Safe: Intended restriction bypass via object references

The Safe (aka Safe.pm) module before 2.25 for Perl allows context-dependent attackers to bypass intended (1) Safe::reval and (2) Safe::rdo access restrictions, and inject and execute arbitrary code, via vectors involving implicitly called methods and implicitly blessed objects, as demonstrated by the (a) DESTROY and (b) AUTOLOAD methods, related to "automagic methods."

PostgreSQL: PL/Tcl Intended restriction bypass

The PL/Tcl implementation in PostgreSQL 7.4 before 7.4.29, 8.0 before 8.0.25, 8.1 before 8.1.21, 8.2 before 8.2.17, 8.3 before 8.3.11, 8.4 before 8.4.4, and 9.0 Beta before 9.0 Beta 2 loads Tcl code from the pltcl_modules table regardless of the table's ownership and permissions, which allows remote authenticated users, with database-creation privileges, to execute arbitrary Tcl code by creating this table and inserting a crafted Tcl script.

rhn_satellite: Improper channel comps information management

Red Hat Network (RHN) Satellite 5.3 and 5.4 exposes a dangerous, obsolete XML-RPC API, which allows remote authenticated users to access arbitrary files and cause a denial of service (failed yum operations) via vectors related to configuration and package group (comps.xml) files for channels.

dbus-glib: property access not validated

DBus-GLib 0.73 disregards the access flag of exported GObject properties, which allows local users to bypass intended access restrictions and possibly cause a denial of service by modifying properties, as demonstrated by properties of the (1) DeviceKit-Power, (2) NetworkManager, and (3) ModemManager services.

MediaWiki: Two security fixes in v1.15.2

thumb.php in MediaWiki before 1.15.2, when used with access-restriction mechanisms such as img_auth.php, does not check user permissions before providing scaled images, which allows remote attackers to bypass intended access restrictions and read private images via unspecified manipulations.

Bugzilla: Sensitive time-tracking information disclosure via specially-crafted URL

Search.pm in Bugzilla 2.17.1 through 3.2.6, 3.3.1 through 3.4.6, 3.5.1 through 3.6, and 3.7 allows remote attackers to obtain potentially sensitive time-tracking information via a crafted search URL, related to a "boolean chart search."

Firefox: Spoofing attacks via vectors involving 'No Content' status code or via a windows.stop call

The startDocumentLoad function in browser/base/content/browser.js in Mozilla Firefox 3.5.x before 3.5.11 and 3.6.x before 3.6.7, and SeaMonkey before 2.0.6, does not properly implement the Same Origin Policy in certain circumstances related to the about:blank document and a document that is currently loading, which allows (1) remote web servers to conduct spoofing attacks via vectors involving a 204 (aka No Content) status code, and allows (2) remote attackers to conduct spoofing attacks via vectors involving a window.stop call.

Mozilla Same-origin bypass using canvas context

Mozilla Firefox before 3.6.7 and Thunderbird before 3.1.1 do not properly implement read restrictions for CANVAS elements, which allows remote attackers to obtain sensitive cross-origin information via vectors involving reference retention and node deletion.

main/acl.c in Asterisk Open Source 1.6.0.x before 1.6.0.25, 1.6.1.x before 1.6.1.17, and 1.6.2.x before 1.6.2.5 does not properly enforce remote host access controls when CIDR notation "/0" is used in permit= and deny= configuration rules, which causes an improper arithmetic shift and might allow remote attackers to bypass ACL rules and access services from unauthorized hosts.

The memory-management implementation in the Virtual Machine Monitor (aka VMM or hypervisor) in Microsoft Virtual PC 2007 Gold and SP1, Virtual Server 2005 Gold and R2 SP1, and Windows Virtual PC does not properly restrict access from the guest OS to memory locations in the VMM work area, which allows context-dependent attackers to bypass certain anti-exploitation protection mechanisms on the guest OS via crafted input to a vulnerable application. NOTE: the vendor reportedly found that only systems with an otherwise vulnerable application are affected, because "the memory areas accessible from the guest cannot be leveraged to achieve either remote code execution or elevation of privilege and ... no data from the host is exposed to the guest OS."

MoinMoin 1.7.1 allows remote attackers to bypass the textcha protection mechanism by modifying the textcha-question and textcha-answer fields to have empty values.

acroread: multiple code execution flaws (APSB10-15)

Adobe Reader and Acrobat 9.x before 9.3.3, and 8.x before 8.2.3 on Windows and Mac OS X, do not restrict the contents of one text field in the Launch File warning dialog, which makes it easier for remote attackers to trick users into executing an arbitrary local program that was specified in a PDF document, as demonstrated by a text field that claims that the Open button will enable the user to read an encrypted message.

The installation for Microsoft Open XML File Format Converter for Mac sets insecure ACLs for the /Applications folder, which allows local users to execute arbitrary code by replacing the executable with a Trojan Horse, aka "Mac Office Open XML Permissions Vulnerability."

perms.cpp in March Hare Software CVSNT 2.0.58, 2.5.01, 2.5.02, 2.5.03 before build 3736, 2.5.04 before build 2862; CVS Suite 2.5.03, 2008 before build 3736, and 2009 before 3729 allows remote attackers to bypass the permissions check, modify arbitrary modules and directories within CVSROOT, and execute arbitrary code via a crafted branch name ACL, possibly related to incorrect inheritance.

Director Agent 6.1 before 6.1.2.3 in IBM Systems Director on AIX and Linux uses incorrect permissions for the (1) diruninstall and (2) opt/ibm/director/bin/wcitinst scripts, which allows local users to gain privileges by executing these scripts.

page/Geolocation.cpp in WebCore in WebKit before r56188 and before 1.2.5 does not properly restrict access to the lastPosition function, which has unspecified impact and remote attack vectors, aka rdar problem 7746357.

WebKit in Apple Safari before 5.0 on Mac OS X 10.5 through 10.6 and Windows, and before 4.1 on Mac OS X 10.4, allows remote attackers to bypass intended restrictions on outbound connections to "non-default TCP ports" via a crafted port number, related to an "integer truncation issue." NOTE: this may overlap CVE-2010-1099.

WebKit in Apple Safari before 5.0 on Mac OS X 10.5 through 10.6 and Windows, and before 4.1 on Mac OS X 10.4, does not properly restrict the reading of a canvas that contains an SVG image pattern from a different web site, which allows remote attackers to read images from other sites via a crafted canvas, related to a "cross-site image capture issue."

JBossEAP status servlet info leak

Red Hat JBoss Enterprise Application Platform (aka JBoss EAP or JBEAP) 4.2 before 4.2.0.CP09 and 4.3 before 4.3.0.CP08 allows remote attackers to obtain sensitive information about "deployed web contexts" via a request to the status servlet, as demonstrated by a full=true query string. NOTE: this issue exists because of a CVE-2008-3273 regression.

rhn-client-tools: authorized information disclosure

yum-rhn-plugin in Red Hat Network Client Tools (aka rhn-client-tools) on Red Hat Enterprise Linux (RHEL) 5 and Fedora uses world-readable permissions for the /var/spool/up2date/loginAuth.pkl file, which allows local users to access the Red Hat Network profile, and possibly prevent future security updates, by leveraging authentication data from this file.