Cumin (aka MRG Management Console), as used in Red Hat Enterprise MRG 2.5, allows attackers with certain database privileges to cause a denial of service (inaccessible page) via a non-ASCII character in the name of a link.

Cumin before 0.1.5444, as used in Red Hat Enterprise Messaging, Realtime, and Grid (MRG) 2.0, uses predictable random numbers to generate session keys, which makes it easier for remote attackers to guess the session key.

Cumin before 0.1.5444, as used in Red Hat Enterprise Messaging, Realtime, and Grid (MRG) 2.0, does not properly restrict access to resources, which allows remote attackers to obtain sensitive information via unspecified vectors related to (1) "web pages," (2) "export functionality," and (3) "image viewing."

Red Hat Network (RHN) Configuration Client (rhncfg-client) in rhncfg before 5.10.27-8 uses weak permissions (world-readable) for /var/log/rhncfg-actions, which allows local users to obtain sensitive information about the rhncfg-client actions by reading the file.

389 Directory Server before 1.2.11.6 (aka Red Hat Directory Server before 8.2.10-3), after the password for a LDAP user has been changed and before the server has been reset, allows remote attackers to read the plaintext password via the unhashed#user#password attribute.

Integer overflow in the ordered_malloc function in boost/pool/pool.hpp in Boost Pool before 3.9 makes it easier for context-dependent attackers to perform memory-related attacks such as buffer overflows via a large memory chunk size value, which causes less memory to be allocated than expected.

Multiple integer overflows in the (1) GC_generic_malloc and (2) calloc functions in malloc.c, and the (3) GC_generic_malloc_ignore_off_page function in mallocx.c in Boehm-Demers-Weiser GC (libgc) before 7.2 make it easier for context-dependent attackers to perform memory-related attacks such as buffer overflows via a large size value, which causes less memory to be allocated than expected.

Oracle Mojarra 2.1.7 does not properly "clean up" the FacesContext reference during startup, which allows local users to obtain context information an access resources from another WAR file by calling the FacesContext.getCurrentInstance function.

The main function in tools/hv/hv_kvp_daemon.c in hypervkvpd, as distributed in the Linux kernel before 3.4.5, does not validate the origin of Netlink messages, which allows local users to spoof Netlink communication via a crafted connector message.

libraries/libldap/tls_m.c in OpenLDAP, possibly 2.4.31 and earlier, when using the Mozilla NSS backend, always uses the default cipher suite even when TLSCipherSuite is set, which might cause OpenLDAP to use weaker ciphers than intended and make it easier for remote attackers to obtain sensitive information.

Multiple heap-based buffer overflows in the XML manifest encryption tag parsing functionality in OpenOffice.org and LibreOffice before 3.5.5 allow remote attackers to cause a denial of service and possibly execute arbitrary code via a crafted Open Document Text (.odt) file with (1) a child tag within an incorrect parent tag, (2) duplicate tags, or (3) a Base64 ChecksumAttribute whose length is not evenly divisible by four.

The sosreport utility in the Red Hat sos package before 2.2-29 does not remove the root user password information from the Kickstart configuration file (/root/anaconda-ks.cfg) when creating an archive of debugging information, which might allow attackers to obtain passwords or password hashes.

extensions/libxt_tcp.c in iptables through 1.4.21 does not match TCP SYN+FIN packets in --syn rules, which might allow remote attackers to bypass intended firewall restrictions via crafted packets. NOTE: the CVE-2012-6638 fix makes this issue less relevant.

Multiple cross-site scripting (XSS) vulnerabilities in Red Hat Certificate System (RHCS) before 8.1.1 and Dogtag Certificate System allow remote attackers to inject arbitrary web script or HTML via unspecified parameters to the (1) System Agent or (2) End Entity pages.

The Active Record component in Ruby on Rails 3.0.x before 3.0.13, 3.1.x before 3.1.5, and 3.2.x before 3.2.4 does not properly implement the passing of request data to a where method in an ActiveRecord class, which allows remote attackers to conduct certain SQL injection attacks via nested query parameters that leverage unintended recursion, a related issue to CVE-2012-2695.

actionpack/lib/action_dispatch/http/request.rb in Ruby on Rails before 3.0.13, 3.1.x before 3.1.5, and 3.2.x before 3.2.4 does not properly consider differences in parameter handling between the Active Record component and the Rack interface, which allows remote attackers to bypass intended database-query restrictions and perform NULL checks via a crafted request, as demonstrated by certain "[nil]" values, a related issue to CVE-2012-2694.

Buffer overflow in the SQLDriverConnect function in unixODBC 2.3.1 allows local users to cause a denial of service (crash) via a long string in the DRIVER option. NOTE: this issue might not be a vulnerability, since the ability to set this option typically implies that the attacker already has legitimate access to cause a DoS or execute code, and therefore the issue would not cross privilege boundaries. There may be limited attack scenarios if isql command-line options are exposed to an attacker, although it seems likely that other, more serious issues would also be exposed, and this issue might not cross privilege boundaries in that context.

Buffer overflow in the SQLDriverConnect function in unixODBC 2.0.10, 2.3.1, and earlier allows local users to cause a denial of service (crash) via a long string in the FILEDSN option. NOTE: this issue might not be a vulnerability, since the ability to set this option typically implies that the attacker already has legitimate access to cause a DoS or execute code, and therefore the issue would not cross privilege boundaries. There may be limited attack scenarios if isql command-line options are exposed to an attacker, although it seems likely that other, more serious issues would also be exposed, and this issue might not cross privilege boundaries in that context.

PostgreSQL 8.3.x before 8.3.19, 8.4.x before 8.4.12, 9.0.x before 9.0.8, and 9.1.x before 9.1.4 allows remote authenticated users to cause a denial of service (server crash) by adding the (1) SECURITY DEFINER or (2) SET attributes to a procedural language's call handler.

arpwatch 2.1a15, as used by Red Hat, Debian, Fedora, and possibly others, does not properly drop supplementary groups, which might allow attackers to gain root privileges by leveraging other vulnerabilities in the daemon.