CORE Security Technologies reports:

A format string error has been found on the vinagre_utils_show_error() function that can be exploited via commands issued from a malicious server containing format string specifiers on the VNC name.

In a web based attack scenario, the user would be required to connect to a malicious server. Successful exploitation would then allow the attacker to execute arbitrary code with the privileges of the Vinagre user.

Marc Schoenefeld and Steve Milner of RedHat SRT and Peter Allor of IBM ISS report:

XSS vulnerability with URLPARAM variable

SEARCH variable allows arbitrary shell command execution

Entry for CVE-2008-5619 says:

html2text.php in RoundCube Webmail (roundcubemail) 0.2-1.alpha and 0.2-3.beta allows remote attackers to execute arbitrary code via crafted input that is processed by the preg_replace function with the eval switch.

MySQL Team reports:

Additional corrections were made for the symlink-related privilege problem originally addressed. The original fix did not correctly handle the data directory pathname if it contained symlinked directories in its path, and the check was made only at table-creation time, not at table-opening time later.

A trapkit reports:

MPlayer contains a stack buffer overflow vulnerability while parsing malformed TwinVQ media files. The vulnerability may be exploited by a (remote) attacker to execute arbitrary code in the context of MPlayer.

Secunia reports:

A security issue has been reported in Ampache, which can be exploited by malicious, local users to perform certain actions with escalated privileges.

The security issue is caused due to the "gather-messages.sh" script handling temporary files in an insecure manner. This can be exploited via symlink attacks to overwrite arbitrary files with the privileges of the user running the script.

The Opera Team reports:

Manipulating certain text-area contents can cause a buffer overflow, which may be exploited to execute arbitrary code.

Certain HTML constructs can cause the resulting DOM to change unexpectedly, which triggers a crash. To inject code, additional techniques will have to be employed.

Exceptionally long host names in file: URLs can cause a buffer overflow, which may be exploited to execute arbitrary code. Remote Web pages cannot refer to file: URLs, so successful exploitation involves tricking users into manually opening the exploit URL, or a local file that refers to it.

When Opera is previewing a news feed, some scripted URLs are not correctly blocked. These can execute scripts which are able to subscribe the user to any feed URL that the attacker chooses, and can also view the contents of any feeds that the user is subscribed to. These may contain sensitive information.

Built-in XSLT templates incorrectly handle escaped content and can cause it to be treated as markup. If a site accepts content from untrusted users, which it then displays using XSLT as escaped strings, this can allow scripted markup to be injected. The scripts will then be executed in the security context of that site.

The MediaWiki development team reports:

Certain unspecified input is not properly sanitised before being returned to the user. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site.

Certain unspecified input related to uploads is not properly sanitised before being used. This can be exploited to inject arbitrary HTML and script code, which will be executed in a user's browser session in context of an affected site when a malicious data is opened. Successful exploitation may require that uploads are enabled and the victim uses an Internet Explorer based browser.

Certain SVG scripts are not properly sanitised before being used. This can be exploited to inject arbitrary HTML and script code, which will be executed in a user's browser session in context of an affected site when a malicious data is opened. Successful exploitation may require that SVG uploads are enabled and the victim uses a browser supporting SVG scripting.

The application allows users to perform certain actions via HTTP requests without performing any validity checks to verify the requests. This can be exploited to perform certain operations when a logged in user visits a malicious site.

The Drupal Project reports:

The update system is vulnerable to Cross site request forgeries. Malicious users may cause the superuser (user 1) to execute old updates that may damage the database.

When an input format is deleted, not all existing content on a site is updated to reflect this deletion. Such content is then displayed unfiltered. This may lead to cross site scripting attacks when harmful tags are no longer stripped from 'malicious' content that was posted earlier.

The Mozilla Foundation reports:

MFSA 2008-69 XSS vulnerabilities in SessionStore

MFSA 2008-68 XSS and JavaScript privilege escalation

MFSA 2008-67 Escaped null characters ignored by CSS parser

MFSA 2008-66 Errors parsing URLs with leading whitespace and control characters

MFSA 2008-65 Cross-domain data theft via script redirect error message

MFSA 2008-64 XMLHttpRequest 302 response disclosure

MFSA 2008-62 Additional XSS attack vectors in feed preview

MFSA 2008-61 Information stealing via loadBindingDocument

MFSA 2008-60 Crashes with evidence of memory corruption (rv:1.9.0.5/1.8.1.19)

The phpMyAdmin Team reports:

A logged-in user can be subject of SQL injection through cross site request forgery. Several scripts in phpMyAdmin are vulnerable and the attack can be made through table parameter.

PHP Developers reports:

Due to a security bug found in the PHP 5.2.7 release, it has been removed from distribution. The bug affects configurations where magic_quotes_gpc is enabled, because it remains off even when set to on.

Secunia reports:

A vulnerability has been reported in Wireshark, which can be exploited by malicious people to cause a DoS.

The vulnerability is caused due to an error in the SMTP dissector and can be exploited to trigger the execution of an infinite loop via a large SMTP packet.

Secunia reports:

Some vulnerabilities have been reported in PHP, where some have an unknown impact and others can potentially be exploited by malicious people to cause a DoS (Denial of Service) or compromise a vulnerable system.

An input validation error exists within the "ZipArchive::extractTo()" function when extracting ZIP archives. This can be exploited to extract files to arbitrary locations outside the specified directory via directory traversal sequences in a specially crafted ZIP archive.

An error in the included PCRE library can be exploited to cause a buffer overflow.

The problem is that the "BG(page_uid)" and "BG(page_gid)" variables are not initialized. No further information is currently available.

The problem is that the "php_value" order is incorrect for Apache configurations. No further information is currently available.

An error in the GD library can be exploited to cause a crash via a specially crafted font file.

Debian reports:

Faxspool in mgetty 1.1.36 allows local users to overwrite arbitrary files via a symlink attack on a /tmp/faxsp.#### temporary file.

Secunia reports:

The security issue is caused due to an input validation error when processing script names. This can be exploited to read or modify arbitrary files having ".sieve" extensions via directory traversal attacks, with the privileges of the attacker's user id.

Secunia reports:

Input passed via the "habari_username" parameter when logging in is not properly sanitised before being returned to the user. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site.

Tobias Klein from TrapKit reports:

The VLC media player contains an integer overflow vulnerability while parsing malformed RealMedia (.rm) files. The vulnerability leads to a heap overflow that can be exploited by a (remote) attacker to execute arbitrary code in the context of VLC media player.

Secunia reports:

EgiX has discovered a vulnerability in Mantis, which can be exploited by malicious users to compromise a vulnerable system.

Input passed to the "sort" parameter in manage_proj_page.php is not properly sanitised before being used in a "create_function()" call. This can be exploited to execute arbitrary PHP code.

Secunia reports:

Some vulnerabilities have been reported in Mantis, which can be exploited by malicious users to compromise a vulnerable system and malicious people to conduct cross-site scripting and request forgery attacks.

Input passed to the "filter_target" parameter in return_dynamic_filters.php is not properly sanitised before being returned to a user. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site.

A vulnerability is caused due to the application allowing users to perform certain actions via HTTP requests without performing any validity checks to verify the request. This can be exploited to e.g. add a new user with administrative privileges by enticing a logged-in administrator to visit a malicious site.

Input passed to the "value" parameter in adm_config_set.php is not properly sanitised before being used in an "eval()" statement. This can be exploited to e.g. execute arbitrary PHP commands via a specially crafted request.

Input passed to the "language" parameter in account_prefs_update.php is not properly verified before being used to include files. This can be exploited to include arbitrary files from local resources.

Squirrelmail team reports:

An issue was fixed that allowed an attacker to send specially- crafted hyperlinks in a message that could execute cross-site scripting (XSS) when the user viewed the message in SquirrelMail.

The OpenOffice Team reports:

A security vulnerability with the way OpenOffice 2.x process WMF files may allow a remote unprivileged user who provides a StarOffice/StarSuite document that is opened by a local user to execute arbitrary commands on the system with the privileges of the user running StarOffice/StarSuite. No working exploit is known right now.

A security vulnerability with the way OpenOffice 2.x process EMF files may allow a remote unprivileged user who provides a StarOffice/StarSuite document that is opened by a local user to execute arbitrary commands on the system with the privileges of the user running StarOffice/StarSuite. No working exploit is known right now.

Secunia reports:

Input passed via the HTTP "Host" header is not properly sanitised before being used. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site if malicious data is viewed.

