Bug 167676

Summary: CVE-2005-4744 Multiple freeradius security issues
Product: Red Hat Enterprise Linux 4 Reporter: Josh Bressers <bressers>
Component: freeradiusAssignee: Thomas Woerner <twoerner>
Status: CLOSED ERRATA QA Contact:
Severity: low Docs Contact:
Priority: medium    
Version: 4.0CC: milan.kerslager, security-response-team
Target Milestone: ---Keywords: Security
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard: impact=low,reported=20050501,source=vendorsec
Fixed In Version: RHSA-2006-0271 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-04-04 08:45:46 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:

Description Josh Bressers 2005-09-06 21:25:00 UTC 
A number of issues have been found in freeradius:

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/lib/token.c.diff?r1=1.17&r2=1.18
http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_sql/drivers/rlm_sql_unixodbc/sql_unixodbc.c.diff?r1=1.13&r2=1.14
* moderate
Off by one error (\0 being written to end of string, probably not even a crash)

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/main/session.c.diff?r1=1.27&r2=1.28
http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_exec/exec.c.diff?r1=1.2&r2=1.3
* low
Use per-system max fd to close, rather than hard-coded number

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/main/xlat.c.diff?r1=1.101&r2=1.102
* low
If strftime(3) returns 0, the contents of the string array is undefined,
therefore it should not be copied.  This could lead to the leaking of 40 bytes
off the stack (random unitialized memory).

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/main/xlat.c.diff?r1=1.72.2.6&r2=1.72.2.7
* moderate/low
Adds boundly length checks, whithout which it could be possible to overflow a
var on the stack.
* As best as I can tell, this is coming from the server, not a malicious source.

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_exec/exec.c.diff?r1=1.1&r2=1.2
* moderate
Adds a boundry length check.  Possible stack buffer overflow. This sould be a
crash only.

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_ldap/rlm_ldap.c.diff?r1=1.153&r2=1.154
http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_ldap/rlm_ldap.c.diff?r1=1.122.2.6&r2=1.122.2.7
* moderate
Properly escape ldap data.  This could possible leak ldap data via a malformed
query, without proper acls.

http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/src/modules/rlm_sql/drivers/rlm_sql_unixodbc/sql_unixodbc.c.diff?r1=1.11.2.1&r2=1.11.2.2
* low/bug
Make things a bit more thread safe by introducing a horrible horrible hack.
Comment 1 Josh Bressers 2005-09-06 21:29:55 UTC 
These issues may also affect RHEL3.
Comment 3 Josh Bressers 2005-09-07 12:53:30 UTC 
These issues should be considered public
Comment 4 Josh Bressers 2005-09-12 12:19:10 UTC 
After reading upstream's response to these issues, I'm marking the severity as
low.  Given I'm not an expert with the FreeRADIUS source, I purposely marked
some of these issues as "moderate" severity to prevent us from not giving enough
attention to something that may have needed it.

http://www.freeradius.org/security/20050909-response-to-suse.txt
Comment 6 Josh Bressers 2006-03-28 14:58:36 UTC 
After a short discussion with MITRE and some analysis, only the off my one error
in sql_unixodbc.c is being considered a security issue.
Comment 9 Red Hat Bugzilla 2006-04-04 08:45:46 UTC 
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2006-0271.html