Re: [provision] RequestType should require requestID.

[Gary P Cole <Gary.P.Cole@Sun.COM>]

I forgot to mention another important reason to require requestID: the 
"Hal Hole".  If the provider executes a requested operation 
asynchronously (either because the requestor said so or because the 
requestor let the provider decide so), then the provider must return a 
requestID that the requestor can use to cancel or to check the status of 
the operation.

If the original request contains a requestID, then the requestor knows 
the requestID (even if the requestor for some reason does not receive 
the provider's response).  If the original request lacks a requestID and 
the requestor for some reason does not receive the provider's response, 
then the requestor must use the 'statusAll' operation (i.e., 'status' 
with no argument) to recover.  This hurts.

I think it would be much simpler (for the requestor, for the provider, 
and for anyone reading the specification) to require requestID in each 
request.

Gary P Cole wrote:

> We talked about making requestID mandatory at the most recent 
> Face-to-Face, but no one (who was present at the time) defended this 
> vigorously. As a result, we decided to make things simple on the 
> requestor and allow requestID to be optional.  I think we should 
> revisit this.
>
> If each request contains a (reasonably unique) requestID, then a 
> provider is free to process requests in any order that is efficient 
> (or otherwise advantageous).
>
> If this is NOT the case (and some requests lack requestID), then a 
> provider must queue requests for that requestor.  The provider must 
> respond to each request from a specific requestor before it responds 
> to another request from that some requestor.  If the provider responds 
> in an order that does not match the order of requests, the requestor 
> may be unable to match the response with the correct request.
>
> We could address this by:
> 1) Requiring requestID in RequestType.
> 2) Normatively specifying that that a requestor that wants to send 
> another request without waiting for a response to the first must 
> specify requestID in each request.
>
> Since a requestID value needs to be only reasonably unique (e.g., a 
> monotonically increasing integer that ultimately rolls over), I favor 
> option 1.
>
> Does anyone think that requiring requestID is too hard on a simple 
> requestor?  If not, can we just go ahead and make it required?
>
> Gary
>
>
> To unsubscribe from this mailing list (and be removed from the roster 
> of the OASIS TC), go to 
> http://www.oasis-open.org/apps/org/workgroup/provision/members/leave_workgroup.php. 
>
>