Bug 913647 (CVE-2013-0785, CVE-2013-0786)

Summary: CVE-2013-0785 CVE-2013-0786 bugzilla: XSS and information leak flaws fixed in 3.6.13/4.0.10/4.2.5/4.4rc2
Product: [Other] Security Response Reporter: Vincent Danen <vdanen>
Component: vulnerabilityAssignee: Red Hat Product Security <security-response-team>
Status: CLOSED ERRATA QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: unspecifiedCC: itamar, xavier
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: bugzilla 3.6.13, bugzilla 4.0.10, bugzilla 4.2.5, bugzilla 4.4rc2 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2021-10-19 22:00:27 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On: 913649    
Bug Blocks:    

Description Vincent Danen 2013-02-21 17:56:57 UTC
Two flaws were reported as fixed in upstream bugzilla [1]:

Vulnerability Details
=====================

Class:       Cross-Site Scripting
Versions:    2.0 to 3.6.12, 3.7.1 to 4.0.9, 4.1.1 to 4.2.4,
             4.3.1 to 4.4rc1
Fixed In:    3.6.13, 4.0.10, 4.2.5, 4.4rc2
Description: When viewing a single bug report, which is the default,
             the bug ID is validated and rejected if it is invalid.
             But when viewing several bug reports at once, which is
             specified by the format=multiple parameter, invalid bug
             IDs can go through and are sanitized in the HTML page
             itself. But when an invalid page format is passed to the
             CGI script, the wrong HTML page is called and data are not
             correctly sanitized, which can lead to XSS.
References:  https://bugzilla.mozilla.org/show_bug.cgi?id=842038
CVE Number:  CVE-2013-0785

Class:       Information Leak
Versions:    2.17.1 to 3.6.12, 3.7.1 to 4.0.9
Fixed In:    3.6.13, 4.0.10
Description: When running a query in debug mode, the generated SQL
             query used to collect the data is displayed. The way this
             SQL query is built permits the user to determine if some
             confidential field value (such as a product name) exists.
             This problem only affects Bugzilla 4.0.9 and older. Newer
             releases are not affected by this issue.
References:  https://bugzilla.mozilla.org/show_bug.cgi?id=824399
CVE Number:  CVE-2013-0786


Currently 4.0.10 is in Fedora 17 testing and 4.2.5 is in Fedora 18 testing.

The latest version of Bugzilla in EPEL6 is 3.4.14; the last build was 20120420 to fix a security flaw but was never pushed to stable.  Likewise EPEL5 is shipping with 3.2.10 and a security fix is sitting in testing.  I recommend that EPEL gets bugzilla updated to the latest 3.6.13 for both versions so that it can continue to receive security fixes from upstream in a more timely fashion (or drop it since the versions in EPEL are out-dated and have known security flaws).

Comment 1 Vincent Danen 2013-02-21 17:58:39 UTC
External References:

http://www.bugzilla.org/security/3.6.12/

Comment 2 Vincent Danen 2013-02-21 17:59:21 UTC
Created bugzilla tracking bugs for this issue

Affects: epel-all [bug 913649]

Comment 3 Frédéric Buclin 2013-02-24 18:12:08 UTC
(In reply to comment #0)
> recommend that EPEL gets bugzilla updated to the latest 3.6.13 for both
> versions so that it can continue to receive security fixes from upstream in
> a more timely fashion

I honestly wouldn't recommend to jump to another branch in older distros as the DB schema and the codebase are different in Bugzilla 3.2, 3.4 and 3.6. If someone made some customizations, it's very likely that they will break during the major upgrade.

Also, upgrading to the 3.6 branch to get new security fixes won't help much as we don't expect any other release on this branch. Bugzilla 4.4 is almost there, and this means the EOL for Bugzilla 3.6.

Comment 4 Vincent Danen 2013-02-26 18:34:28 UTC
So you think it is better to keep older/insecure packages?  Because current EPEL packages are also missing CVE-2012-1969, CVE-2012-3981, and CVE-2012-4747 fixes (in addition to these).  They've not been touched in almost a year.