Samba Team reports:

Samba 3.0.29 and beyond contain a change to deal with gcc 4 optimizations. Part of the change modified range checking for client-generated offsets of secondary trans, trans2 and nttrans requests. These requests are used to transfer arbitrary amounts of memory from clients to servers and back using small SMB requests and contain two offsets: One offset (A) pointing into the PDU sent by the client and one (B) to direct the transferred contents into the buffer built on the server side. While the range checking for offset (B) is correct, a cut and paste error lets offset (A) pass completely unchecked against overflow.

The buffers passed into trans, trans2 and nttrans undergo higher-level processing like DCE/RPC requests or listing directories. The missing bounds check means that a malicious client can make the server do this higher-level processing on arbitrary memory contents of the smbd process handling the request. It is unknown if that can be abused to pass arbitrary memory contents back to the client, but an important barrier is missing from the affected Samba versions.

Secunia reports:

A security issue has been reported in hplip, which can be exploited by malicious, local users to cause a DoS.

The security issue is caused due to an error within hpssd.py when parsing certain requests. This can be exploited to crash the service by sending specially crafted requests to the default port 2207/TCP.

CUPS reports:

The PNG image reading code did not validate the image size properly, leading to a potential buffer overflow (STR #2974)

Secunia reports:

A vulnerability has been discovered in imlib2, which can be exploited by malicious people to potentially compromise an application using the library.

The vulnerability is caused due to a pointer arithmetic error within the "load()" function provided by the XPM loader. This can be exploited to cause a heap-based buffer overflow via a specially crafted XPM file.

Successful exploitation may allow execution of arbitrary code.

Secunia reports:

A boundary error exists within http_parse_sc_header() in lib/http.c when parsing an overly long HTTP header starting with "Zwitterion v".

A boundary error exists within http_get_pls() in lib/http.c when parsing a specially crafted pls playlist containing an overly long entry.

A boundary error exists within http_get_m3u() in lib/http.c when parsing a specially crafted m3u playlist containing an overly long "File" entry.

The mantis Team reports:

When configuring a web application to use only ssl (e. g. by forwarding all http-requests to https), a user would expect that sniffing and hijacking the session is impossible. Though, for this to be secure, one needs to set the session cookie to have the secure flag. Else the cookie will be transferred through http if the victim's browser does a single http-request on the same domain.

Timo Sirainen reports in dovecot 1.1.4 release notes:

ACL plugin fixes: Negative rights were actually treated as positive rights. 'k' right didn't prevent creating parent/child/child mailbox. ACL groups weren't working.

Secunia reports:

Two vulnerabilities have been reported in Libxml2, which can be exploited by malicious people to cause a DoS (Denial of Service) or to potentially compromise an application using the library.

1) An integer overflow error in the "xmlSAX2Characters()" function can be exploited to trigger a memory corruption via a specially

Successful exploitation may allow execution of arbitrary code, but requires e.g. that the user is tricked into processing an overly large XML file (2GB or more).

2) An integer overflow error in the "xmlBufferResize()" function can be exploited to trigger the execution of an infinite loop. The vulnerabilities are reported in version 2.7.2.

Other versions may also be affected.

Andreas Kurtz reports:

The jabber server Openfire (<= version 3.6.0a) contains several serious vulnerabilities. Depending on the particular runtime environment these issues can potentially even be used by an attacker to execute code on operating system level.

1. Authentication bypass - This vulnerability provides an attacker full access to all functions in the admin webinterface without providing any user credentials. The Tomcat filter which is responsible for authentication could be completely circumvented.
2. SQL injection - It is possible to pass SQL statements to the backend database through a SQL injection vulnerability. Depending on the particular runtime environment and database permissions it is even possible to write files to disk and execute code on operating system level.
3. Multiple Cross-Site Scripting - Permits arbitrary insertion of HTML- and JavaScript code in login.jsp. An attacker could also manipulate a parameter to specify a destination to which a user will be forwarded to after successful authentication.

Florian Grandel reports:

I have not had the time to analyze all of syslog-ng code. But by reading the code section near the chroot call and looking at strace results I believe that syslog-ng does not chdir to the chroot jail's location before chrooting into it.

This opens up ways to work around the chroot jail.

Ulf Harnhammar of Secunia Research reports:

Stack-based buffer overflow in the read_special_escape function in src/psgen.c in GNU Enscript 1.6.1 and 1.6.4 beta, when the -e (aka special escapes processing) option is enabled, allows user-assisted remote attackers to execute arbitrary code via a crafted ASCII file, related to the setfilename command.

SecurityFocus reports:

GnuTLS is prone to a security-bypass vulnerability because the application fails to properly validate chained X.509 certificates. Successfully exploiting this issue allows attackers to perform man-in-the-middle attacks by impersonating trusted servers. Unsuspecting users may be under a false sense of security that can aid attackers in launching further attacks.

Wes Hardaker reports through sourceforge.net forum:

SECURITY ISSUE: A bug in the getbulk handling code could let anyone with even minimal access crash the agent. If you have open access to your snmp agents (bad bad bad; stop doing that!) or if you don't trust everyone that does have access to your agents you should updated immediately to prevent potential denial of service attacks.

Description at cve.mitre.org additionally clarifies:

Integer overflow in the netsnmp_create_subtree_cache function in agent/snmp_agent.c in net-snmp 5.4 before 5.4.2.1, 5.3 before 5.3.2.3, and 5.2 before 5.2.5.1 allows remote attackers to cause a denial of service (crash) via a crafted SNMP GETBULK request, which triggers a heap-based buffer overflow, related to the number of responses or repeats.

The Mozilla Foundation reports:

MFSA 2008-58 Parsing error in E4X default namespace

MFSA 2008-57 -moz-binding property bypasses security checks on codebase principals

MFSA 2008-56 nsXMLHttpRequest::NotifyEventListeners() same-origin violation

MFSA 2008-55 Crash and remote code execution in nsFrameManager

MFSA 2008-54 Buffer overflow in http-index-format parser

MFSA 2008-53 XSS and JavaScript privilege escalation via session restore

MFSA 2008-52 Crashes with evidence of memory corruption (rv:1.9.0.4/1.8.1.18)

MFSA 2008-51 file: URIs inherit chrome privileges when opened from chrome

MFSA 2008-50 Crash and remote code execution via __proto__ tampering

MFSA 2008-49 Arbitrary code execution via Flash Player dynamic module unloading

MFSA 2008-48 Image stealing via canvas and HTTP redirect

MFSA 2008-47 Information stealing via local shortcut files

MFSA 2008-46 Heap overflow when canceling newsgroup message

MFSA 2008-44 resource: traversal vulnerabilities

MFSA 2008-43 BOM characters stripped from JavaScript before execution

MFSA 2008-42 Crashes with evidence of memory corruption (rv:1.9.0.2/1.8.1.17)

MFSA 2008-41 Privilege escalation via XPCnativeWrapper pollution

MFSA 2008-38 nsXMLDocument::OnChannelRedirect() same-origin violation

MFSA 2008-37 UTF-8 URL stack buffer overflow

CVE reports:

Heap-based buffer overflow in the decodeMP4file function (frontend/main.c) in FAAD2 2.6.1 and earlier allows remote attackers to cause a denial of service (crash) and possibly execute arbitrary code via a crafted MPEG-4 (MP4) file.

Emacs developers report:

The Emacs command `run-python' launches an interactive Python interpreter. After the Python process starts up, Emacs automatically sends it the line:

import emacs

which normally imports a script named emacs.py which is distributed with Emacs. This script, which is typically located in a write-protected installation directory with other Emacs program files, defines various functions to help the Python process communicate with Emacs.

The vulnerability arises because Python, by default, prepends '' to the module search path, so modules are looked for in the current directory. If the current directory is world-writable, an attacker may insert malicious code by adding a fake Python module named emacs.py into that directory.

Advisory from Moritz Jodeit, November 8th, 2008:

ClamAV contains an off-by-one heap overflow vulnerability in the code responsible for parsing VBA project files. Successful exploitation could allow an attacker to execute arbitrary code with the privileges of the `clamd' process by sending an email with a prepared attachment.

A VBA project file embedded inside an OLE2 office document send as an attachment can trigger the off-by-one.

Entry from Thu Oct 30 13:52:42 CET 2008 (acab) in ChangeLog:

libclamav/vba_extract.c: get_unicode_name off-by-one, bb#1239 reported by Moritz Jodeit >moritz*jodeit.org<

Trac development team reports:

0.11.2 is a new stable maintenance release. It contains several security fixes and everyone is recommended to upgrade their installations.

