Re: [Tsvwg] I-D Action:draft-ietf-tsvwg-rsvp-user-error-spec-05.txt

["Adrian Farrel" <[e-mail]>]

Hi Francois,

Yes, that works and achieves backward compatibility.

Thanks,
Adrian
----- Original Message ----- 
From: "Francois Le Faucheur IMAP" <[e-mail]>
To: "Adrian Farrel" <[e-mail]>
Cc: "Francois Le Faucheur IMAP" <[e-mail]>; "tsvwg IETF list" 
<[e-mail]>
Sent: Tuesday, April 01, 2008 7:25 PM
Subject: Re: [Tsvwg] I-D Action:draft-ietf-tsvwg-rsvp-user-error-spec-05.txt


>
> On 1 Apr 2008, at 18:38, Adrian Farrel wrote:
>
>> Well, what do you think?
>> We intended to allow error values to be defined in the future if  someone
>> comes up with a reason.
>
> how about you do define one Error Value in this draft (eg "0  = all  error 
> information is encoded inside USER_ERROR_SPEC") ?
> This way:
> - implementations will know that they must set value to 0 (until  they 
> have something more meaningful to put inside Error Value)
> - implementation will know that when they receive 0, it means there  is no 
> meaningful info
> - the registry in IANA will show a single value for now (making it  clear 
> that it the value to set/expect for now) and other values can  be added 
> there later.
> I'd think this would make things clearer/cleaner than not specifying  any 
> value. You could argue it kinds of wastes one value but  considering this 
> is 16 bits field and there is no identified use yet,  that seems bearable.
>
> Makes sense?
>
> Francois
>
>>
>> The precedent in RFC 2205 is that only Error Values that are  defined are
>> mentioned. Error Codes that have no Error Values are defined with  no 
>> mention
>> of the Error Value. For example, Error Code 3, 4, or 6.
>>
>> We could curtail this explicitly in the draft if there is demand  from 
>> the
>> group.
>>
>> Cheers,
>> Adrian
>>
>> ----- Original Message ----- From: "Francois Le Faucheur IMAP" 
>> <[e-mail]>
>> To: "tsvwg IETF list" <[e-mail]>
>> Sent: Tuesday, April 01, 2008 5:14 PM
>> Subject: Re: [Tsvwg] I-D Action:draft-ietf-tsvwg-rsvp-user-error- 
>> spec-05.txt
>>
>>
>>> Hi,
>>>
>>> I think this document is ready to move forward.
>>>
>>> One small suggestion perhaps. It probably wouldn't hurt to be   explicit 
>>> about the "Error Value" field of the ERROR_SPEC (when  Error  Code = 
>>> "User Error Spec"). I understand this field is  basically not  to serve 
>>> any purpose ("No Error Values are  defined), suggesting a  "MUST be set 
>>> to zero on transmit and MUST  be ignored on receipt" .
>>>
>>> Francois
>>>
>>>
>>> ===============================================================
>>> Updated after discussion with Jukka.
>>> Adrian
>>>
>>> A New Internet-Draft is available from the on-line Internet-Drafts 
>>> directories.
>>> Title           : User-Defined Errors for RSVP
>>> Author(s)       : G. Swallow, A. Farrel
>>> Filename        : draft-ietf-tsvwg-rsvp-user-error-spec-05.txt
>>> Pages           : 8
>>> Date            : 2008-04-01
>>>
>>> The Resource ReserVation Protocol (RSVP) defines an ERROR_SPEC object
>>> for communicating errors.  That object has a defined format that
>>> permits the definition of 256 error codes.  As RSVP has been
>>> developed and extended, the convention has been to be conservative in
>>> defining new error codes.  Further, no provision for user-defined
>>> errors exists in RSVP.
>>>
>>> This document defines a USER_ERROR_SPEC to be used in addition to the
>>> ERROR_SPEC to carry additional user information related to errors.
>>>
>>> A URL for this Internet-Draft is:
>>> http://www.ietf.org/internet-drafts/draft-ietf-tsvwg-rsvp-user- error- 
>>> spec-05.txt
>>>
>>
>>
>
>