Re: lower_pass = after problems

2004-04-04 Thread Alan DeKok
Mike Lampson [EMAIL PROTECTED] wrote:
  Both.  The after thing runs the packet through the server twice,
  which is problematic.  The lower_pass thing can be done in a module.
 
 I would urge you *not* to do this.  We SHA1 encrypt our lowercased,
 MySQL-stored passwords.  We then lowercase the incoming password from the
 authentication request before the PAP module SHA1-encodes it for comparison.
 I would rather not process the incoming authentication request through an
 additional module prior to performing the authentication.

  Whether it's done in the server core or another module is irrelevant
to CPU time and memory.

  The problem is that the server core shouldn't have atrocious hacks
like this in it.  That functionality already exists in the module
infrastructure, in a more stable and more configurable form.

  Alan DeKok.

- 
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html


Re: lower_pass = after problems

2004-03-26 Thread Federico Giannici
Alan DeKok wrote:
Federico Giannici [EMAIL PROTECTED] wrote:

I have noticed that the lower_pass = after configuration command is 
implemented simply executing a second time the entire sequence of 
authorization/authentication operations.


  Yes.  The feature is a hack, and should be removed from the server.

  Similarly, the lower_user feature should also be deleted.
Hummm...
Do you want to remove only the after option (the real hack) or the 
entire command?

I'd like to know this so, in the latter case, I'll have to implement 
this functionality in our custom module I'm writing...

Thanks.

--
___
__
   |-  [EMAIL PROTECTED]
   |ederico Giannici  http://www.neomedia.it
___
- 
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html


Re: lower_pass = after problems

2004-03-22 Thread Alan DeKok
Federico Giannici [EMAIL PROTECTED] wrote:
 I have noticed that the lower_pass = after configuration command is 
 implemented simply executing a second time the entire sequence of 
 authorization/authentication operations.

  Yes.  The feature is a hack, and should be removed from the server.

  Similarly, the lower_user feature should also be deleted.

  Alan DeKok.

- 
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html