[prev in list] [next in list] [prev in thread] [next in thread] 

List:       full-disclosure
Subject:    Re: [Full-disclosure] ProFTPD mod_tls pre-authentication buffer
From:       Mark Wadham <mark.wadham () areti ! net>
Date:       2006-11-29 10:39:23
Message-ID: 456D635B.9000805 () areti ! net
[Download RAW message or body]

Hi

Can i get this straight, vendor was notified on the 16th of November, 
but this vulnerability has been part of VulnDisco since Jan 2006?  Is 
that actually correct?  This was known about ten months ago but not 
disclosed until now?

Mark

research@gleg.net wrote:
> Name:         ProFTPD mod_tls pre-authentication buffer overflow
> Vendor:       http://www.proftpd.org
> Release date: 28 Nov, 2006
> Author:       Evgeny Legerov <research@gleg.net>
>
> I. DESCRIPTION
>
> A remote buffer overflow vulnerability has been found in mod_tls module of
> ProFTPD server.
> The vulnerability could allow a remote un-authenticated attacker to gain root
> privileges.
>
> II. DETAILS
>
> Let's have a look at the code (ProFTPD version 1.3.0):
>
> contrib/mod_tls.c:
> """
> static char *tls_x509_name_oneline(X509_NAME *x509_name) {
>   static char buf[256] = {'\0'};
>
>   /* If we are using OpenSSL 0.9.6 or newer, we want to use
>  * X509_NAME_print_ex()
>    * instead of X509_NAME_oneline().
>    */
>
> #if OPENSSL_VERSION_NUMBER < 0x000906000L
>   memset(&buf, '\0', sizeof(buf));
>   return X509_NAME_oneline(x509_name, buf, sizeof(buf));
> #else
>
>   /* Sigh...do it the hard way. */
>   BIO *mem = BIO_new(BIO_s_mem());
>   char *data = NULL;
>   long datalen = 0;
>   int ok;
>
>   if ((ok = X509_NAME_print_ex(mem, x509_name, 0, XN_FLAG_ONELINE)))
> [1]  datalen = BIO_get_mem_data(mem, &data);
>
>   if (data) {
>     memset(&buf, '\0', sizeof(buf));
> [2] memcpy(buf, data, datalen);
>     buf[datalen] = '\0';
>     buf[sizeof(buf)-1] = '\0';
>
>     BIO_free(mem);
>     return buf;
>   }
>
>   BIO_free(mem);
>   return NULL;
> #endif /* OPENSSL_VERSION_NUMBER >= 0x000906000 */
> }
> """
>
> The value of 'datalen' parameter is fully controlled by us (see [1]).
> On line [2] we will be able to overflow the 'buf' buffer with our data.
>
> III. VENDOR RESPONSE
>
> Vendor has been notified on Nov 16, 2006 but ProFTPD 1.3.0a is still vulnerable.
>
> IV. CREDIT
>
> Discovered by Evgeny Legerov.
>
> The vulnerability is a part of VulnDisco Pack Professional since Jan, 2006.
>
>
>
>
>   


-- 
Mark Wadham
e: mark.wadham@areti.net t: +44 (0)20 8315 5800 f: +44 (0)20 8315 5801
Areti Internet Ltd., http://www.areti.net/ 

===================================================================
Areti Internet Ltd: BS EN ISO 9001:2000
Providing corporate Internet solutions for more than 10 years.
===================================================================

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/
[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic