Re: Fix for VLV draft.

["Jim Sermersheim" <jimse@novell.com>]

>From my read of draft-armijo-ldap-control-error-00.txt, the proposal is to ammend RFC 2251 in a way that cannot be done. I may be misreading it. 

Maybe it intends to say that this result code MUST only be used when the operation is accompanied by a control that specifies its use. If this is the case, I think it should be re-worded, and progressed so that other I-Ds needing to consume it can be progressed.

Otherwise, I don't mind introducing a new error code in the VLV draft itself that is only used in the presence of the VLV control. 

Jim

>>> "Kurt D. Zeilenga" <Kurt@OpenLDAP.org> 1/29/01 10:49:32 AM >>>
I think it's a bad idea to overload the semantics of the 'other'
(or any other existing) result code.  'other' should only imply
some implementation specific error.   IIRC, Michael previously
suggested introduction a new result code indicating that an error
occurred during control processing.  I believe we should pursue
this approach.

Kurt

At 10:16 AM 1/29/01 -0700, Jim Sermersheim wrote:
>Back in September, as the Virtual List View draft was going through standardization process, Michael Armijo noted the following problem:
> 
>>In section 3.2 we have details about what the virtualListViewResult
>>codes mean, but we still do not define what result code should be
>>returned in the actual SearchResult in the case where there is a value
>>other then success(0) in the virtualListViewResult.
>>
>>We should return something other then success(0)in the SearchResultDone
>>if there is a result code other then success(0) in the
>>virtualListViewResult and what that result code is should be defined (or
>>referenced) in the VLV draft.
> 
>I propose that we fix this by inserting the following statement as paragraph 4 of Section 3.2
>If the LDAP SearchResultDone message has a resultCode of other (80), 
>the virtualListViewResponse MAY be included and MAY hold a non-zero 
>value in the virtualListViewResult field.
>If there are no objections, I'd like to add this and allow the draft to progress.
> 
>Jim