A memory leak was found in the way libpng processed malformed Portable Network Graphics (PNG) images with Physical Scale (sCAL) extension. A remote attacker could create a specially-crafted PNG image and trick the local user into opening it in an application, using the libpng library, leading to denial of service (relevant libpng-based application crash).
References: [1] http://www.libpng.org/pub/png/libpng.html
CVE Request: [2] http://www.openwall.com/lists/oss-security/2010/06/28/2
Description of problem: Due to bad parsing of malformed X.25 facilities, a remote attacker can cause a kernel panic due to heap corruption (assuming both parties are communicating using X.25).
Reference: http://www.spinics.net/lists/netdev/msg145786.html http://www.spinics.net/lists/netdev/msg145873.html
Acknowledgements:
Red Hat would like to thank Dan Rosenberg for reporting this issue.
Description of problem: In bcmconnect() (in net/can/bcm.c), there is the following code:
sprintf(bo->procname, "%p", sock);
"procname" is a 9-byte char array. On 64-bit platforms, up to 17 bytes may be copied into the buffer. Fortunately, structure padding will most likely prevent this from being a problem, except for the trailing NULL byte, which may overwrite the first byte of the next heap object.
Reference: http://www.spinics.net/lists/netdev/msg145791.html
Acknowledgements:
Red Hat would like to thank Dan Rosenberg for reporting this issue.
Last updated 24 July 2024
Description of problem: INET-DIAG is inconsistent about how it looks up the bytecode contained in a netlink message, making it possible for a user to cause the kernel to execute unaudited INET-DIAG bytecode. This can be abused to make the kernel enter an infinite loop, and possibly other consequences.
Reference: http://www.spinics.net/lists/netdev/msg145899.html
Acknowledgements:
Red Hat would like to thank Nelson Elhage for reporting this issue.
Description of problem: Reported by Nelson Elhage.
CVE-2010-3848 AFECONET kernel stack overflow CVE-2010-3849 AFECONET saddr->cookie NULL dereferences CVE-2010-3850 AFECONET SIOCSIFADDR ioctl does not check privileges
The two main vulnerabilities lie in the econetsendmsg() function. If CONFIGECONETAUNUDP, the function declares a variable-length array of iovec's on the stack:
static int econetsendmsg(struct kiocb iocb, struct socket sock, struct msghdr msg, sizet len) { ... #ifdef CONFIGECONETAUNUDP ... struct iovec iov[msg->msgiovlen+1]; ... #endif
"msg->msgiovlen" is a user-controlled value <= 1024; 1024 'struct iovec's is enough to overflow the kernel stack and clobber the 'threadinfo' struct on either 32- or 64-bit systems with or without CONFIG4KSTACKS.
The code populates 'iov' with user-controlled values, so this is almost certainly exploitable for privesc.
The second bug is conveniently labeled as such for anyone who cares to look:
/ BUG: saddr may be NULL / eb->cookie = saddr->cookie;
In fact, 'saddr' comes from userspace and may be NULL (if non-NULL, it will have been copied to kernelspace), leading to a NULL-pointer dereference. This saddr->cookie dereference appears twice, once in the "hardware" case and once in the "UDP emulation" case. This is easily exploitable for an oops, but probably not anything more.
The final issue is in the ecdevioctl function. The SIOCSIFADDR ioctl does not check privileges, allowing an unprivileged user to assign econet addresses to arbitrary interfaces. It is possible this is intentional, but AFAIK it would be unusual for SIOCSIFADDR to be unprivileged, so this may be an accident.
Note that it is necessary to use this property in order to make the first two bugs exploitable, since econetsendmsg will return almost immediately if no econet addresses are configured.
Reference: http://seclists.org/oss-sec/2010/q4/236
Description of problem: Changes to tty to use struct pid happened here: ab521dc0f8e117fd808d3e425216864d60390500
mrg-1/rhel-6 are missing: 1) redo locking of tty->pgrp 47f86834bbd4193139d61d659bebf9ab9d691e37 2) tty: fix race in ttyfasync 703625118069f9f8960d356676662d3db5a9d116 3) fnctl: fmodown should call writelockirqsave/restore b04da8bfdfbbd79544cab2fadfdc12e87eb01600
Description of problem: The CONFIGSTACKGROWSDOWN variant of setupargpages() does not check the size of the argument/environment area on the stack. When it is unworkably large, shiftargpages() hits its BUGON. This is exploitable with a very large RLIMITSTACK limit, to create a crash pretty easily. Check that the initial stack is not too large to make it possible to map in any executable. We're not checking that the actual executable (or intepreter, for binfmtelf) will fit. So those mappings might clobber part of the initial stack mapping. But that is just userland lossage that userland made happen, not a kernel problem.