Bug fixes:

Fixes potential DOS vulnerability with certain wiki markup.

The VLC Team reports:

The VLC media player contains a stack overflow vulnerability while parsing malformed cue files. The vulnerability may be exploited by a (remote) attacker to execute arbitrary code in the context of VLC media player.

Opera reports:

When certain parameters are passed to Opera's History Search, they can cause content not to be correctly sanitized. This can allow scripts to be injected into the History Search results page. Such scripts can then run with elevated privileges and interact with Opera's configuration, allowing them to execute arbitrary code.

The links panel shows links in all frames on the current page, including links with JavaScript URLs. When a page is held in a frame, the script is incorrectly executed on the outermost page, not the page where the URL was located. This can be used to execute scripts in the context of an unrelated frame, which allows cross-site scripting.

Aurelien Jarno reports:

CVE-2008-4539: fix a heap overflow in Cirrus emulation

The code in hw/cirrus_vga.c has changed a lot between CVE-2007-1320 has been announced and the patch has been applied. As a consequence it has wrongly applied and QEMU is still vulnerable to this bug if using VNC.

SecurityFocus reports:

phpMyAdmin is prone to a cross-site scripting vulnerability because it fails to sufficiently sanitize user-supplied data.

An attacker may leverage this issue to execute arbitrary script code in the browser of an unsuspecting user in the context of the affected site. This may allow the attacker to steal cookie-based authentication credentials and to launch other attacks.

Opera reports:

Certain constructs are not escaped correctly by Opera's History Search results. These can be used to inject scripts into the page, which can then be used to look through the user's browsing history, including the contents of the pages they have visited. These may contain sensitive information.

If a link that uses a JavaScript URL triggers Opera's Fast Forward feature, when the user activates Fast Forward, the script should run on the current page. When a page is held in a frame, the script is incorrectly executed on the outermost page, not the page where the URL was located. This can be used to execute scripts in the context of an unrelated frame, which allows cross-site scripting.

When Opera is previewing a news feed, some scripts are not correctly blocked. These scripts are able to subscribe the user to any feed URL that the attacker chooses, and can also view the contents of any feeds that the user is subscribed to. These may contain sensitive information.

CVE reports:

Heap-based buffer overflow in the SPF_dns_resolv_lookup function in Spf_dns_resolv.c in libspf2 before 1.2.8 allows remote attackers to execute arbitrary code via a long DNS TXT record with a modified length field.

Secunia reports:

OpenX can be exploited by malicious people to conduct SQL injection attacks.

Input passed to the "bannerid" parameter in www/delivery/ac.php is not properly sanitised before being used in SQL queries. This can be exploited to manipulate SQL queries by injecting arbitrary SQL code.

The Flyspray Project reports:

Flyspray is affected by a Cross Site scripting Vulnerability due to an error escaping PHP's $_SERVER['QUERY_STRING'] superglobal, that can be maliciously used to inject arbitrary code into the savesearch() javascript function.

There is an XSS problem in the history tab, the application fails to sanitize the "details" parameter correctly, leading to the possibility of arbitrary code injection into the getHistory() javascript function.

Flyspray is affected by a Cross Site scripting Vulnerability due missing escaping of SQL error messages. By including HTML code in a query and at the same time causing it to fail by submitting invalid data, an XSS hole can be exploited.

There is an XSS problem in the task history attached to comments, since the application fails to sanitize the old_value and new_value database fields for changed task summaries.

Input passed via the "item_summary" parameter to "index.php?do=details" is not properly sanitised before being returned to the user. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site.

The Wordpress development team reports:

A vulnerability in the Snoopy library was announced today. WordPress uses Snoopy to fetch the feeds shown in the Dashboard. Although this seems to be a low risk vulnerability for WordPress users, we wanted to get an update out immediately.

The Drupal Project reports:

On a server configured for IP-based virtual hosts, Drupal may be caused to include and execute specifically named files outside of its root directory. This bug affects both Drupal 5 and Drupal 6.

The title of book pages is not always properly escaped, enabling users with the "create book content" permission or the permission to edit any node in the book hierarchy to insert arbitrary HTML and script code into pages. Such a Cross site scripting attack may lead to the attacker gaining administrator access. This bug affects Drupal 6.

xine team reports:

A new xine-lib version is now available. This release contains some security fixes, notably a DoS via corrupted Ogg files (CVE-2008-3231), some related fixes, and fixes for a few possible buffer overflows.

Adobe Product Security Incident Response Team reports:

Potential vulnerabilities have been identified in Adobe Flash Player 9.0.124.0 and earlier that could allow an attacker who successfully exploits these potential vulnerabilities to bypass Flash Player security controls. Adobe recommends users update to the most current version of Flash Player available for their platform.

Secunia reports:

Two vulnerabilities have been reported in Libxml2, which can be exploited by malicious people to cause a DoS (Denial of Service) or potentially compromise an application using the library.

1) A recursion error exists when processing certain XML content. This can be exploited to e.g. exhaust all available memory and CPU resources by tricking an application using Libxml2 into processing specially crafted XML documents.

2) A boundary error in the processing of long XML entity names in parser.c can be exploited to cause a heap-based buffer overflow when specially crafted XML content is parsed.

Successful exploitation may allow execution of arbitrary code.

The Drupal Project reports:

A logic error in the core upload module validation allowed unprivileged users to attach files to content. Users can view files attached to content which they do not otherwise have access to. If the core upload module is not enabled, your site will not be affected.

A deficiency in the user module allowed users who had been blocked by access rules to continue logging into the site under certain conditions. If you do not use the 'access rules' functionality in core, your site will not be affected.

The BlogAPI module does not implement correct validation for certain content fields, allowing for values to be set for fields which would otherwise be inaccessible on an internal Drupal form. We have hardened these checks in BlogAPI module for this release, but the security team would like to re-iterate that the 'Administer content with BlogAPI' permission should only be given to trusted users. If the core BlogAPI module is not enabled, your site will not be affected.

A weakness in the node module API allowed for node validation to be bypassed in certain circumstances for contributed modules implementing the API. Additional checks have been added to ensure that validation is performed in all cases. This vulnerability only affects sites using one of a very small number of contributed modules, all of which will continue to work correctly with the improved API. None of them were found vulnerable, so our correction is a preventative measure.

The release note of cups 1.3.9 reports:

It contains the following fixes:

• SECURITY: The HP-GL/2 filter did not range check pen numbers (STR #2911)
• SECURITY: The SGI image file reader did not range check 16-bit run lengths (STR #2918)
• SECURITY: The text filter did not range check cpi, lpi, or column values (STR #2919)

Exploitation of this vulnerability results in the execution of arbitrary code with the privileges of the affected service.

Opera reports:

If a malicious page redirects Opera to a specially crafted address (URL), it can cause Opera to crash. Given sufficient address content, the crash could cause execution of code controlled by the attacking page.

Once a Java applet has been cached, if a page can predict the cache path for that applet, it can load the applet from the cache, causing it to run in the context of the local machine. This allows it to read other cache files on the computer or perform other normally more restrictive actions. These files could contain sensitive information, which could then be sent to the attacker.

Thomas Henlich reports:

The mysql command-line client does not quote HTML special characters like < in its output. This allows an attacker who is able to write data into a table to hide or modify records in the output, and to inject potentially dangerous code, e. g. Javascript to perform cross-site scripting or cross-site request forgery attacks.

The oCERT team reports:

The MPlayer multimedia player suffers from a vulnerability which could result in arbitrary code execution and at the least, in unexpected process termination. Three integer underflows located in the Real demuxer code can be used to exploit a heap overflow, a specific video file can be crafted in order to make the stream_read function reading or writing arbitrary amounts of memory.

Lighttpd seurity annoucement:

lighttpd 1.4.19, and possibly other versions before 1.5.0, does not decode the url before matching against rewrite and redirect patterns, which allows attackers to bypass rewrites rules. this can be a security problem in certain configurations if these rules are used to hide certain urls.

lighttpd 1.4.19, and possibly other versions before 1.5.0, does not lowercase the filename after generating it from the url in mod_userdir on case insensitive (file)systems.

As other modules are case sensitive, this may lead to information disclosure; for example if one configured php to handle files ending on ".php", an attacker will get the php source with http://example.com/~user/file.PHP

lighttpd 1.4.19 does not always release a header if it triggered a 400 (Bad Request) due to a duplicate header.

Secunia reports:

Some security issues have been reported in BitlBee, which can be exploited by malicious people to bypass certain security restrictions and hijack accounts.

