Bug 444179 (CVE-2012-5533) – <www-servers/lighttpd-1.4.32 : HTTP Header Processing Denial of Service Vulnerability (CVE-2012-5533)

• Product: Gentoo Security
• Status: RESOLVED FIXED

[Description Agostino Sarubbo 2012-11-21 16:35:17 UTC]

From https://secunia.com/advisories/51268/ :

Description
A vulnerability has been reported in lighttpd, which can be exploited by malicious people to cause 
a DoS (Denial of Service).

The vulnerability is caused due to an error in the "http_request_split_value()" function 
(src/request.c) when processing certain HTTP headers. This can be exploited to cause an endless 
loop denying further request processing via a specially crafted "Connection" header.

The vulnerability is reported in version 1.4.31 only.


Solution
Update to version 1.4.32.

Provided and/or discovered by
The vendor credits Jesse Sipprell, McClatchy Interactive, Inc.

Original Advisory
http://download.lighttpd.net/lighttpd/security/lighttpd_sa_2012_01.txt

[Comment 1 Markos Chandras (RETIRED) 2012-11-21 18:54:28 UTC]

Ebuild in portage. Please test with various use flag combinations

[Comment 2 Agostino Sarubbo 2012-11-22 09:57:19 UTC]

(In reply to comment #1)
> Ebuild in portage. Please test with various use flag combinations

http://bpaste.net/raw/59688/ should be enough.


Arches, please test and mark stable:
=www-servers/lighttpd-1.4.32
Target keywords : "alpha amd64 arm hppa ia64 ppc ppc64 sparc x86"

[Comment 3 Markos Chandras (RETIRED) 2012-11-22 10:03:30 UTC]

(In reply to comment #2)
> (In reply to comment #1)
> > Ebuild in portage. Please test with various use flag combinations
> 
> http://bpaste.net/raw/59688/ should be enough.
> 
> 
> Arches, please test and mark stable:
> =www-servers/lighttpd-1.4.32
> Target keywords : "alpha amd64 arm hppa ia64 ppc ppc64 sparc x86"

I asked you multiple times to not use online paste services as they may not be available in the future (or when are teams get around to handle this bug)

[Comment 4 Agostino Sarubbo 2012-11-22 10:11:21 UTC]

(In reply to comment #3)
> I asked you multiple times to not use online paste services as they may not
> be available in the future (or when are teams get around to handle this bug)

This paste will never be useful in the future, is just to say it passed multiple compile test

[Comment 5 Agostino Sarubbo 2012-11-22 10:33:13 UTC]

amd64 stable

[Comment 6 Jeroen Roovers (RETIRED) 2012-11-22 15:26:46 UTC]

Stable for HPPA.

[Comment 7 Jeroen Roovers (RETIRED) 2012-11-22 15:28:10 UTC]

(In reply to comment #4)
> (In reply to comment #3)
> > I asked you multiple times to not use online paste services as they may not
> > be available in the future (or when are teams get around to handle this bug)
> 
> This paste will never be useful in the future, is just to say it passed
> multiple compile test

   "You should never use URL to point to pastebins for error messages, logs,
    emerge --info output, screenshots or similar information. Instead, these
    should always be attached to the bug."

http://www.gentoo.org/doc/en/bugzilla-howto.xml

[Comment 8 Anthony Basile 2012-11-22 17:30:45 UTC]

stable ppc ppc64

[Comment 9 Mark Reiche 2012-11-22 22:02:56 UTC]

x86 ok.
emerge, test and run successfully.
repoman complains about file.size of lighttpd-1.4.29-mod_uploadprogress.patch

my use-flags: bzip2 gdbm pcre rrdtool ssl {test} webdav zlib -doc -fam -ipv6 -kerberos -ldap -libev -lua -memcache -minimal -mmap -mysql -php (-selinux) -uploadprogress -xattr

[Comment 10 Anthony Basile 2012-11-23 04:56:58 UTC]

stable arm

[Comment 11 Raúl Porcel (RETIRED) 2012-11-25 19:03:20 UTC]

alpha/ia64/sh/sparc/x86 stable

[Comment 12 Sean Amoss (RETIRED) 2012-11-26 01:16:32 UTC]

Thanks, everyone.

GLSA vote: yes.

[Comment 13 GLSAMaker/CVETool Bot 2012-11-28 22:37:58 UTC]

CVE-2012-5533 (http://nvd.nist.gov/nvd.cfm?cvename=CVE-2012-5533):
  The http_request_split_value function in request.c in lighttpd 1.4.32 allows
  remote attackers to cause a denial of service (infinite loop) via a request
  with a header containing an empty token, as demonstrated using the
  "Connection: TE,,Keep-Alive" header.

[Comment 14 Stefan Behte (RETIRED) 2012-12-16 21:49:31 UTC]

Vote: yes.
GLSA request filed.

[Comment 15 GLSAMaker/CVETool Bot 2014-06-13 20:44:09 UTC]

This issue was resolved and addressed in
 GLSA 201406-10 at http://security.gentoo.org/glsa/glsa-201406-10.xml
by GLSA coordinator Sergey Popov (pinkbyte).