Re: userPassword question

[Mark Smith <mcs@netscape.com>]

Chris Newman wrote:
> 
> On Wed, 29 Jul 1998, Mark Smith wrote:
> > database.  In fact, one of the deficiencies of CRAM-MD5 is that as far
> > as I can tell it requires that the LDAP server store the password in the
> > clear (or in a way that allows it to easily recover the clear text
> > password).  But that's a different topic.
> 
> Not true.  You can store the intermediate MD5 state from the two MD5
> operations in the HMAC.  It makes CRAM-MD5 marginally better than
> APOP, CHAP and their ilk although it's still plaintext equivalent.

Makes sense.  Thanks for the clarification.

> ...
> > c) On bind operations, we compare the submitted value (which must be a
> > clear text password) against all userPassword values -- hashed or
> > unhashed -- in an entry.
> 
> So what happens if the client supplies a simple bind password of
> "{crypt}CRYPTED-VALUE"?

It would be rejected.  The actual "compare" algorithm we use for
userPassword values looks something like:

    foreach userPassword value V present in an entry
        1. determine the encoding used in V, e.g., crypt
        2. apply that encoding to the value presented in the bind
operation
        3. compare V to the result of step 2. and return SUCCESS if they
match
    return FAILURE

In your example, if the value {crypt}CRYPTED-VALUE was ALSO present in
the entry that is the target of the bind operation, we would apply the
UNIX crypt algorithm a second time to the password supplied by the
client.  Bottom line: clients should always present passwords in clear
text.

-Mark