The security issues are caused due to unspecified errors, which can be exploited to overwrite existing accounts.

The Mozilla Foundation reports:

MFSA 2008-37
UTF-8 URL stack buffer overflow

MFSA 2008-38
nsXMLDocument::OnChannelRedirect() same-origin violation

MFSA 2008-39
Privilege escalation using feed preview page and XSS flaw

MFSA 2008-40
Forced mouse drag

MFSA 2008-41
Privilege escalation via XPCnativeWrapper pollution

MFSA 2008-42
Crashes with evidence of memory corruption (rv:1.9.0.2/1.8.1.17)

MFSA 2008-43
BOM characters stripped from JavaScript before execution

MFSA 2008-44
resource: traversal vulnerabilities

MFSA 2008-45
XBM image uninitialized memory reading

Hanno Boeck reports:

When configuring a web application to use only ssl (e.g. by forwarding all http-requests to https), a user would expect that sniffing and hijacking the session is impossible.

Though, for this to be secure, one needs to set the session cookie to have the secure flag. Otherwise the cookie will be transferred through HTTP if the victim's browser does a single HTTP request on the same domain.

Squirrelmail does not set that flag. It is fixed in the 1.5 test versions, but current 1.4.15 is vulnerable.

Secunia reports:

The vulnerability is caused due to the application truncating an overly long FTP command, and improperly interpreting the remainder string as a new FTP command. This can be exploited to execute arbitrary FTP commands with the privileges of another user by e.g. tricking the user into following malicious link.

Secunia reports:

An error exists in the "PMA_escapeJsString()" function in libraries/js_escape.lib.php, which can be exploited to bypass certain filters and execute arbitrary HTML and script code in a user's browser session in context of an affected site when e.g. Microsoft Internet Explorer is used.

Secunia reports:

An error in the handing of ZIP archives with symbolic links can be exploited to disclose the contents of arbitrary files.

Input from uploaded Flash animations is not properly sanitised before being used. This can be exploited to insert arbitrary HTML and script code, which is executed in a user's browser session in context of an affected site when the malicious data is viewed.

A phpMyAdmin security announcement:

The server_databases.php script was vulnerable to an attack coming from a user who is already logged-on to phpMyAdmin, where he can execute shell code (if the PHP configuration permits commands like exec).

Th1nk3r reports:

The version of TWiki installed on the remote host allows access to the 'configure' script and fails to sanitize the 'image' parameter of that script of directory traversal sequences before returning the file contents when the 'action' parameter is set to 'image'. An unauthenticated attacker can leverage this issue to view arbitrary files on the remote host subject to the privileges of the web server user id. .

Joe Orton reports:

A NULL pointer deference in the Digest authentication support in neon versions 0.28.0 through 0.28.2 inclusive allows a malicious server to crash a client application, resulting in possible denial of service.

Hanno Boeck reports:

A fuzzing test showed weakness in the chm parser of clamav, which can possibly be exploited. The clamav team has disabled the chm module in older versions though freshclam updates and has released 0.94 with a fixed parser.

Secunia reports:

Some vulnerabilities have been reported in various Horde products, which can be exploited by malicious people to conduct script insertion attacks

Input via MIME attachment linking is not properly sanitised in the MIME library before being used. This can be exploited to execute arbitrary HTML and script code in a user's browser session if e.g. a malicious email is viewed.

Certain unspecified input in HTML messages is not properly sanitised before being used. This can be exploited to execute arbitrary HTML and script in a user's browser session if e.g. a malicious HTML email is viewed.

Secunia reports:

Some vulnerabilities have been reported in Python, where some have unknown impact and others can potentially be exploited by malicious people to cause a DoS (Denial of Service) or to compromise a vulnerable system.

Various integer overflow errors exist in core modules e.g. stringobject, unicodeobject, bufferobject, longobject, tupleobject, stropmodule, gcmodule, mmapmodule.

An integer overflow in the hashlib module can lead to an unreliable cryptographic digest results.

Integer overflow errors in the processing of unicode strings can be exploited to cause buffer overflows on 32-bit systems.

An integer overflow exists in the PyOS_vsnprintf() function on architectures that do not have a "vsnprintf()" function.

An integer underflow error in the PyOS_vsnprintf() function when passing zero-length strings can lead to memory corruption.

SecurityFocus reports:

MySQL is prone to a security-bypass vulnerability. An attacker can exploit this issue to overwrite existing table files in the MySQL data directory, bypassing certain security restrictions.

Jonathan Weiss reports, that it is possible to perform an SQL injection in Rails applications via not correctly sanitized :limit and :offset parameters. It is possible to change arbitrary values in affected tables or gain access to the sensitive data.

The Wordpress development team reports:

With open registration enabled, it is possible in WordPress versions 2.6.1 and earlier to craft a username such that it will allow resetting another users password to a randomly generated password. The randomly generated password is not disclosed to the attacker, so this problem by itself is annoying but not a security exploit. However, this attack coupled with a weakness in the random number seeding in mt_rand() could be used to predict the randomly generated password.

Problem Description:

In case of an incoming ICMPv6 'Packet Too Big Message', there is an insufficient check on the proposed new MTU for a path to the destination.

Impact:

When the kernel is configured to process IPv6 packets and has active IPv6 TCP sockets, a specifically crafted ICMPv6 'Packet Too Big Message' could cause the TCP stack of the kernel to panic.

Workaround:

Systems without INET6 / IPv6 support are not vulnerable and neither are systems which do not listen on any IPv6 TCP sockets and have no active IPv6 connections.

Filter ICMPv6 'Packet Too Big Messages' using a firewall, but this will at the same time break PMTU support for IPv6 connections.

Problem Description:

Various user defined input such as mount points, devices, and mount options are prepared and passed as arguments to nmount(2) into the kernel. Under certain error conditions, user defined data will be copied into a stack allocated buffer stored in the kernel without sufficient bounds checking.

Impact:

If the system is configured to allow unprivileged users to mount file systems, it is possible for a local adversary to exploit this vulnerability and execute code in the context of the kernel.

Workaround:

It is possible to work around this issue by allowing only privileged users to mount file systems by running the following sysctl(8) command:

# sysctl vfs.usermount=0

Problem Description:

If a General Protection Fault happens on a FreeBSD/amd64 system while it is returning from an interrupt, trap or system call, the swapgs CPU instruction may be called one extra time when it should not resulting in userland and kernel state being mixed.

Impact:

A local attacker can by causing a General Protection Fault while the kernel is returning from an interrupt, trap or system call while manipulating stack frames and, run arbitrary code with kernel privileges.

The vulnerability can be used to gain kernel / supervisor privilege. This can for example be used by normal users to gain root privileges, to break out of jails, or bypass Mandatory Access Control (MAC) restrictions.

Workaround:

No workaround is available, but only systems running the 64 bit FreeBSD/amd64 kernels are vulnerable.

Systems with 64 bit capable CPUs, but running the 32 bit FreeBSD/i386 kernel are not vulnerable.

The Opera Team reports:

Scripts are able to change the addresses of framed pages that come from the same site. Due to a flaw in the way that Opera checks what frames can be changed, a site can change the address of frames on other sites inside any window that it has opened. This allows sites to open pages from other sites, and display misleading information on them.

Custom shortcut and menu commands can be used to activate external applications. In some cases, the parameters passed to these applications are not prepared correctly, and may be created from uninitialized memory. These may be misinterpreted as additional parameters, and depending on the application, this could allow execution of arbitrary code.

Successful exploitation requires convincing the user to modify their shortcuts or menu files appropriately, pointing to an appropriate target application, then to activate that shortcut at an appropriate time. To inject code, additional means will have to be employed.

When insecure pages load content from secure sites into a frame, they can cause Opera to incorrectly report the insecure site as being secure. The padlock icon will incorrectly be shown, and the security information dialog will state that the connection is secure, but without any certificate information.

As a security precaution, Opera does not allow Web pages to link to files on the user's local disk. However, a flaw exists that allows Web pages to link to feed source files on the user's computer. Suitable detection of JavaScript events and appropriate manipulation can unreliably allow a script to detect the difference between successful and unsuccessful subscriptions to these files, to allow it to discover if the file exists or not. In most cases the attempt will fail.

It has been reported that when a user subscribes to a news feed using the feed subscription button, the page address can be changed. This causes the address field not to update correctly. Although this can mean that misleading information can be displayed in the address field, it can only leave the attacking page's address in the address bar, not a trusted third party address.

Secunia reports:

A vulnerability has been reported in GnuTLS, which can potentially be exploited by malicious people to cause a DoS (Denial of Service).