Upstream commit: http://git.kernel.org/linus/1b528181b2ffa14721fb28ad1bd539fe1732c583
Reference: http://grsecurity.net/~spender/64bitdos.c
Acknowledgements:
Red Hat would like to thank Brad Spengler for reporting this issue.
Last updated 24 July 2024
Apache Portable Runtime (APR) is vulnerable to a denial of service, caused by an error in the aprfnmatch() function when processing specific patterns with the "" wildcard. A remote attacker could exploit this vulnerability to consume all available CPU and memory resources resulting in a denial of service.
Devin Carraway of the Debian Security Team discovered that the upstream fix for the CVE-2008-2079 is incomplete and still makes it possible for local users to create tables via INDEX/DATA DIRECTORY directives in the MySQL data directory (/var/lib/mysql) via directory symlinks.
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=480292#25
CVE-2008-2079 was tracked via bug bug #445222.
An attacker needs following to exploit this issue: - MySQL database account with privileges to create tables - shell access to the host running MySQL database with write access to a directory accessible by the mysqld daemon process
PHP before 5.2.12 and 5.3.x before 5.3.1 does not restrict the number of temporary files created when handling a multipart/form-data POST request, which allows remote attackers to cause a denial of service (resource exhaustion), and makes it easier for remote attackers to exploit local file inclusion vulnerabilities, via multiple requests, related to lack of support for the maxfileuploads directive.
The exifreaddata function in the Exif module in PHP before 5.2.10 allows remote attackers to cause a denial of service (crash) via a malformed JPEG image with invalid offset fields, a different issue than CVE-2005-3353.
In his upcoming Blackhat paper and presentation Dan Kaminsky highlights some more issues he has found relating to SSL hash collisions and related vulnerabilities.
His second issue is all about inconsistencies in the interpretation of subject x509 names in certificates. Specifically "issue 2, attack 2c" regarding NULL terminators in a Common Name field. An attacker could create a malicious certificate containing a NULL, which, if they were able to get it signed, could confuse a client into accepting it by mistake.
According to the paper this is said to affect Firefox.
The kvmarchvcpuioctlsetsregs function in the KVM in Linux kernel 2.6 before 2.6.30, when running on x86 systems, does not validate the page table root in a KVMSETSREGS call, which allows local users to cause a denial of service (crash or hang) via a crafted cr3 value, which triggers a NULL pointer dereference in the gfntormap function.
libxml2 before 2.7.8, as used in Google Chrome before 7.0.517.44, Apple Safari 5.0.2 and earlier, and other products, reads from invalid memory locations during processing of malformed XPath expressions, which allows context-dependent attackers to cause a denial of service (application crash) via a crafted XML document.
The upstream PHP 5.2.14 and 5.3.3 releases corrected an information disclosure flaw where the varexport() function would disclose data if a fatal error occurred due to recursion, memorylimit, or execution time. The buffer is never cleared and is flushed to the user, regardless of the configured displayerrors setting, because it is considered part of the output. This could lead to a disclosure of possibly sensitive information.
The name CVE-2010-2531 has been assigned to this issue.
An example to test:
% cat test.php #!/usr/bin/php <?php @$obj->p =& $obj; varexport($obj, true); ?> % php test.php PHP Fatal error: Nesting level too deep - recursive dependency? in test.php on line 4 stdClass::setstate(array( 'p' => stdClass::setstate(array( 'p' => stdClass::setstate(array( 'p' => stdClass::setstate(array( % cat test2.php #!/usr/bin/php <?php $a[] =& $a; varexport($a, true); ?> % php test2.php PHP Fatal error: Nesting level too deep - recursive dependency? in test.php on line 4 array ( 0 => array ( 0 => array ( 0 => array ( 0 => array (
The upstream changes prevent any output from displaying, so should only display the "PHP Fatal error".
The upstream fix:
http://svn.php.net/viewvc?view=revision&revision=301143
Heap-based buffer overflow in the InsIUP function in truetype/ttinterp.c in FreeType before 2.4.0, when TrueType bytecode support is enabled, allows remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted font file.
