[asterisk-dev] [Code Review] SIP: for an INVITE transaction, treat all 2XX responses the same as a 200

Olle E. Johansson oej at edvina.net
Thu Sep 23 05:53:41 CDT 2010


23 sep 2010 kl. 12.28 skrev sst at sil.at:

> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/935/#review2773
> -----------------------------------------------------------
> 
> 
> 
> /trunk/channels/chan_sip.c
> <https://reviewboard.asterisk.org/r/935/#comment5970>
> 
>    i am not sure but AFAIK there are only two 2xx responses. 200 Ok and 202 Accept.
>    A 202 in a Refer dialog is handled on Row 19727.
> 
>    again AFAIK there could only be a 202 for notifys. So maybe we here it should be handle_response_notify and not invite?
> 
> 
> 
> 
> /trunk/channels/chan_sip.c
> <https://reviewboard.asterisk.org/r/935/#comment5971>
> 
>    same here.
> 
> 
> - schmidts
> 
> 
> On 2010-09-22 17:03:18, David Vossel wrote:
>> 
>> -----------------------------------------------------------
>> This is an automatically generated e-mail. To reply, visit:
>> https://reviewboard.asterisk.org/r/935/
>> -----------------------------------------------------------
>> 
>> (Updated 2010-09-22 17:03:18)
>> 
>> 
>> Review request for Asterisk Developers.
>> 
>> 
>> Summary
>> -------
>> 
>> Currently if Asterisk receives any 2XX response to an outgoing INVITE other than a 200 it is ignored.  I have changed this behavior to treat all 2XX responses the same as a 200.  RFC3261 section 13.2.2.4 which outlines how a 2XX response should be handled in response to an INVITE request seems to treat all 2XX responses the same.  I don't see any reason why our implementation should be specific to only 200 Ok responses. 
>> 
>> 
For an INVITE, I have never seen any other response. As Stefan says, for REFERs we already handle the valid alternative.
What case are you considering here, David?
A 202 Accept to an INVITE is not a 200 OK, really.
/O




More information about the asterisk-dev mailing list