[INFOD-WG] Faults

[Ronny Fehling]

you are right.
We are treating the responses inconsistent.
Even though you might not leave a call-back ERP, I assume that the 
response handling should be part of the request-response message no 
matter what the protocol is.
So I think we should put MUST in every response.

Comments?
R

Fisher, SM (Steve) wrote:
> Ronny,
>
> But as we have it the registry does not even have to try to explain why
> it has ignored a message!
>
> Are you also saying the mechanism to return a fault is quite different
> from that to return a success message to the extent that it may not be
> possible to send a faulyt but always possible to send a success message?
>
>
> Steve
>
>   
>> -----Original Message-----
>> From: Ronny Fehling [mailto:ronny.fehling at oracle.com] 
>> Sent: 02 April 2007 15:52
>> To: Fisher, SM (Steve)
>> Cc: infod-wg at ggf.org
>> Subject: Re: [INFOD-WG] Faults
>>
>> I think we had put this in because we do not require the 
>> calling service to have a call-back number
>>
>> Fisher, SM (Steve) wrote:
>>     
>>> Chapter 2 says in each response section:
>>>
>>> "If the INFOD registry accepts ... MUST ...
>>> Instead ... MAY send the following faults ..."
>>>
>>>
>>> Don't we want to say:
>>>
>>> "If the INFOD registry accepts ... MUST ...
>>> otherwise ... MUST send one of the following faults ..."
>>>
>>>
>>> Of course we cannot guarantee succesful transmission of the fault - 
>>> but nor can we be sure that the happy response wil get though.
>>>
>>> Steve
>>> --
>>>   infod-wg mailing list
>>>   infod-wg at ogf.org
>>>   http://www.ogf.org/mailman/listinfo/infod-wg
>>>   
>>>