Multiple buffer overflows in demo programs in FreeType before 2.4.0 allow remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted font file.
Heap-based buffer overflow in the MacReadPOSTResource function in base/ftobjs.c in FreeType before 2.4.0 allows remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted length value in a POST fragment header in a font file.
Robert Swiecki reported an invalid free flaw, that could possibly corrupt the heap, in freetype. This could cause applications linked against freetype to crash (SIGABRT) or, possibly, lead to the execution of arbitrary code if an attacker were able to get a victim to load a malicious font file.
The affected code is not present in Red Hat Enterprise Linux 3 (freetype 2.1.4). In Red Hat Enterprise Linux 4 and 5 this code is present, but does not cause a crash.
This issue has been given the name CVE-2010-2498.
Upstream bug report:
http://savannah.nongnu.org/bugs/index.php?30106
Upstream commit that fixes the issue:
http://git.savannah.gnu.org/cgit/freetype/freetype2.git/commit/?id=8d22746c9e5af80ff4304aef440986403a5072e2
Acknowledgements:
Red Hat would like to thank Robert Święcki of the Google Security Team for the discovery of this issue.
Integer underflow in glyph handling in FreeType before 2.4.0 allows remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted font file.
Buffer overflow in the MacReadPOSTResource function in base/ftobjs.c in FreeType before 2.4.0 allows remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted LaserWriter PS font file with an embedded PFB fragment.
Integer overflow in the grayrenderspan function in smooth/ftgrays.c in FreeType before 2.4.0 allows remote attackers to cause a denial of service (application crash) or possibly execute arbitrary code via a crafted font file.
Drupal versions 5.x and 6.x has open redirection
TYPO3 before 4.5.4 allows Information Disclosure in the backend.
Integer overflow in the exifprocessIFDTAG function in exif.c in the exif extension in PHP 5.4.0beta2 on 32-bit platforms allows remote attackers to read the contents of arbitrary memory locations or cause a denial of service via a crafted offsetval value in an EXIF header in a JPEG file, a different vulnerability than CVE-2011-0708.
MediaWiki before 1.17.1 does not check for read permission before handling action=ajax requests, which allows remote attackers to obtain sensitive information by (1) leveraging the SpecialUpload::ajaxGetExistsWarning function, or by (2) leveraging an extension, as demonstrated by the CategoryTree, ExtTab, and InlineEditor extensions.
An signedness error, leading to out of stack-based buffer read flaw was found in the way lighttpd, a lightning fast webserver with light system requirements, processed certain invalid base64 HTTP authentication tokens. A remote attacker could provide a specially crafted HTTP authentication request, leading to denial of service (lighttpd daemon crash due to an signedness error while processing the token).
Upstream bug report: [1] http://redmine.lighttpd.net/issues/2370
Upstream patch (with testcase and NEWS update): [2] http://redmine.lighttpd.net/projects/lighttpd/repository/revisions/2806/diff
References: [3] http://download.lighttpd.net/lighttpd/security/lighttpdsa201101.txt (upstream advisory) [4] http://www.openwall.com/lists/oss-security/2011/11/29/8 (CVE request) [5] http://www.openwall.com/lists/oss-security/2011/11/29/13 (CVE assignment)
An information disclosure flaw was found in the way MediaWiki, the wiki engine, processed 'curid' and 'oldid' request paramaters. A remote attacker could use this flaw to enumerate page titles on private MediaWiki installations.
Upstream bug report: [1] https://bugzilla.wikimedia.org/showbug.cgi?id=32276
An information disclosure flaw was found in the way MediaWiki, the wiki engine, performed action=ajax requests dispatching to relevant internal functions. These requests were dispatched without any read permissions checks being done. A remote attacker could use this flaw to obtain data on private MediaWiki installations.
Upstream bug report: [2] https://bugzilla.wikimedia.org/showbug.cgi?id=32616
References: [3] http://lists.wikimedia.org/pipermail/mediawiki-announce/2011-November/000104.html [4] http://www.openwall.com/lists/oss-security/2011/11/29/6 [5] https://bugs.gentoo.org/showbug.cgi?id=392383
Upstream patch (covering both of the issues): [6] http://www.mediawiki.org/wiki/Special:Code/MediaWiki/104506