Re: [sipcore] Draft new version: keep

["Rifaat Shekh-Yusef" <[e-mail]>]

>  
>>I read your document (sipcore-keep-01) and I have the following
questions (I am sorry if these questions were asked before):
>> 
>>RFC 5626 recommends the STUN Keep-Alive technique with connection-less
transports and CRLF Keep-Alive technique with connection-oriented
transports.
>>Is this document recommending the client should use STUN Keep-Alive
technique for both connection-oriented and connection-less transports?
>
>No.
>
>The idea is to use the keep-alive mechanism defined in RFC 5626.
>
>The purpose of the draft is only to allow entities to negotiate the
usage of keep-alive, if they don't support outbound.

Then section 5 should be clarified to reflect this, as this is not clear
from the current text.


>>Why did you choose STUN over CRLF?
>>
>>The header of Section 8 implies that there is an overlap between this 
>>document and the connect-reuse draft, but when you read the section
you realize that this is not the case.
>>Why do you think that this section is needed? Does it matter if the 
>>connection between the UA and the proxy uses one TCP connection or two
TCP connections?
>
>At one point I was asked to add that section, just to point out the
differences between keep-alive and connect-reuse.

This section does not seem to add any value to this document.


Regards,
 Rifaat

 


________________________________

From: [e-mail] [mailto:[e-mail]] On
Behalf Of Christer Holmberg
Sent: Tuesday, December 01, 2009 6:18 AM
To: [e-mail]
Subject: [sipcore] Draft new version: keep





Hi, 

I've submitted a new version of the keep draft. There are no changes
from the previous version, but it had expired. 

I will produce a new version, with new content, once we get the info
issues resolved - it has really taken more time than expected... 

A new version of the info draft will be submitted tomorrow. 

Regards, 

Christer