The vulnerability is caused due to a use-after-free error when an application calls "gnutls_handshake()" for an already valid session and can potentially be exploited, e.g. during re-handshakes.

Joomla project reports:

A flaw in the reset token validation mechanism allows for non-validating tokens to be forged. This will allow an unauthenticated, unauthorized user to reset the password of the first enabled user (lowest id). Typically, this is an administrator user. Note, that changing the first users username may lessen the impact of this exploit (since the person who changed the password does not know the login associated with the new password). However, the only way to completely rectify the issue is to upgrade to 1.5.6 (or patch the /components/com_user/models/reset.php file).

NASA Goddard Space Flight Center reports:

The libraries for the scientific data file format, Common Data Format (CDF) version 3.2 and earlier, have the potential for a buffer overflow vulnerability when reading specially-crafted (invalid) CDF files. If successful, this could trigger execution of arbitrary code within the context of the CDF-reading program that could be exploited to compromise a system, or otherwise crash the program. While it's unlikely that you would open CDFs from untrusted sources, we recommend everyone upgrade to the latest CDF libraries on their systems, including the IDL and Matlab plugins. Most worrisome is any service that enables the general public to submit CDF files for processing.

The vulnerability is in the CDF library routines not properly checking the length tags on a CDF file before copying data to a stack buffer. Exploitation requires the user to explicitly open a specially-crafted file. CDF users should not open files from untrusted third parties until the patch is applied (and continue then to exercise normal caution for files from untrusted third parties).

The Drupal Project reports:

A bug in the output filter employed by Drupal makes it possible for malicious users to insert script code into pages (cross site scripting or XSS). A bug in the private filesystem trusts the MIME type sent by the browser, enabling malicious users with the ability to upload files to execute cross site scripting attacks.

The BlogAPI module does not validate the extension of uploaded files, enabling users with the "administer content with blog api" permission to upload harmful files. This bug affects both Drupal 5.x and 6.x.

Drupal forms contain a token to protect against cross site request forgeries (CSRF). The token may not be validated properly for cached forms and forms containing AHAH elements. This bug affects Drupal 6.x.

User access rules can be added or deleted upon accessing a properly formatted URL, making such modifications vulnerable to cross site request forgeries (CSRF). This may lead to unintended addition or deletion of an access rule when a sufficiently privileged user visits a page or site created by a malicious person. This bug affects both Drupal 5.x and 6.x.

The Upload module in Drupal 6 contains privilege escalation vulnerabilities for users with the "upload files" permission. This can lead to users being able to edit nodes which they are normally not allowed to, delete any file to which the webserver has sufficient rights, and download attachments of nodes to which they have no access. Harmful files may also be uploaded via cross site request forgeries (CSRF). These bugs affect Drupal 6.x.

The official ruby site reports:

Several vulnerabilities in safe level have been discovereds:.

• untrace_var is permitted at safe level 4;
• $PROGRAM_NAME may be modified at safe level 4;
• insecure methods may be called at safe level 1-3;
• syslog operations are permitted at safe level 4;
• dl doesn't check taintness, so it could allow attackers to call dangerous functions.

The official ruby site reports:

WEBrick::HTTP::DefaultFileHandler is faulty of exponential time taking requests due to a backtracking regular expression in WEBrick::HTTPUtils.split_header_value.

The official ruby site reports:

resolv.rb allow remote attackers to spoof DNS answers. This risk can be reduced by randomness of DNS transaction IDs and source ports.

A Bugzilla Security Advisory reports:

When importing bugs using importxml.pl, the --attach_path option can be specified, pointing to the directory where attachments to import are stored. If the XML file being read by importxml.pl contains a malicious

../relative_path/to/local_file

node, the script follows this relative path and attaches the local file pointed by it to the bug, making the file public. The security fix makes sure the relative path is always ignored.

James Yonan reports:

Security Fix - affects non-Windows OpenVPN clients running OpenVPN 2.1-beta14 through 2.1-rc8 (OpenVPN 2.0.x clients are NOT vulnerable nor are any versions of the OpenVPN server vulnerable).

An OpenVPN client connecting to a malicious or compromised server could potentially receive an "lladdr" or "iproute" configuration directive from the server which could cause arbitrary code execution on the client. A successful attack requires that (a) the client has agreed to allow the server to push configuration directives to it by including "pull" or the macro "client" in its configuration file, (b) the client succesfully authenticates the server, (c) the server is malicious or has been compromised and is under the control of the attacker, and (d) the client is running a non-Windows OS.

A phpMyAdmin security announcement:

A logged-in user, if abused into clicking a crafted link or loading an attack page, would create a database he did not intend to, or would change his connection character set.

The Drupal Project reports:

Free tagging taxonomy terms can be used to insert arbitrary script and HTML code (cross site scripting or XSS) on node preview pages. A successful exploit requires that the victim selects a term containing script code and chooses to preview the node. This issue affects Drupal 6.x only. Some values from OpenID providers are output without being properly escaped, allowing malicious providers to insert arbitrary script and HTML code (XSS) into user pages. This issue affects Drupal 6.x only. filter_xss_admin() has been hardened to prevent use of the object HTML tag in administrator input.

Translated strings (5.x, 6.x) and OpenID identities (6.x) are immediately deleted upon accessing a properly formatted URL, making such deletion vulnerable to cross site request forgeries (CSRF). This may lead to unintended deletion of translated strings or OpenID identities when a sufficiently privileged user visits a page or site created by a malicious person.

When contributed modules such as Workflow NG terminate the current request during a login event, user module is not able to regenerate the user's session. This may lead to a session fixation attack, when a malicious user is able to control another users' initial session ID. As the session is not regenerated, the malicious user may use the 'fixed' session ID after the victim authenticates and will have the same access. This issue affects both Drupal 5 and Drupal 6.

Schema API uses an inappropriate placeholder for 'numeric' fields enabling SQL injection when user-supplied data is used for such fields.This issue affects Drupal 6 only.

Problem Description:

The BIND DNS implementation does not randomize the UDP source port when doing remote queries, and the query id alone does not provide adequate randomization.

Impact:

The lack of source port randomization reduces the amount of data the attacker needs to guess in order to successfully execute a DNS cache poisoning attack. This allows the attacker to influence or control the results of DNS queries being returned to users from target systems.

Workaround:

Limiting the group of machines that can do recursive queries on the DNS server will make it more difficult, but not impossible, for this vulnerability to be exploited.

To limit the machines able to perform recursive queries, add an ACL in named.conf and limit recursion like the following:

acl example-acl {

   192.0.2.0/24;

};

options {

	recursion yes;

	allow-recursion { example-acl; };

};

Felipe Andres Manzano reports:

The libpoppler pdf rendering library, can free uninitialized pointers, leading to arbitrary code execution. This vulnerability results from memory management bugs in the Page class constructor/destructor.

Pylons team reports:

The error.py controller uses paste.fileapp to serve the static resources to the browser. The default error.py controller uses os.path.join to combine the id from Routes with the media path. Routes prior to 1.8 double unquoted the PATH_INFO, resulting in FileApp returning files from the filesystem that can be outside of the intended media path directory.

An attacker can craft URL's which utilize the double escaping to pass in a name to the error.py controller which contains a leading slash thus escaping the intended media path and serving files from any location on the filesystem that the Pylons application has access to.

Secunia reports:

• An integer overflow error exists in the processing of PFB font files. This can be exploited to cause a heap-based buffer overflow via a PFB file containing a specially crafted "Private" dictionary table.
• An error in the processing of PFB font files can be exploited to trigger the "free()" of memory areas that are not allocated on the heap.
• An off-by-one error exists in the processing of PFB font files. This can be exploited to cause a one-byte heap-based buffer overflow via a specially crafted PFB file.
• An off-by-one error exists in the implementation of the "SHC" instruction while processing TTF files. This can be exploited to cause a one-byte heap-based buffer overflow via a specially crafted TTF file.

Successful exploitation of the vulnerabilities may allow execution of arbitrary code.

Matthias Andree reports:

2008-06-24 1.2 also fixed issue in report_complete (reported by Petr Uzel)

Secunia report:

Some vulnerabilities have been reported in phpMyAdmin, which can be exploited by malicious people to conduct cross-site scripting attacks.

Input passed via unspecified parameters to files in /libraries is not properly sanitised before being returned to the user. This can be exploited to execute arbitrary HTML and script code in a user's browser session in context of an affected site.

Successful exploitation requires that "register_globals" is enabled and support for ".htaccess" files is disabled.

