Re: local naming vs directory naming
Date: Fri, 09 May 2008 08:55:23 -0500
I believe that both are correct, though not necessarily DB logins for #2. The first introduction of OID was done as a product in the middleware security area to provide an LDAPv3 compliant server that could be used for many purposes. I think the first integrations with other Oracle products were
1) TNS name resolution and
2) a Portal (3.0.x timeframe) user authentication alternative.
At the time, Portal users were authenticated using a table in the Portal database. Naturally, many customers wanted to integrate with other logins, like Windows, or perhaps another enterprise directory server. OID was the first step towards that by pulling the identities into a standards-compliant LDAP directory.
A database release or two after OID was introduced, Enterprise User Security arrived. At first, it only supported certificate authentication (v8 timeframe). Starting with 9i, it added password-based enterprise users and starting with 10g, they refactored the licenses to allow password-based EUS to be included as part of DB EE licenses (previously it was part of the Advanced Security Option).
Probably more than you wanted to know. I've proposed a session on EUS, OID and AD integration for OOW. If you haven't started voting yet, signup or login to Mix and vote for it! https://mix.oracle.com/ideas/27775-integrating-oid-ad-with-database-logins-enterprise-user-security
Dennis Williams wrote:
> My understanding was that originally there were 2 main purposes for OID.
> 1) Database lookup.
> 2) User logins. I thought the idea was for users to be able to log in
> and be authenticated with OID. Am I misremembering in my old age?
> Obviously too lazy to do the research :-)
> Dennis Williams