SYSTEM privileges. Users can mitigate the impact of the first vulnerability by disabling the "Autodial.DLL" library; this will prevent access to the vulnerable API. The second vulnerability can be mitigated by blocking ATMA, TXT, X25, HINFO, and ISDN DNS record responses at the network perimeter.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: Opening a malicious Microsoft Office document containing a specially-crafted Visual Basic document properties could result in arbitrary code execution with the privileges of the current user. In most common configurations, Office documents are not opened automatically. Users are advised to not open documents received from untrusted sources. Note that the vulnerable document properties are not currently publicly known.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. All sites plan to deploy the patch during their next regularly scheduled system update process.
Description: Opening a malicious Microsoft PowerPoint document containing specially-crafted document properties could result in arbitrary arbitrary code execution with the privileges of the current user. In most common configurations other than PowerPoint 2000, PowerPoint documents are not opened automatically. Users are advised to not open documents received from untrusted sources. Some technical details for this vulnerability have been publicly posted. This update patches a vulnerability mentioned in a previous @RISK entry.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. All sites plan to deploy the patch during their next regularly scheduled system update process.
Description: Microsoft Windows is vulnerable to an remote code execution vulnerability through the Windows kernel's exception handling facility. This attack could be exploited by a malicious web site. No other technical details about this vulnerability are publicly available.
Status: Microsoft confirmed, updates available.
Council Site Actions: All reporting council sites are responding to all Microsoft issues. Several sites are addressing this issue on a fast track schedule and have already pushed the patches. Others are deploying the patch during their next regularly scheduled system update process. A few sites are addressing desktops on an urgent basis and servers on a standard basis.
Description: SAP Internet Graphics Service, a component of the SAP suite that allows graphical output, contains a remotely-exploitable buffer overflow. No technical details for this vulnerability have been publicly posted, other than that the vulnerability is exploitable via the HTTP protocol.
Status: SAP confirmed, updates available.
Council Site Actions: The affected software and/or configuration are not in production or widespread use, or are not officially supported at any of the council sites. They reported that no action was necessary.
Description: Ruby-on-Rails, a popular Ruby-based web application development platform, is vulnerable to an undisclosed vulnerability. No technical details about this vulnerability have been publicly posted, except that the issue is considered critical by the Ruby-on-Rails team. Because Ruby-on-Rails is open source software, some technical details may be easily available by analyzing the source code.
Status: Ruby-on-Rails confirmed, updates available.
Council Site Actions: One of the reporting council sites is using the affected software; however it is not supported by their central IT department. In most cases, the owner of the system is active in the Rails community and was able to receive the vendor notice and complete an update within a day or two.
This list is compiled by Qualys ( www.qualys.com ) as part of that company's ongoing effort to ensure its vulnerability management web service tests for all known vulnerabilities that can be scanned. As of this week Qualys scans for 5121 unique vulnerabilities. For this special SANS community listing, Qualys also includes vulnerabilities that cannot be scanned remotely.
*************************************************************************
BONUS SECTION: Drive By Port Scanning and Exploitation of Internal Networks
SPI Labs has discovered a technique to use JavaScript to portscan an internal network, fingerprint all the web-enabled devices found, and send attacks or commands to those devices. All the code uses parts of the JavaScript standard that are almost ten years old. Accordingly, the code can execute in nearly any Web browser on nearly any platform when a user opens a Webpage that contains the JavaScript. Since this is not exploiting any browser bug or vulnerability, there is no patch or defense for the end user other than turning off JavaScript support in the browser. The code can be part of a Cross-site scripting (XSS) attack payload, thereby increasing the damage XSS can do.
Simply viewing a page with an embedded scanner will download the JavaScript along with the HTML to a user's browser, automatically executing the code. The scanner can be included in a site an attacker controls, or injected into popular sites using XSS vulnerabilities. The scanner finds targets by implementing a "ping" feature using the JavaScript Image object and an IFrame tag. Uses a blend of these two objects allows the scanner to quickly detect hosts and confirm they are serving HTTP content. Once the scanner has detected a host with a web interface, the scanner tries to fingerprint the Web server to determine its type and version number. This is done using the Image object to retrieve graphics from well known locations on the device. For example, most Microsoft IIS Web server's have an image /pagerror.gif that is 36 by 48 pixels in size, Linksys WRK54G wireless routers have an image /UI_Linksys.gif that is 165 by 57 pixels, and Plone wiki applications have an image /plone_powered.gif that is 80 by 15 pixels. Once the scanner knows what applications exist on the intranet, it can send attacks to exploit known vulnerabilities in the applications. By dynamically building HTML forms and automatically submitting them, the scanner can send attacks using either GET or POST against the application. At the very least, the information collected from scanning and fingerprinting can be sent to the attacker to assist in planning another attack.
SPI Labs has created a proof of concept web page that implements the detection and fingerprinting functionality of a full scanner. This site is available to the public and is listed at the end of this article. The scanner does not automatically start scanning or attacking any internal applications.
Most of these traditional XSS attacks target the Website where the XSS vulnerability exists and the damage of the attack is limited by the features of that Website. For example, session hijacking is only damaging if the site that has the XSS vulnerability actually issues session state and does something meaningful with it. The danger is that scanning and attacking internal applications or systems targets the end user. This means any XSS vulnerability on any site can be used to attack the end user, regardless of the features of the vulnerable site. There is no longer any such thing as a harmless XSS vulnerability.
More information: Complete Whitepaper: http://www.spidynamics.com/assets/documents/JSportscan.pdf Proof of Concept: http://www.spidynamics.com/spilabs/js-port-scan/ Upcoming BlackHat Presentation, Jeremiah Grossman, WhiteHat Security: http://www.blackhat.com/html/bh-usa-06/bh-usa-06-speakers.html#Grossman Upcoming BlackHat Presentation, Billy Hoffman, SPI Dynamics: http://www.blackhat.com/html/bh-usa-06/bh-usa-06-speakers.html#Hoffman2
______________________________________________________________________
(c) 2006. All rights reserved. The information contained in this newsletter, including any external links, is provided "AS IS," with no express or implied warranty, for informational purposes only. In some cases, copyright for material in this newsletter may be held by a party other than Qualys (as indicated herein) and permission to use such material must be requested from the copyright owner.
==end==
Subscriptions: @RISK is distributed free of charge to people responsible for managing and securing information systems and networks. You may forward this newsletter to others with such responsibility inside or outside your organization.