[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: general aliasing



Robert Streich wrote:
> 
> > That raises a good question, should the deref return a candidate that
> > contains a DN that is not within the requested base and fails the
> > original base and scope checks.  My original instinct was no, the
> > required base and scope should be adhered to or the client may be
> > confused, but perhaps I am wrong.  Anyone else have comments on this?
> 
> X.518(93) Section 19.3.2.2.3 says yes.

fair enough.  I will put the change in as time permits.

> But you can have entirely different subtrees in the database. Why would
> they not scope it to one of those? The only thing that makes it a
> single tree is the null node. Under that you could have a set of
> subtrees that are completely disparate.

Well, for a given search you have to specify a base that is a
subordinate of the suffix for the back-end or the back-end cannot be
determined.  The suffix is the null node for that back-end, so as long
as a back-end is selected the subtree scope includes everything within
it.

This is another point that will get you into trouble if the alias points
to a base that is superior to the specified search base.  The back-end
could switch on you and the current search method is geared to a
determined back-end. 

-- 
Will Ballantyne     GEMS Technical Architect
mailto:Will.Ballantyne@gems1.gov.bc.ca