RE: LCUP Issue: UUID

["Jim Sermersheim" <jimse@xxxxxxxxxx>]

I agree. if the attribute is called entryUUID it should be unique within
the universe of the global DIT. Otherwise, call it an lcupUID.

Jim

>>> "Kurt D. Zeilenga" <Kurt@xxxxxxxxxxxx> 09/04/02 01:54PM >>>

At 12:25 PM 2002-09-04, Liben, Michael (GTS) wrote:
>The UUID only needs to be unique in the DIT in which it has meaning
(the context).

LDAP and X.500 are designed to support a global DIT.

>As long as the server(s) holding the DIT create the UUID using a
consistent algorithm (server-assigned per the spec),
>we shouldn't care which algorithm is employed.

Consider a server which masters one context and shadows a subordinate
context.  UUIDs in the spanning LCUP context must be unique.  This
requires that both servers (this one and the own who masters the
shadowed context) need to agree on the algorithm to be employed.
To ensure interoperability, a standard algorithm should be implemented
by both.  A REQUIRED algorithm is needed.

In my opinion, UUIDs in LDAP should conform to ISO/IEC 11578:1996.

>Combined with a server-assigned LCUP cookie, the opportunity of
collision is academic at best. 

If entryUUID were only to be used by LCUP, I might agree
with you.  But I believe the desire is to define entryUUID
so that they have broader applicability.  (Which is one of
the reasons why I believe entryUUID and entryCSN should be
detailed in a separate document.)

Kurt