[CVE-2006-0996] phpinfo() Cross Site Scripting

Debian Bug report logs - #361853
[CVE-2006-0996] phpinfo() Cross Site Scripting

Toggle useless messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Oliver Paulus <oliver@code-project.org>

To: submit@bugs.debian.org

Subject: phpinfo() Cross Site Scripting PHP 5.1.2 and 4.4.2

Date: Mon, 10 Apr 2006 23:12:38 +0200

[Message part 1 (text/plain, inline)]

Package: php4
Version: 4:4.3.10-16
Severity: grave

for more informaton see:
http://securityreason.com/achievement_securityalert/34

--
Oliver Paulus

OpenPGP
Key id: 28D9C44F
Fingerprint: EADA 62FC 07DC 3361 A3D6  4174 2DE3 C027 28D9 C44F
Public Key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x28D9C44F

[Message part 2 (application/pgp-signature, inline)]

Message #12 received at 361853@bugs.debian.org (full text, mbox, reply):

From: Ondrej Sury <ondrej@sury.org>

To: 361853@bugs.debian.org

Subject: Re: [php-maint] Bug#361853: phpinfo() Cross Site Scripting PHP 5.1.2 and 4.4.2

Date: Tue, 11 Apr 2006 09:58:36 +0200

[Message part 1 (text/plain, inline)]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

[phpinfo() Cross Site Scripting PHP 5.1.2 and 4.4.2]

Author: Maksymilian Arciemowicz (cXIb8O3)
Date:
- -Written: 26.2.2006
- -Public: 8.4.2006
from SecurityReason.Com
CVE-2006-0996

- --- 0.Description ---
PHP is an HTML-embedded scripting language. Much of its syntax is
borrowed from C, Java and
Perl with a couple of unique PHP-specific features thrown in. The goal
of the language is to
allow web developers to write dynamically generated pages quickly.

A nice introduction to PHP by Stig Sæther Bakken can be found at
http://www.zend.com/zend/art/intro.php on the Zend website. Also, much
of the PHP Conference
Material is freely available. 

- --- 1. Cross Site Scripting ---
In phpinfo() you can see all Varibles like:

file: standard/info.c
- -630-636---
php_print_gpcse_array("_REQUEST", sizeof("_REQUEST")-1 TSRMLS_CC);
php_print_gpcse_array("_GET", sizeof("_GET")-1 TSRMLS_CC);
php_print_gpcse_array("_POST", sizeof("_POST")-1 TSRMLS_CC);
php_print_gpcse_array("_FILES", sizeof("_FILES")-1 TSRMLS_CC);
php_print_gpcse_array("_COOKIE", sizeof("_COOKIE")-1 TSRMLS_CC);
php_print_gpcse_array("_SERVER", sizeof("_SERVER")-1 TSRMLS_CC);
php_print_gpcse_array("_ENV", sizeof("_ENV")-1 TSRMLS_CC);
- -630-636---

Function php_print_gpcse_array() for any arrays check 4096b of varible. 

file: standard/info.c
- -135-154---
if (Z_TYPE_PP(tmp) == IS_ARRAY) {
zval *tmp3;
MAKE_STD_ZVAL(tmp3);
if (!sapi_module.phpinfo_as_text) {
PUTS("<pre>");
}
php_start_ob_buffer(NULL, 4096, 1 TSRMLS_CC);
zend_print_zval_r(*tmp, 0);
php_ob_get_buffer(tmp3 TSRMLS_CC);
php_end_ob_buffer(0, 0 TSRMLS_CC);

elem_esc = php_info_html_esc(Z_STRVAL_P(tmp3) TSRMLS_CC);
PUTS(elem_esc);
efree(elem_esc);
zval_ptr_dtor(&tmp3);

if (!sapi_module.phpinfo_as_text) {
PUTS("</pre>");
}
} else if (Z_TYPE_PP(tmp) != IS_STRING) {
- -135-154---

So if we create array longer like 4096, html tags don't be remove.

Exploit:
If in php script is function phpinfo() try create some varibles (array)
like

phpinfo.php?cx[][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][][]=[XSS
]

or 

phpinfo.php?cx[]=ccccc..~4096chars...ccc[XSS]

- --- 2. How to fix ---
CVS
http://cvs.php.net/viewcvs.cgi/php-src/NEWS

- --- 3. Greets ---

For: sp3x
and
p_e_a, pi3, eax ;]

- --- 4. Contact ---
Author: SecurityReason.Com [ Maksymilian Arciemowicz ( cXIb8O3 ) ]
Email: max [at] jestsuper [dot] pl or cxib [at] securityreason [dot] com
GPG: http://securityreason.com/key/Arciemowicz.Maksymilian.gpg
SecurityReason.Com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (FreeBSD)

iD8DBQFEOAIl3Ke13X/fTO4RAo4LAJ0fBxJWN64vWrDYJEuhGkqc/OC42QCbBxip
f35+6LHjuBoqP5D2JV84ufs=
=iz3m
-----END PGP SIGNATURE-----
-- 
Ondrej Sury <ondrej@sury.org>

