Re: Admission field in QSPECRE: Review of draft-ietf-nsis-qspec-18.txt

On 25 Feb 2008, at 23:49, TARAPORE, PERCY S, ATTLABS wrote:

> Ken,
>
> You state that Francois has "one way" that supposedly defines a  
> "common structure to hold the values and the ability to set the  
> values accordingly" as John desires.

As I said before, we are recommending that the Admission Priority  
value be handled in a similar way to how Preemption Priority is  
already handled today in NSIS and RSVP: it is up to the network  
administrator to decide what values should be used for what. If the  
network administrator wants to use values as defined in Y.2171, this  
is all fine and good. But if the network administrator decides to use  
different values, this is also fine too.

Francois


> I am somewhat at a loss here - what exactly has Francois described  
> that takes an incoming CAC priority value as desired by a customer  
> at the initial domain which then passes it on to the other domains  
> in the path. After all, Y.2171 is simply a vehicle for providing a  
> CAC priority value. It does not mandate how that value can be acted  
> upon in each domain.
>
> Thanks
>
> Percy
>
> From: ken carlberg [mailto:[email protected]]
> Sent: Monday, February 25, 2008 3:28 PM
> To: [email protected]
> Cc: [email protected]; [email protected]; TARAPORE, PERCY S, ATTLABS;  
> [email protected]
> Subject: Re: Admission field in QSPECRE: [NSIS] Review of draft- 
> ietf-nsis-qspec-18.txt
>
> John,
>
>> I think wee need to be able to support a few different scenarios  
>> and usages for the admission priority.  Is there a way that we  
>> could have a  common structure to hold the values, and the ability  
>> to set the values accordingly?
>
> we have one way, as described by Francois.
>
> however, if there is a segment that is absolutely adamant in having  
> a field reflect the defined values of Y.1541 or Y.2171bis in the  
> QSPEC, then I think a new field should be defined in that document  
> to contain those values.  The authors of the other documents can  
> decide if they too wish to add an ITU dependent field.
>
> the other option is to disregard any relation between the Admission  
> Priority field of the QSPEC and tsvwg-emergency-rsvp, which would  
> be a shame.
>
> -ken
>
>
>
>

_______________________________________________
nsis mailing list
[email protected]
http://www.ietf.org/mailman/listinfo/nsis