< Previous by Date Date Index Next by Date >
< Previous in Thread Thread Index Next in Thread >

Re: [reSIProcate] re-post: hangup before receiving any response


Hi,

You should check out the latest revision since there is a fix for that
(actually, the ACK case always execute break with or without a To tag).
This was discussed few mails ago.

Regards,
Dom.

----- Original Message ----- 
From: "kaiduan xie" <kaiduanx@xxxxxxxx>
To: <resiprocate-devel@xxxxxxxxxxxxxxxxxxx>
Sent: Friday, September 17, 2004 1:16 AM
Subject: [reSIProcate] re-post: hangup before receiving any response


> Hi, all,
>
> This is a re-post of my last email for the case where
> user hangup before receiving ANY response.
>
> In transaction layer, after sending INVITE for 7
> times, transaction layer will post a 408 to TU. In
> DialogSet::dispatch() (#357)
>
>    switch (response.header(h_CSeq).method())
>    {
>       case INVITE:
>       case SUBSCRIBE:
>       case BYE:
>       case ACK:
>           if (response.header(h_To).exists(p_tag))
>           {
>              break; // dialog creating/handled by
> dialog
>           }
>           else
>           {
>              // throw away, informational stauts
> message eventually
>              return;
>           }
>
> Since this 408 has no tag in To, the message is throw
> away. So, we still need to put guard.destroy() in
> DialogSet::cancel().
>
> DialogSet::cancel()
> {
>     if (mDialogs.empty())
>     {
>        if (mReceivedProvisional && getCreator())
>        {
>            ....
>
>        }
>        else
>        {
>             //xkd to handle the case where user hangup
> before receiving ANY response, including 100
>            guard.destroy();
>            return;
>        }
>     }
> }
>
> The scenario did happen while testing with Free World
> Dialup.
>
> kaiduan
>
>
>
>
> ______________________________________________________________________
> Post your free ad now! http://personals.yahoo.ca
> _______________________________________________
> resiprocate-devel mailing list
> resiprocate-devel@xxxxxxxxxxxxxxxxxxx
> https://list.sipfoundry.org/mailman/listinfo/resiprocate-devel
>