Apache HTTP server project reports:

The following potential security flaws are addressed:

• CVE-2008-2364: mod_proxy_http: Better handling of excessive interim responses from origin server to prevent potential denial of service and high memory usage. Reported by Ryujiro Shibuya.
• CVE-2007-6420: mod_proxy_balancer: Prevent CSRF attacks against the balancer-manager interface

According to Maksymilian Arciemowicz research, it is possible to bypass security restrictions of safe_mode in various functions via directory traversal vulnerability. The attacker can use this attack to gain access to sensitive information. Functions utilizing expand_filepath() may be affected.

It should be noted that this vulnerability is not considered to be serious by the FreeBSD Security Team, since safe_mode and open_basedir are insecure by design and should not be relied upon.

Rdancer.org reports:

Improper quoting in some parts of Vim written in the Vim Script can lead to arbitrary code execution upon opening a crafted file.

The official ruby site reports:

Multiple vulnerabilities in Ruby may lead to a denial of service (DoS) condition or allow execution of arbitrary code.

Matthias Andree reports:

Gunter Nau reported fetchmail crashing on some messages; further debugging by Petr Uzel and Petr Cerny at Novell/SUSE Czech Republic dug up that this happened when fetchmail was trying to print, in -v -v verbose level, headers exceeding 2048 bytes. In this situation, fetchmail would resize the buffer and fill in further parts of the message, but forget to reinitialize its va_list typed source pointer, thus reading data from a garbage address found on the stack at addresses above the function arguments the caller passed in; usually that would be the caller's stack frame.

Matthieu Herrb of X.Org reports:

Several vulnerabilities have been found in the server-side code of some extensions in the X Window System. Improper validation of client-provided data can cause data corruption.

Exploiting these overflows will crash the X server or, under certain circumstances allow the execution of arbitray machine code.

When the X server is running with root privileges (which is the case for the Xorg server and for most kdrive based servers), these vulnerabilities can thus also be used to raise privileges.

All these vulnerabilities, to be exploited successfully, require either an already established connection to a running X server (and normally running X servers are only accepting authenticated connections), or a shell access with a valid user on the machine where the vulnerable server is installed.

MoinMoin team reports:

A check in the userform processing was not working as expected and could be abused for ACL and superuser privilege escalation.

Secunia reports:

A vulnerability has been reported in the Courier Authentication Library, which can be exploited by malicious people to conduct SQL injection attacks.

Input passed via e.g. the username to the library is not properly sanitised before being used in SQL queries. This can be exploited to manipulate SQL queries by injecting arbitrary SQL code and e.g. potentially bypass authentication.

Successful exploitation requires that a MySQL database is used for authentication and that a Non-Latin character set is selected.

The ikiwiki development team reports:

Until version 2.48, ikiwiki stored passwords in cleartext in the userdb. That risks exposing all users' passwords if the file is somehow exposed. To pre-emtively guard against that, current versions of ikiwiki store password hashes (using Eksblowfish).

The ikiwiki development team reports:

This hole allowed ikiwiki to accept logins using empty passwords to openid accounts that didn't use a password.

Upgrading to a non-vulnerable ikiwiki version immediatly is recommended if your wiki allows both password and openid logins.

Adobe Product Security Incident Response Team reports:

An exploit appears to be taking advantage of a known vulnerability, reported by Mark Dowd of the ISS X-Force and wushi of team509, that was resolved in Flash Player 9.0.124.0 (CVE-2007-0071). This exploit does NOT appear to include a new, unpatched vulnerability as has been reported elsewhere - customers with Flash Player 9.0.124.0 should not be vulnerable to this exploit.

Secunia reports:

A vulnerability has been reported in Nagios, which can be exploited by malicious people to conduct cross-site scripting attacks.

Spamdyke Team reports:

Fixed smtp_filter() to reject the DATA command if no valid recipients have been specified. Otherwise, a specific scenario could result in every spamdyke installation being used as an open relay. If the remote server connects and gives one or more recipients that are rejected (for relaying or blacklisting), then gives the DATA command, spamdyke will ignore all other commands, assuming that message data is being transmitted. However, because all of the recipients were rejected, qmail will reject the DATA command. From that point on, the remote server can give as many recipients as it likes and spamdyke will ignore them all -- they will not be filtered at all. After that, the remote server can give the DATA command and send the actual message data. Because spamdyke is controlling relaying, the RELAYCLIENT environment variable is set and qmail won't check for relaying either. Thanks to Mirko Buffoni for reporting this one.

Nico Golde discovered that PeerCast, a P2P audio and video streaming server, is vulnerable to a buffer overflow in the HTTP Basic Authentication code, allowing a remote attacker to crash PeerCast or execure arbitrary code.

Red Hat reports:

Will Drewry of the Google Security Team reported several flaws in the way libvorbis processed audio data. An attacker could create a carefully crafted [Vorbis] audio file in such a way that it could cause an application linked with libvorbis to crash, or execute arbitrary code when it was opened.

Django project reports:

The Django administration application will, when accessed by a user who is not sufficiently authenticated, display a login form and ask the user to provide the necessary credentials before displaying the requested page. This form will be submitted to the URL the user attempted to access, by supplying the current request path as the value of the form's "action" attribute.

The value of the request path was not being escaped, creating an opportunity for a cross-site scripting (XSS) attack by leading a user to a URL which contained URL-encoded HTML and/or JavaScript in the request path.

Secunia reports:

A vulnerability has been reported in vorbis-tools, which can potentially be exploited by malicious people to compromise a user's system.

The vulnerability is caused due to an input validation error when processing Speex headers, which can be exploited via a specially crafted Speex stream containing a negative "modeID" field in the header.

Successful exploitation may allow execution of arbitrary code.

Secunia reports:

A vulnerability has been reported in QEMU, which can be exploited by malicious, local users to bypass certain security restrictions.

The vulnerability is caused due to the "drive_init()" function in vl.c determining the format of a disk from data contained in the disk's header. This can be exploited by a malicious user in a guest system to e.g. read arbitrary files on the host by writing a fake header to a raw formatted disk image.

Secunia reports:

A vulnerability has been reported in swfdec, which can be exploited by malicious people to disclose sensitive information.

The vulnerability is caused due to swfdec not properly restricting untrusted sandboxes from reading local files, which can be exploited to disclose the content of arbitrary local files by e.g. tricking a user into visiting a malicious website.

FrSIRT reports:

A vulnerability has been identified in mt-daapd which could be exploited by remote attackers to cause a denial of service or compromise an affected system. This issue is caused by a buffer overflow error in the ws_getpostvars() function when processing a negative Content-Length: header value, which could be exploited by remote unauthenticated attackers to crash an affected application or execute arbitrary code.

Secunia reports:

Two vulnerabilities have been reported in SDL_image, which can be exploited by malicious people to cause a Denial of Service or potentially compromise an application using the library.

A boundary error within the LWZReadByte() function in IMG_gif.c can be exploited to trigger the overflow of a static buffer via a specially crafted GIF file.

A boundary error within the "IMG_LoadLBM_RW()" function in IMG_lbm.c can be exploited to cause a heap-based buffer overflow via a specially crafted IFF ILBM file.

Secunia reports:

A vulnerability has been reported in GnuPG, which can potentially be exploited to compromise a vulnerable system.

The vulnerability is caused due to an error when importing keys with duplicated IDs. This can be exploited to cause a memory corruption when importing keys via --refresh-keys or --import.

Successful exploitation potentially allows execution of arbitrary code, but has not been proven yet.

Extmail team reports:

Emergency update #4 fixes a serious security vulnerability.

Successful exploit of this vulnerability would allow attacker to change user's password without knowing it by using specifically crafted HTTP request.

Secunia reports:

A vulnerability has been reported in Mailman, which can be exploited by malicious users to conduct script insertion attacks.

Certain input when editing the list templates and the list info attribute is not properly sanitised before being stored. This can be exploited to insert arbitrary HTML and script code, which is executed in a user's browser session in context of an affected site when the malicious website is accessed.

Secunia reports:

The vulnerability is caused due to an error when attaching to a TTY via the -T command line switch. This can be exploited to execute arbitrary commands with the privileges of the user running mksh via characters previously written to the attached virtual console.

Hanno Boeck reports:

The installer of serendipity 1.3 has various Cross Site Scripting issues. This is considered low priority, as attack scenarios are very unlikely.

Various path fields are not escaped properly, thus filling them with javascript code will lead to XSS. MySQL error messages are not escaped, thus the database host field can also be filled with javascript.

