bugtraq - Re: [ECHO_ADV_34$2006] W-Agora (Web-Agora) <= 4.2.0 (inc

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <200606270201.k5R211Rn019996@faron.mitre.org>
Date: Mon, 26 Jun 2006 22:01:01 -0400 (EDT)
From: "Steven M. Christey" <coley@...re.org>
To: bugtraq@...urityfocus.com
Subject: Re: [ECHO_ADV_34$2006] W-Agora (Web-Agora) <= 4.2.0 (inc_dir) Remote File Inclusion

>Successful exploitation requires that "register_globals= Off ".

That seems very strange, doesn't it?

Especially if you look at the source code.

Let's start with search.php, one of the vulnerable vectors:

  <?php

  ...

  require ("init.inc");

and in init.inc:

  require ("globals.inc");

  ...

  require ("$inc_dir/misc_func.$ext");

and in globals.inc:

  if (defined("_GLOBALS")) return;
  define('_GLOBALS', 1);

  ...

  $inc_dir    = 'include';

So - how could an attacker POSSIBLY modify the $inc_dir variable AFTER
it's set to a static value and when register_globals is off?

The answer is dynamic variable evaluation, which I posted about a
couple months ago at:

  http://www.securityfocus.com/archive/1/432828

If we look more closely at globals.inc, later on, we see this:

  if (!get_cfg_var("register_globals") ) {
      include "include/register_globals.$ext";
  }

and then if we look at register_globals.php, we see code like this:

	if (isset($HTTP_GET_VARS)) {
		reset($HTTP_GET_VARS);
		while ( list($var, $val) = each($HTTP_GET_VARS) ) {
			$$var=$val;
		}
	}

as well as for $HTTP_GET_VARS, $HTTP_POST_VARS, $HTTP_POST_FILES,
$HTTP_COOKIE_VARS.

This highlights the insidious nature of dynamic evaluation.

So - for those of us who have been questioning vulnerability reports
because we were seeing static values being set in include files - we
need to see if dynamic variable evaluation happens *after* the
variable is set, but *before* it's used.

Nasty, nasty stuff...

- Steve