[Live-devel] Re: bug in openRTSP?

[Norbert Donath]

On Mon, Jan 19, 2004 at 10:31:24 -0800, Ross Finlayson wrote:

>> I loaded a stream with openRTSP where the SDP description shows a
>> permanent stream (t=0 0). There was no r= in the SDP. I know that this
>> is a strange behaviour of the server as the stream is only 60 sec and
>> it is terminated by the server after 60 sec with a RTCP BYE.
>>
>> The problem ist that openRTSP shows the following:
>> Receiving streamed data (for up to 234 seconds)...
> 
> That's odd.  Is the "rtsp://" URL publically accessible, so we can try this 
> for ourselves?

Only from within a German mobile network.


> If not, could you please post the output from running "openRTSP -V" on this 
> URL?

I took a closer look on this problem and discovered the reason for
this message but I'm not quite sure if it is a bug or a feature. From
the description on your web page I would say it's a feature while from
what I expected as a user I would call it a bug ;-)

As I call openRTSP from a Tcl-Script with a defined end time (it is
automatically killed if it runs longer) I always call openRTSP with a
positive value for the -e parameter so that openRTSP always returns
right before the Tcl-Script would be killed.

In this case the SDP description says this:
a=range:npt=0-54.200000
for the audio and video subsession. I called openRTSP with -e 234. I
expected that openRTSP stopps downloading a stream if the stream is
shorter than the maximum duration given with -e but it looks like it
tries to download exactly this time even if the stream is shorter. Is
this behaviour intended?


Best Regards,
Norbert Donath


-- 
Visit us at "3 GSM World Congress 2004" in Cannes,
February 23rd - 26th, hall 2, stand F 36

SIGOS Systemintegration GmbH
 - TESTING IS OUR COMPETENCE - 
Fon +49 911 95168-0
www.sigos.de