In the referrer plugin of the blog application serendipity, the referrer string is not escaped, thus leading to a permanent XSS.

Mozilla Foundation reports:

Fixes for security problems in the JavaScript engine described in MFSA 2008-15 introduced a stability problem, where some users experienced crashes during JavaScript garbage collection. This is being fixed primarily to address stability concerns. We have no demonstration that this particular crash is exploitable but are issuing this advisory because some crashes of this type have been shown to be exploitable in the past.

Secunia reports:

Tavis Ormandy has reported a vulnerability in libpng, which can be exploited by malicious people to cause a Denial of Service, disclose potentially sensitive information, or potentially compromise an application using the library.

The vulnerability is caused due to the improper handling of PNG chunks unknown to the library. This can be exploited to trigger the use of uninitialized memory in e.g. a free() call via unknown PNG chunks having a length of zero.

Successful exploitation may allow execution of arbitrary code, but requires that the application calls the png_set_read_user_chunk_fn() function or the png_set_keep_unknown_chunks() function under specific conditions.

Secunia reports:

A vulnerability has been reported in Openfire, which can be exploited by malicious people to cause a Denial of Service.

The vulnerability is caused due to an unspecified error and can be exploited to cause a Denial of Service.

CVE reports:

Integer overflow in PHP 5.2.5 and earlier allows context-dependent attackers to cause a denial of service and possibly have unspecified other impact via a printf format parameter with a large width specifier, related to the php_sprintf_appendstring function in formatted_print.c and probably other functions for formatted strings (aka *printf functions).

Justin Ferguson reports:

Integer signedness error in the zlib extension module in Python 2.5.2 and earlier allows remote attackers to execute arbitrary code via a negative signed integer, which triggers insufficient memory allocation and a buffer overflow.

The PostgreSQL developers report:

PostgreSQL allows users to create indexes on the results of user-defined functions, known as "expression indexes". This provided two vulnerabilities to privilege escalation: (1) index functions were executed as the superuser and not the table owner during VACUUM and ANALYZE, and (2) that SET ROLE and SET SESSION AUTHORIZATION were permitted within index functions. Both of these holes have now been closed.

PostgreSQL allowed malicious users to initiate a denial-of-service by passing certain regular expressions in SQL queries. First, users could create infinite loops using some specific regular expressions. Second, certain complex regular expressions could consume excessive amounts of memory. Third, out-of-range backref numbers could be used to crash the backend.

DBLink functions combined with local trust or ident authentication could be used by a malicious user to gain superuser privileges. This issue has been fixed, and does not affect users who have not installed DBLink (an optional module), or who are using password authentication for local access. This same problem was addressed in the previous release cycle, but that patch failed to close all forms of the loophole.

A phpMyAdmin security announcement report:

It is possible to read the contents of any file that the web server's user can access. The exact mechanism to achieve this won't be disclosed. If a user can upload on the same host where phpMyAdmin is running a PHP script that can read files with the rights of the web server's user, the current advisory does not describe an additional threat.

A phpMyAdmin security announcement report:

phpMyAdmin saves sensitive information like the MySQL username and password and the Blowfish secret key in session data, which might be unprotected on a shared host.

xine Team reports:

A new xine-lib version is now available. This release contains a security fix (an unchecked array index that could allows remote attackers to execute arbitrary code via a header structure containing a negative offset, which is used to dereference a function pointer.)

Secunia reports:

Some vulnerabilities have been reported in ClamAV, which can be exploited by malicious people to cause a DoS (Denial of Service) or to compromise a vulnerable system.

1) A boundary error exists within the "cli_scanpe()" function in libclamav/pe.c. This can be exploited to cause a heap-based buffer overflow via a specially crafted "Upack" executable.

Successful exploitation allows execution of arbitrary code.

2) A boundary error within the processing of PeSpin packed executables in libclamav/spin.c can be exploited to cause a heap-based buffer overflow.

Successful exploitation may allow execution of arbitrary code.

3) An unspecified error in the processing of ARJ files can be exploited to hang ClamAV.

Secunia reports:

A vulnerability has been reported in lighttpd, which can be exploited by malicious people to cause a DoS (Denial of Service).

The vulnerability is caused due to lighttpd not properly clearing the OpenSSL error queue. This can be exploited to close concurrent SSL connections of lighttpd by terminating one SSL connection.

The ikiwiki development team reports:

Cross Site Request Forging could be used to construct a link that would change a logged-in user's password or other preferences if they clicked on the link. It could also be used to construct a link that would cause a wiki page to be modified by a logged-in user.

postfix-policyd-weight does not check for symlink for its working directory. If the working directory is not already setup by the super root, an unprivileged user can link it to another directories in the system. This results in ownership/permission changes on the target directory.

If the system random number generator can be predicted by its past output, then an attacker may spoof Recursor to accept mallicious data. This leads to DNS cache poisoning and client redirection.

Multiple local privilege escalation are found in the symlink verification code. An attacker may use it to run a PHP script with the victim's privilege. This attack is a little harder when suphp operates in paranoid mode. For suphp that runs in owner mode which is the default in ports, immediate upgrade to latest version is advised.

Opera Software reports of multiple security issues in Opera. All of them can lead to arbitrary code execution. Details are as the following:

• Newsfeed prompt can cause Opera to execute arbitrary code
• Resized canvas patterns can cause Opera to execute arbitrary code

The Mozilla Foundation reports of multiple security issues in Firefox, Seamonkey, and Thunderbird. Several of these issues can probably be used to run arbitrary code with the privilege of the user running the program.

• MFSA 2008-19 XUL popup spoofing variant (cross-tab popups)
• MFSA 2008-18 Java socket connection to any local port via LiveConnect
• MFSA 2008-17 Privacy issue with SSL Client Authentication
• MFSA 2008-16 HTTP Referrer spoofing with malformed URLs
• MFSA 2008-15 Crashes with evidence of memory corruption (rv:1.8.1.13)
• MFSA 2008-14 JavaScript privilege escalation and arbitrary code execution

Core Security Technologies reports:

A remote buffer overflow vulnerability found in a library used by both the SILC server and client to process packets containing cryptographic material may allow an un-authenticated client to executearbitrary code on the server with the privileges of the user account running the server, or a malicious SILC server to compromise client systems and execute arbitrary code with the privileges of the user account running the SILC client program.

SecurityFocus reports:

The 'bzip2' application is prone to a remote file-handling vulnerability because the application fails to properly handle malformed files.

Exploit attempts likely result in application crashes.

Ian Jackson reports on the debian-security mailinglist:

When a block device read or write request is made by the guest, nothing checks that the request is within the range supported by the backend, but the code in the backend typically assumes that the request is sensible.

Depending on the backend, this can allow the guest to read and write arbitrary memory locations in qemu, and possibly gain control over the qemu process, escaping from the emulation/virtualisation.

Dovecot reports:

Security hole in blocking passdbs (MySQL always. PAM, passwd and shadow if blocking=yes) where user could specify extra fields in the password. The main problem here is when specifying "skip_password_check" introduced in v1.0.11 for fixing master user logins, allowing the user to log in as anyone without a valid password.

The Mplayer team reports:

A buffer overflow was found in the code used to extract album titles from CDDB server answers. When parsing answers from the CDDB server, the album title is copied into a fixed-size buffer with insufficient size checks, which may cause a buffer overflow. A malicious database entry could trigger a buffer overflow in the program. That can lead to arbitrary code execution with the UID of the user running MPlayer.

A buffer overflow was found in the code used to escape URL strings. The code used to skip over IPv6 addresses can be tricked into leaving a pointer to a temporary buffer with a non-NULL value; this causes the unescape code to reuse the buffer, and may lead to a buffer overflow if the old buffer is smaller than required. A malicious URL string may be used to trigger a buffer overflow in the program, that can lead to arbitrary code execution with the UID of the user running MPlayer.

A buffer overflow was found in the code used to parse MOV file headers. The code read some values from the file and used them as indexes into as array allocated on the heap without performing any boundary check. A malicious file may be used to trigger a buffer overflow in the program. That can lead to arbitrary code execution with the UID of the user running MPlayer.

Chris Evans from the Google Security Team reports:

Severity: parsing of evil PostScript file will result in arbitrary code execution.

A stack-based buffer overflow in the zseticcspace() function in zicc.c allows remote arbitrary code execution via a malicious PostScript file (.ps) that contains a long Range array.

