dovecot - Jul 2004 - weakness in hash salt generation

Dovecot has routines for password hashing; two of these, crypt_generate
and md5_generate, both use sizeof(salt_chars) when reducing a random
string to salt.

I became suspicious when I noted that all salts generated are composed
only of "./01".

Unfortunately salt_char is declared static const char * rather than 
static const char [], and so sizeof(salt_chars) is the size of a pointer.

Recommend:


diff -u -r1.8 password-scheme.c
--- password-scheme.c	30 May 2004 03:57:15 -0000	1.8
+++ password-scheme.c	23 Jul 2004 17:05:36 -0000
@@ -15,7 +15,7 @@
 #  include <openssl/sha.h>
 #endif
 
-static const char *salt_chars +static const char salt_chars[]  
"./0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
 
 static buffer_t *schemes_buf;
 

joshua.

-- 
Joshua Goodall                           "as modern as tomorrow
afternoon"
joshua at roughtrade.net                                       - FW109
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
URL:
<http://dovecot.org/pipermail/dovecot/attachments/20040724/17a6a715/attachment-0001.bin>

On 23.7.2004, at 20:11, Joshua Goodall wrote:
> Dovecot has routines for password hashing; two of these, crypt_generate
> and md5_generate, both use sizeof(salt_chars) when reducing a random
> string to salt.
So it seems. But how did you notice it? I don't think those functions 
are ever called by Dovecot itself? They're there just in case some day 
they would be useful..
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
URL:
<http://dovecot.org/pipermail/dovecot/attachments/20040723/e394f39d/attachment-0001.bin>