[sork] Passwd H3 Beta

[Eric Rostetter]

Quoting Julez <blades1 at ix.netcom.com>:

> Good evening, all.  I've got the Horde 3 framework installed, and I'm
> running IMP H3 (4.0.3).  I'm using Courier-IMAP, Qmail, and Vpopmail on the
> backend.  I'm trying to get Passwd H3 working, but whenever I try to change
> a virtual user's password, I get "Incorrect Password" returned.  From what I
> can tell it's coming from my passwd/lib/Driver.php, where it tries to
> compare my "oldPassword" with the one it pulls from the database.

This sounds correct.

> I'm
> guessing that it has to do with the encryption scheme, but I'm not sure.

I would guess so also.

> I'm not absolutely sure, but I think I'm using Crypt-md5 for Vpopmail (is
> that what Vpopmail uses by default?).

No idea.  I'd try all the different md5-type drivers and see if any of them
work.

> I know I have PHP compiled with
> Crypt_md5.  I've tried various encryption options in my backends.php file,
> but nothing works.

I'd try them all, and insert a debugging statement in lib/Driver.php for
each case in the comparePasswords function which prints out the encrypted
and plaintext passwords, as well as the results of any functions it calls
there (e.g. md5(), base64_encode(mHas()), etc.  Chances are from that you
will at least be able to tell if something is really wrong, and if not 
what the correct format is likely to be, etc.

> Is there more information I can provide that would be helpful?  Any patches
> yet for Passwd H3?  Any ideas what I can do to test what the password gets
> encrypted as from the website?

I'd say insert debugging statements in the case as explained above, and
examine that output.  You should be able to see if the formats match
(base64 is rather distinct, as is md5, etc. so you should be able to
determine what formats to exclude or consider doing this).
 
> Thanks in advance, and if I need to provide more info, let me know.

Can't help much as I've never used vpopmail.  I think debugging via
echoing variables is about the best I can recommend.
 
> /Julian

-- 
Eric Rostetter
The Department of Physics
The University of Texas at Austin
 
Why get even? Get odd!