A phpMyAdmin security announcement report:

phpMyAdmin used the $_REQUEST superglobal as a source for its parameters, instead of $_GET and $_POST. This means that on most servers, a cookie with the same name as one of phpMyAdmin's parameters can interfere.

Another application could set a cookie for the root path "/" with a "sql_query" name, therefore overriding the user-submitted sql_query because by default, the $_REQUEST superglobal imports first GET, then POST then COOKIE data.

Mitigation factor

An attacker must trick the victim into visiting a page on the same web server where he has placed code that creates a malicious cookie.

PCRE developers report:

A character class containing a very large number of characters with codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer overflow.

xine Team reports:

A new xine-lib version is now available. This release contains a security fix (array index vulnerability which may lead to a stack buffer overflow.

Coppermine Security advisory

The development team is releasing a security update for Coppermine in order to counter a recently discovered cross-site-scripting vulnerability.

MoinMoin Security advisory

XSS issue in login action

XSS issue in AttachFile action

XSS issue in RenamePage/DeletePage action

XSS issue in gui editor

Opera Software ASA reports about multiple security fixes:

• Fixed an issue where simulated text inputs could trick users into uploading arbitrary files, as reported by Mozilla.
• Image properties can no longer be used to execute scripts, as reported by Max Leonov.
• Fixed an issue where the representation of DOM attribute values could allow cross site scripting, as reported by Arnaud.lb.

The Mozilla Foundation reports of multiple security issues in Firefox, Seamonkey, and Thunderbird. Several of these issues can probably be used to run arbitrary code with the privilege of the user running the program.

• Web forgery overwrite with div overlay
• URL token stealing via stylesheet redirect
• Mishandling of locally-saved plain text files
• File action dialog tampering
• Possible information disclosure in BMP decoder
• Web browsing history and forward navigation stealing
• Directory traversal via chrome: URI
• Stored password corruption
• Privilege escalation, XSS, Remote Code Execution
• Multiple file input focus stealing vulnerabilities
• Crashes with evidence of memory corruption (rv:1.8.1.12)

Secunia Advisory reports:

A vulnerability has been reported in OpenLDAP, which can be exploited by malicious users to cause a DoS (Denial of Service).

iDefense Security Advisory 02.12.08:

Remote exploitation of an integer overflow vulnerability in Clam AntiVirus' ClamAV, as included in various vendors' operating system distributions, allows attackers to execute arbitrary code with the privileges of the affected process.

The vulnerability exists within the code responsible for parsing and scanning PE files. While iterating through all sections contained in the PE file, several attacker controlled values are extracted from the file. On each iteration, arithmetic operations are performed without taking into consideration 32-bit integer wrap.

Since insufficient integer overflow checks are present, an attacker can cause a heap overflow by causing a specially crafted Petite packed PE binary to be scanned. This results in an exploitable memory corruption condition.

Exploitation of this vulnerability results in the execution of arbitrary code with the privileges of the process using libclamav. In the case of the clamd program, this will result in code execution with the privileges of the clamav user. Unsuccessful exploitation results in the clamd process crashing.

Workaround

Disabling the scanning of PE files will prevent exploitation.

If using clamscan, this can be done by running clamscan with the '--no-pe' option.

If using clamdscan, set the 'ScanPE' option in the clamd.conf file to 'no'.

The cacti development team reports:

Multiple security vulnerabilities have been discovered in Cacti's web interface:

• XSS vulnerabilities
• Path disclosure vulnerabilities
• SQL injection vulnerabilities
• HTTP response splitting vulnerabilities

The ikiwiki development team reports:

The htmlscrubber did not block javascript in uris. This was fixed by adding a whitelist of valid uri types, which does not include javascript. Some urls specifyable by the meta plugin could also theoretically have been used to inject javascript; this was also blocked.

zenphoto project reports:

A new zenphoto version is now available. This release contains security fixes for HTML, XSS, and SQL injection vulnerabilities.

Greg Wilkins reports:

jetty allows remote attackers to bypass protection mechanisms and read the source of files via multiple '/' characters in the URI.

xine project reports:

A new xine-lib version is now available. This release contains a security fix (remotely-expoitable buffer overflow, CVE-2006-1664). (This is not the first time that that bug has been fixed...) It also fixes a few more recent bugs, such as the audio output problems in 1.1.9.

Matthieu Herrb of X.Org reports:

Several vulnerabilities have been identified in server code of the X window system caused by lack of proper input validation on user controlled data in various parts of the software, causing various kinds of overflows.

Exploiting these overflows will crash the X server or, under certain circumstances allow the execution of arbitray machine code.

When the X server is running with root privileges (which is the case for the Xorg server and for most kdrive based servers), these vulnerabilities can thus also be used to raise privileges.

All these vulnerabilities, to be exploited succesfully, require either an already established connection to a running X server (and normally running X servers are only accepting authenticated connections), or a shell access with a valid user on the machine where the vulnerable server is installed.

Gentoo reports:

A remote attacker could entice a user to install a specially crafted "rc" file to execute arbitrary code via long strings in the "Name" and "Comment" fields or via unspecified vectors involving the second vulnerability.

Nico Golde reports:

A local attacker could exploit this vulnerability to conduct symlink attacks to overwrite files with the privileges of the user running Claws Mail.

Secunia reports:

A vulnerability has been reported in IRC Services, which can be exploited by malicious people to cause a Denial of Service. The vulnerability is caused due to the improper handling of overly long passwords within the "default_encrypt()" function in encrypt.c and can be exploited to crash an affected server.

xine project reports:

A new xine-lib version is now available. This release contains a security fix (remotely-expoitable buffer overflow, CVE-2008-0225). It also contains a read-past-end fix for an internal library function which is only used if the OS does not supply it and a rendering fix for Darwin/PPC.

Geeklog reports:

MustLive pointed out a possible XSS in the form to email an article to a friend that we're fixing with this release.

Please note that this problem only exists in Geeklog 1.4.0 - neither Geeklog 1.4.1 nor any older versions (1.3.x series) have that problem.

The Drupal Project reports:

The aggregator module fetches items from RSS feeds and makes them available on the site. The module provides an option to remove items from a particular feed. This has been implemented as a simple GET request and is therefore vulnerable to cross site request forgeries. For example: Should a privileged user view a page containing an <img> tag with a specially constructed src pointing to a remove items URL, the items would be removed.

The Drupal Project reports:

When outputting plaintext Drupal strips potentially dangerous HTML tags and attributes from HTML, and escapes characters which have a special meaning in HTML. This output filtering secures the site against cross site scripting attacks via user input.

Certain byte sequences that are invalid in the UTF8 specification are not handled properly by Internet Explorer 6 and may lead it to see a multibyte start character where none is present. Internet Explorer 6 then consumes a number of subsequent UTF-8 characters. This may lead to unsafe attributes that were outside a tag for the filter to appear inside a tag for Internet Explorer 6. This behaviour can then be used to insert and execute javascript in the context of the website.

The Drupal Project reports:

When theme .tpl.php files are accessible via the web and the PHP setting register_globals is set to enabled, anonymous users are able to execute cross site scripting attacks via specially crafted links.

Drupal's .htaccess attempts to set register_globals to disabled and also prevents access to .tpl.php files. Only when both these measures are not effective and your PHP interpreter is configured with register_globals set to enabled, will this issue affect you.

Secunia reports:

A vulnerability has been reported in MaraDNS, which can be exploited by malicious people to cause a Denial of Service.

The vulnerability is caused due to an error within the handling of certain DNS packets. This can be exploited to cause a resource rotation by sending specially crafted DNS packets, which cause an authoritative CNAME record to not resolve, resulting in a Denial of Sevices.

Secunia reports:

Multiple vulnerabilities have been reported in RealPlayer/RealOne/HelixPlayer, which can be exploited by malicious people to compromise a user's system.

An input validation error when processing .RA/.RAM files can be exploited to cause a heap corruption via a specially crafted .RA/.RAM file with an overly large size field in the header.

An error in the processing of .PLS files can be exploited to cause a memory corruption and execute arbitrary code via a specially crafted .PLS file.

An input validation error when parsing .SWF files can be exploited to cause a buffer overflow via a specially crafted .SWF file with malformed record headers.

A boundary error when processing rm files can be exploited to cause a buffer overflow.

Adobe Security bulletin:

Critical vulnerabilities have been identified in Adobe Flash Player that could allow an attacker who successfully exploits these potential vulnerabilities to take control of the affected system. A malicious SWF must be loaded in Flash Player by the user for an attacker to exploit these potential vulnerabilities. Users are recommended to update to the most current version of Flash Player available for their platform.