blob: 41fe3ae0bab66fa13128592fb2b121274d21efc7 [file] [log] [blame]
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE glsa SYSTEM "http://www.gentoo.org/dtd/glsa.dtd">
<glsa id="200608-02">
<title>Mozilla SeaMonkey: Multiple vulnerabilities</title>
<synopsis>
The Mozilla Foundation has reported numerous security vulnerabilities
related to Mozilla SeaMonkey.
</synopsis>
<product type="ebuild">SeaMonkey</product>
<announced>August 03, 2006</announced>
<revised>August 03, 2006: 01</revised>
<bug>141842</bug>
<access>remote</access>
<affected>
<package name="www-client/seamonkey" auto="yes" arch="*">
<unaffected range="ge">1.0.3</unaffected>
<vulnerable range="lt">1.0.3</vulnerable>
</package>
</affected>
<background>
<p>
The Mozilla SeaMonkey project is a community effort to deliver
production-quality releases of code derived from the application
formerly known as "Mozilla Application Suite".
</p>
</background>
<description>
<p>
The following vulnerabilities have been reported:
</p>
<ul>
<li>Benjamin Smedberg discovered that chrome URL's could be made to
reference remote files.</li>
<li>Developers in the Mozilla community
looked for and fixed several crash bugs to improve the stability of
Mozilla clients, which could lead to the execution of arbitrary code by
a remote attacker.</li>
<li>"shutdown" reports that cross-site
scripting (XSS) attacks could be performed using the construct
XPCNativeWrapper(window).Function(...), which created a function that
appeared to belong to the window in question even after it had been
navigated to the target site.</li>
<li>"shutdown" reports that scripts
granting the UniversalBrowserRead privilege can leverage that into the
equivalent of the far more powerful UniversalXPConnect since they are
allowed to "read" into a privileged context.</li>
<li>"moz_bug_r_a4"
reports that A malicious Proxy AutoConfig (PAC) server could serve a
PAC script that can execute code with elevated privileges by setting
the required FindProxyForURL function to the eval method on a
privileged object that leaked into the PAC sandbox.</li>
<li>"moz_bug_r_a4" discovered that Named JavaScript functions have a
parent object created using the standard Object() constructor
(ECMA-specified behavior) and that this constructor can be redefined by
script (also ECMA-specified behavior).</li>
<li>Igor Bukanov and
shutdown found additional places where an untimely garbage collection
could delete a temporary object that was in active use.</li>
<li>Georgi
Guninski found potential integer overflow issues with long strings in
the toSource() methods of the Object, Array and String objects as well
as string function arguments.</li>
<li>H. D. Moore reported a testcase
that was able to trigger a race condition where JavaScript garbage
collection deleted a temporary variable still being used in the
creation of a new Function object.</li>
<li>A malicious page can hijack
native DOM methods on a document object in another domain, which will
run the attacker's script when called by the victim page.</li>
<li>Secunia Research has discovered a vulnerability which is caused due
to an memory corruption error within the handling of simultaneously
happening XPCOM events. This leads to use of a deleted timer
object.</li>
<li>An anonymous researcher for TippingPoint and the Zero
Day Initiative showed that when used in a web page Java would reference
properties of the window.navigator object as it started up.</li>
<li>Thilo Girmann discovered that in certain circumstances a JavaScript
reference to a frame or window was not properly cleared when the
referenced content went away.</li>
</ul>
</description>
<impact type="normal">
<p>
A user can be enticed to open specially crafted URLs, visit webpages
containing malicious JavaScript or execute a specially crafted script.
These events could lead to the execution of arbitrary code, or the
installation of malware on the user's computer.
</p>
</impact>
<workaround>
<p>
There is no known workaround at this time.
</p>
</workaround>
<resolution>
<p>
All Thunderbird users should upgrade to the latest version:
</p>
<code>
# emerge --sync
# emerge --ask --oneshot --verbose &quot;&gt;=www-client/seamonkey-1.0.3&quot;</code>
</resolution>
<references>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3113">CVE-2006-3113</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3677">CVE-2006-3677</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3801">CVE-2006-3801</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3802">CVE-2006-3802</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3803">CVE-2006-3803</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3804">CVE-2006-3804</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3805">CVE-2006-3805</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3806">CVE-2006-3806</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3807">CVE-2006-3807</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3808">CVE-2006-3808</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3809">CVE-2006-3809</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3810">CVE-2006-3810</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3811">CVE-2006-3811</uri>
<uri link="http://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2006-3812">CVE-2006-3812</uri>
</references>
<metadata tag="requester" timestamp="Fri, 28 Jul 2006 14:37:24 +0000">
DerCorny
</metadata>
<metadata tag="submitter" timestamp="Fri, 28 Jul 2006 18:00:11 +0000">
dizzutch
</metadata>
<metadata tag="bugReady" timestamp="Thu, 03 Aug 2006 16:55:20 +0000">
DerCorny
</metadata>
</glsa>