[pjsip] [SIPRTP] lot of percent loss for TX results
bennylp at pjsip.org
Mon May 19 05:27:13 EDT 2008
On Mon, May 19, 2008 at 8:18 AM, antoine duclaud
<antoine.duclaud at gmail.com> wrote:
> For high, I have for example a rtt up to 1000 ms and jitter up to 7ms...
> sometimes more... I think it's just a peak at the beginning but i'm not
Okay. But 7 ms is not bad IMO, considering that we're running siprtp
on general purpose OS and not a real-time one (we use
pj_thread_sleep() for scheduling packet). Also the RTCP will ignore
the first 25 RTP packets (25 is controllable via
PJMEDIA_RTCP_IGNORE_FIRST_PACKETS setting) on jitter calculation.
Not sure about the RTT value. But check if you have "clock skew
detected" message in the log. When this happens it will screw up the
RTT calculation. What OS are you running siprtp on? The clock skew
happens a lot on Windows, but if I recall correctly it doesn't happen
> I have some packets in reorder too for the RX stats.
Okay. It's good then at least siprtp is able to report them. :)
> 2008/5/16 Benny Prijono <bennylp at pjsip.org>:
>> On Fri, May 16, 2008 at 2:12 PM, antoine duclaud
>> <antoine.duclaud at gmail.com> wrote:
>> > Hello!
>> > I use siprtp for test a liaison between two sites, and I have see that
>> > for
>> > the TX results, I have a lot of percent loss, 99% at the beginning of
>> > the
>> > test...
>> > But the results of RTT and Jitter are high too. How the results of RTT
>> > and
>> > jitter can be calculated if there are all packet losts?
>> Jitter is only calculated when the sequence numbers of the subsequent
>> RTP packets are adjacent, so packet lost should not affect the
>> calculation. And RTT shouldn't get affected by packet lost either, as
>> all the information is conceived in one RTCP RR packet.
>> How high is high? (what's the value reported by siprtp?)
>> > Is it a problem in the siprtp or it's my liaison that causes that.??
>> > Thanks!!
>> > _______________________________________________
>> > 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
>> Visit our blog: http://blog.pjsip.org
>> pjsip mailing list
>> pjsip at lists.pjsip.org
> Visit our blog: http://blog.pjsip.org
> pjsip mailing list
> pjsip at lists.pjsip.org
More information about the pjsip