[PATCH] off by one error in nis-publickey

[PATCH] off by one error in nis-publickey

[Thorsten Kukuk]

Hi,

there is a off by one error in nis-publickey:

2005-10-28  Thorsten Kukuk  <kukuk@suse.de>

	* nis/nss_nis/nis-publickey.c(_nss_nis_getpublickey): fix off
	by one error.

--- nis/nss_nis/nis-publickey.c	26 Aug 2002 06:20:04 -0000	1.13
+++ nis/nss_nis/nis-publickey.c	28 Oct 2005 12:53:02 -0000
@@ -120,7 +120,7 @@
 
       ++p;
       strncpy (buf, p, 2 * (HEXKEYBYTES + 1));
-      buf[2 * (HEXKEYBYTES + 1)] = '\0';
+      buf[2 * HEXKEYBYTES + 1] = '\0';
       if (!xdecrypt (buf, passwd))
 	return NSS_STATUS_SUCCESS;
 

-- 
Thorsten Kukuk         http://www.suse.de/~kukuk/      kukuk@suse.de
SUSE LINUX Products GmbH       Maxfeldstr. 5       D-90409 Nuernberg
--------------------------------------------------------------------    
Key fingerprint = A368 676B 5E1B 3E46 CFCE  2D97 F8FD 4E23 56C6 FB4B

Re: [PATCH] off by one error in nis-publickey

[Ulrich Drepper]

[-- Attachment #1: Type: text/plain, Size: 515 bytes --]

Thorsten Kukuk wrote:
> there is a off by one error in nis-publickey:

While this is correct for the current definition of the array the
question is why is the array and the strncpy not defined as 2 *
HEXKEYBYTES + 1.

The memcmp starts at HEXKEYBYTES and continues for KEYCHECKSUMSIZE.  In
fact, the array should be HEXKEYBYTES + KEYCHECKSUMSIZE in size (no NUL
byte needed).  So, why doesn't it do it like this?

-- 
➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 251 bytes --]

Re: [PATCH] off by one error in nis-publickey

[Thorsten Kukuk]

On Fri, Oct 28, Ulrich Drepper wrote:

> Thorsten Kukuk wrote:
> > there is a off by one error in nis-publickey:
> 
> While this is correct for the current definition of the array the
> question is why is the array and the strncpy not defined as 2 *
> HEXKEYBYTES + 1.
> 
> The memcmp starts at HEXKEYBYTES and continues for KEYCHECKSUMSIZE.  In
> fact, the array should be HEXKEYBYTES + KEYCHECKSUMSIZE in size (no NUL
> byte needed).  So, why doesn't it do it like this?

Because you ignores the data you get from the entry, the xdecrypt() 
call and the fact, that the NIS definition is very unclear about the
NUL byte and different implementations handle that different.

  Thorsten

-- 
Thorsten Kukuk         http://www.suse.de/~kukuk/      kukuk@suse.de
SUSE LINUX Products GmbH       Maxfeldstr. 5       D-90409 Nuernberg
--------------------------------------------------------------------    
Key fingerprint = A368 676B 5E1B 3E46 CFCE  2D97 F8FD 4E23 56C6 FB4B