[signature.asc (application/pgp-signature, inline)]

Message #21 received at 361853-done@bugs.debian.org (full text, mbox, reply):

From: Steve Langasek <vorlon@debian.org>

To: Oliver Paulus <oliver@code-project.org>, 361854-done@bugs.debian.org, 361855-done@bugs.debian.org, 361856-done@bugs.debian.org, 361853-done@bugs.debian.org

Subject: Re: Bug#361854: function *() php/apache Crash PHP 4.4.2 and 5.1.2

Date: Mon, 10 Apr 2006 22:01:02 -0700

[Message part 1 (text/plain, inline)]

Version: 4:4.4.2-1
severity 361854 important
severity 361855 important
severity 361856 important
thanks

On Mon, Apr 10, 2006 at 11:14:43PM +0200, Oliver Paulus wrote:
> for more informaton see:
> http://securityreason.com/achievement_securityalert/34

On Mon, Apr 10, 2006 at 11:14:43PM +0200, Oliver Paulus wrote:
> for more informaton see:
> http://securityreason.com/achievement_securityalert/35

On Mon, Apr 10, 2006 at 11:16:11PM +0200, Oliver Paulus wrote:
> for more informaton see:
> http://securityreason.com/achievement_securityalert/36

On Mon, Apr 10, 2006 at 11:17:52PM +0200, Oliver Paulus wrote:

> for more informaton see:
> http://securityreason.com/achievement_securityalert/37

It is my understanding that all of these bugs are fixed in the etch version
of php4; I'm accordingly marking them as closed.

In addition, except for the cross-site scripting bug, none of these appear
to warrant severity: grave.

-- 
Steve Langasek                   Give me a lever long enough and a Free OS
Debian Developer                   to set it on, and I can move the world.
vorlon@debian.org                                   http://www.debian.org/

[signature.asc (application/pgp-signature, inline)]

Message #36 received at 361853@bugs.debian.org (full text, mbox, reply):

From: David Sterratt <david.c.sterratt@ed.ac.uk>

To: 361853@bugs.debian.org

Subject: This bug may not to be an issue for php4-4.3.10

Date: Wed, 17 May 2006 15:54:24 +0100

There appears to have been some ambiguity with the vulnerability
announcements, with some sites saying PHP 5.1.2 and prior
PHP 4.4.2 and prior
(http://securityreason.com/achievement_securityalert/34) and others just
saying PHP 5.1.2 and 4.4.2
http://www.securityfocus.com/archive/1/archive/1/430449/100/0/threaded

However, in php_print_gpcse_array() in ext/standard/info.c in the debian
sarge package there is no reference to the 4096 byte buffer which seemed
to be the cause of the problem. 

David.

Message #41 received at 361853@bugs.debian.org (full text, mbox, reply):

From: Stefan Fritsch <sf@sfritsch.de>

To: 361853@bugs.debian.org, control@bugs.debian.org

Subject: not fixed in etch

Date: Mon, 14 Aug 2006 22:30:35 +0200

found 361853 4:4.4.2-1.1
thanks

according to secunia [1], this has been fixed in 4.4.3, not in 4.4.2

[1] http://secunia.com/advisories/19599

Message #48 received at 361853@bugs.debian.org (full text, mbox, reply):

From: Thijs Kinkhorst <thijs@debian.org>

To: 361853@bugs.debian.org

Cc: Stefan Fritsch <sf@sfritsch.de>

Subject: Re: not fixed in etch

Date: Tue, 15 Aug 2006 11:14:37 +0200

[Message part 1 (text/plain, inline)]

Hello Stefan,

> according to secunia [1], this has been fixed in 4.4.3, not in 4.4.2
> 
> [1] http://secunia.com/advisories/19599

I've verified that the bug is indeed marked as fixed in the 4.4.3
changelog of PHP.

However, phpinfo() is a debug tool. I don't know why you would want to
use it on a production system and inside a context where cookies contain
security relevant information at the same time. If you ask me, this is
'important' at most. Secunia labels it as "not critical".


Thijs

[signature.asc (application/pgp-signature, inline)]

Message #53 received at 361853@bugs.debian.org (full text, mbox, reply):

From: Stefan Fritsch <sf@sfritsch.de>

To: Thijs Kinkhorst <thijs@debian.org>, control@bugs.debian.org

Cc: 361853@bugs.debian.org

Subject: Re: not fixed in etch

Date: Wed, 16 Aug 2006 21:15:10 +0200

severity 361853 important
thanks

On Tuesday 15 August 2006 11:14, Thijs Kinkhorst wrote:
> However, phpinfo() is a debug tool. I don't know why you would want
> to use it on a production system and inside a context where cookies
> contain security relevant information at the same time. If you ask
> me, this is 'important' at most. Secunia labels it as "not
> critical".

You are right. For some reason, I thought is was already marked 
important by Steve's mail.

Cheers,
Stefan

Send a report that this bug log contains spam.