Re: [Sip] Possible bug in "Non-INVITE Client Transaction" - 17.1.2.2

[Peter Hanratty <[e-mail]>]

While youre at it, the INVITE server transaction could use some 
rethinking.  Terminating on a 2xx but on no other final response is a 
bit daft and makes people like me want to cut corners rather than use 
other objects such as the dialog or session to handle the ACK, which 
doesn't really make for a nice logical model :)  The ACK is attached to 
a single transaction and should handled by that transaction (as it is 
for a non-2xx final response).


Peter Hanratty
Senior Software Developer

Pinnacle Telecom Group PLC
Brooke House
4 The Lakes
Bedford Road
Northampton
NN4 7YD

Tel:  [phone]
Fax:  [phone]


On 08/04/2011 08:56, Iñaki Baz Castillo wrote:
> 2011/4/8 Dean Willis<[e-mail]>:
>    
>> I think you're right that the spec is written incorrectly. I believe it should describe that the multiplier on T1 doubles with each reset. This is not clear in the existing text. So for example, if 8*T1<  T2, then the third reset is 8*T1, and if 16*T1<  T2, then the 4th reset is 16*T2
>>
>> Otherwise said, MIN(2^N*T1,T2) where N is the repetition iterator.
>>      
> That would clarify it, right.
>
>
>    
>>> PS: Sorry for the cross-posting, I don't know which maillist is better
>>> to report it.
>>>        
>> [e-mail] would probably be the right place.
>>      
> Finally I've reported an errata for RFC 3261 (hope it's also a good place):
>
>    http://www.rfc-editor.org/errata_search.php?rfc=3261
>
>
> Thanks a lot.
>
>    
>
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 9.0.894 / Virus Database: 271.1.1/3545 - Release Date: 04/01/11 19:36:00
>
>