Hi Nikolay,
Just fixed this bug in SVN trunk r2996.
Thanks for such detail report!
BR,
nanang
On Tue, Nov 3, 2009 at 9:52 PM, Nikolay Popok <nikolay.popok at gmail.com> wrote:
> Hi guys. I used search, but didnt find anything about this.
> I'm using pjproject 1.4.5. Looks like after remote side makes hold/resume,
> local sequence number for rtp packets is reset. I've investigated the
> problem a little bit.
> In pjsua_media.c :
>
> static void stop_media_session(pjsua_call_id call_id)
> {
> ?...
>
> ??? if (pjmedia_session_get_stream_stat(call->session, 0, &stat)
> ??? ??? == PJ_SUCCESS)
> ??? {
> ??? ??? /* Save RTP timestamp & sequence, so when media session is
> ??? ???? * restarted, those values will be restored as the initial
> ??? ???? * RTP timestamp & sequence of the new media session. So in
> ??? ???? * the same call session, RTP timestamp and sequence are
> ??? ???? * guaranteed to be contigue.
> ??? ???? */
> ??? ??? call->rtp_tx_seq_ts_set = 1 | (1 << 1);
> ??? ??? call->rtp_tx_seq = stat.rtp_tx_last_seq;
> ??? ??? call->rtp_tx_ts = stat.rtp_tx_last_ts;
> ??? }
>
>
> If remote side makes hold, this code is called twice. First time everyting
> is ok, but second time stat.rtp_tx_last_seq == 0. So after resume next
> sequence of rtp packet will be 0 too.
>
>
> _______________________________________________
> Visit our blog: http://blog.pjsip.org
>
> pjsip mailing list
> pjsip at lists.pjsip.org
> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
>
>
