SQL injection vulnerability in the Backend History module in TYPO3 4.5.x before 4.5.21, 4.6.x before 4.6.14, and 4.7.x before 4.7.6 allows remote authenticated backend users to execute arbitrary SQL commands via unspecified vectors.

wireshark: Infinite loop in the DICOM dissector

The dissect_dcm_main function in epan/dissectors/packet-dcm.c in the DICOM dissector in Wireshark 1.2.x before 1.2.17 and 1.4.x before 1.4.7 allows remote attackers to cause a denial of service (infinite loop) via an invalid PDU length.

libxslt: two DoS issues fixed in 1.1.28

libxslt before 1.1.28 allows remote attackers to cause a denial of service (NULL pointer dereference and crash) via an (1) empty match attribute in a XSL key to the xsltAddKey function in keys.c or (2) uninitialized variable to the xsltDocumentFunction function in functions.c.

subscription-manager: rhn-migrate-classic-to-rhsm missing SSL certificate verification

rhn-migrate-classic-to-rhsm tool in Red Hat subscription-manager does not verify the Red Hat Network Classic server's X.509 certificate when migrating to a Certificate-based Red Hat Network, which allows remote man-in-the-middle attackers to obtain sensitive information such as user credentials.

Stack-based buffer overflow in utp.cpp in libutp, as used in Transmission before 2.74 and possibly other products, allows remote attackers to cause a denial of service (crash) and possibly execute arbitrary code via crafted "micro transport protocol packets."

Puppet: Directory /var/log/puppet is world readable

Red Hat OpenStack Essex and Folsom creates the /var/log/puppet directory with world-readable permissions, which allows local users to obtain sensitive information such as Puppet log files.

Candlepin: Re-enable manifest signature checking

Candlepin before 0.7.24, as used in Red Hat Subscription Asset Manager before 1.2.1, does not properly check manifest signatures, which allows local users to modify manifests.

Conductor: Unprivileged user can change their own Maximum Running Instances quota

The Administer tab in Aeolus Conductor allows remote authenticated users to bypass intended quota restrictions by updating the Maximum Running Instances quota user setting.

Configserver: Passwords from application blueprint stored plaintext in configserver.log

Aeolus Configuration Server, as used in Red Hat CloudForms Cloud Engine before 1.1.2, uses world-readable permissions for /var/log/aeolus-configserver/configserver.log, which allows local users to read plaintext passwords by reading the log file.

rhev: rhevm-manage-domains logs admin passwords

The domain management tool (rhevm-manage-domains) in Red Hat Enterprise Virtualization Manager (RHEV-M) 3.1 and earlier, when the validate action is enabled, logs the administrative password to a world-readable log file, which allows local users to obtain sensitive information by reading this file.

rubygem-rack: parsing Content-Disposition header DoS

lib/rack/multipart.rb in Rack before 1.1.4, 1.2.x before 1.2.6, 1.3.x before 1.3.7, and 1.4.x before 1.4.2 uses an incorrect regular expression, which allows remote attackers to cause a denial of service (infinite loop) via a crafted Content-Disposion header.

blog/rsslib.php in Moodle 2.1.x before 2.1.10, 2.2.x before 2.2.7, 2.3.x before 2.3.4, and 2.4.x before 2.4.1 continues to provide a blog RSS feed after blogging is disabled, which allows remote attackers to obtain sensitive information by reading this feed.

Multiple open redirect vulnerabilities in Moodle 2.2.x before 2.2.7, 2.3.x before 2.3.4, and 2.4.x before 2.4.1 allow remote attackers to redirect users to arbitrary web sites and conduct phishing attacks via vectors related to (1) backup/backupfilesedit.php, (2) comment/comment_post.php, (3) course/switchrole.php, (4) mod/wiki/filesedit.php, (5) tag/coursetags_add.php, or (6) user/files.php.

grade/edit/outcome/edit_form.php in Moodle 1.9.x through 1.9.19, 2.1.x before 2.1.10, 2.2.x before 2.2.7, 2.3.x before 2.3.4, and 2.4.x before 2.4.1 does not properly enforce the moodle/grade:manage capability requirement, which allows remote authenticated users to convert custom outcomes into standard site-wide outcomes by leveraging the teacher role and using the re-editing feature.

cronie: fd leak in 1.4.8

File descriptor leak in cronie 1.4.8, when running in certain environments, might allow local users to read restricted files, as demonstrated by reading /etc/crontab.

activemq: Multiple XSS flaws in web demos

Multiple cross-site scripting (XSS) vulnerabilities in the web demos in Apache ActiveMQ before 5.8.0 allow remote attackers to inject arbitrary web script or HTML via (1) the refresh parameter to PortfolioPublishServlet.java (aka demo/portfolioPublish or Market Data Publisher), or vectors involving (2) debug logs or (3) subscribe messages in webapp/websocket/chat.js. NOTE: AMQ-4124 is covered by CVE-2012-6551.

qemu: e1000 driver buffer overflow when processing large packets when SBP and LPE flags are disabled

Buffer overflow in the e1000_receive function in the e1000 device driver (hw/e1000.c) in QEMU 1.3.0-rc2 and other versions, when the SBP and LPE flags are disabled, allows remote attackers to cause a denial of service (guest OS crash) and possibly execute arbitrary guest code via a large packet.

Jenkins: cross-site scripting vulnerability

Cross-site scripting (XSS) vulnerability in Jenkins before 1.491, Jenkins LTS before 1.480.1, and Jenkins Enterprise 1.424.x before 1.424.6.13, 1.447.x before 1.447.4.1, and 1.466.x before 1.466.10.1 allows remote authenticated users with write access to inject arbitrary web script or HTML via unspecified vectors.

Jenkins: HTTP response splitting

CRLF injection vulnerability in Jenkins before 1.491, Jenkins LTS before 1.480.1, and Jenkins Enterprise 1.424.x before 1.424.6.13, 1.447.x before 1.447.4.1, and 1.466.x before 1.466.10.1 allows remote attackers to inject arbitrary HTTP headers and conduct HTTP response splitting attacks via unspecified vectors.

xulrunner: Google CityHash hash function collisions (oCERT-2012-001)

Google CityHash computes hash values without properly restricting the ability to trigger hash collisions predictably, which allows context-dependent attackers to cause a denial of service (CPU consumption) via crafted input to an application that maintains a hash table, as demonstrated by a universal multicollision attack.