NvMMLiteOpen : Block : BlockType = 261 not getting displayed

• Hardware Platform (Jetson / GPU) = Jetson Tx2
• DeepStream Version = deepstream 5
• JetPack Version (valid for Jetson only) = 32.4.3
• TensorRT Version =7.1.3

Hello all, I am working on a application where i take rtsp stream as input and do processing but i am facing a issue in that when i try to run the application the block is not getting initialized .
NvMMLiteOpen : Block : BlockType = 261
NVMEDIA: Reading vendor.tegra.display-size : status: 6
NvMMLiteBlockCreate : Block : BlockType = 261
NvMMLiteOpen : Block : BlockType = 261
NVMEDIA: Reading vendor.tegra.display-size : status: 6
NvMMLiteBlockCreate : Block : BlockType = 261

It just loads the model but does to start processing it .

using the code of deepstream-test3 for my application (link: deepstream_python_apps/apps/deepstream-test3 at master · NVIDIA-AI-IOT/deepstream_python_apps · GitHub)

I am not sure what is the issue .
can anyone help to identify the issue and how to resolve it .

thank you

These logs are fine. Have you modified deepstream_test_3.py? Can you use “export GST_DEBUG=3” to get more log?

This should be printed right but this is not getting printed . it stops after model loading

Can your board run with other deepstream cases? This log means start to use HW video decoder, if there is no such log, you need to check whether the RTSP streams has been parsed and passed to video decoder.

Yes it works for other case like for simple video file .

You need to check whether the RTSP streams has been parsed and passed to video decoder.

okay how to check that ?

Can you use “export GST_DEBUG=3” or “export GST_DEBUG=5” to get more log?

@krupalpurohit1997

Any update? Please help to provide more log to move this issue forwrd.

Thanks

Hello @kayccc, i have tried with setting GST_DEBUG =3 and with that i am getting following logs

2020-11-30 14:44:10,216 - INFO - PIPELINE STARTING…
0:01:13.654395462 32614 0x7ea0009680 FIXME default gstutils.c:3981:gst_pad_create_stream_id_internal:fakesrc0:src Creating random stream-id, consider implementing a deterministic way of creating a stream-id
0:01:13.654480709 32614 0x7e98005de0 FIXME default gstutils.c:3981:gst_pad_create_stream_id_internal:fakesrc2:src Creating random stream-id, consider implementing a deterministic way of creating a stream-id
0:01:13.654384710 32614 0x7ea0008ed0 FIXME default gstutils.c:3981:gst_pad_create_stream_id_internal:fakesrc1:src Creating random stream-id, consider implementing a deterministic way of creating a stream-id
0:01:39.129828774 32614 0x7e94031450 WARN rtspsrc gstrtspsrc.c:3155:on_timeout: source 00000000, stream 00000000 in session 0 timed out
0:01:41.011575602 32614 0x7e94006230 WARN rtspsrc gstrtspsrc.c:3155:on_timeout: source 00000000, stream 00000000 in session 0 timed out
0:01:41.184811443 32614 0x7ea0008f70 WARN rtspsrc gstrtspsrc.c:3155:on_timeout: source 00000000, stream 00000000 in session 0 timed out

There is no rtsp packet received in source0.

So it may be the RTSP link problem. You need to check your RTSP server settings.

Hello @Fiona.Chen, which server setting is needed to be checked for this issue ?

You need to consult your vendor - e.g. the camera manufacturer if you are using ip cameras.

ok no problem so what question should i ask them for this issue ? and also is there any workaround possible for this in deepstream app code ?

as this issue is not occurring everytime , we are facing this issue few time a day , other then this few times the deepstream is running .

You may just describe your problem to your vendor. They provide the RTSP server functions, they must know RTSP related knowledge.
To get more log for you to report the bug to your vendor. You can use the following command line to reproduce the issue and get detailed log from client side.
gst-launch-1.0 rtspsrc location=rtsp://xxxxxxxx debug=1 ! queue ! rtph264depay ! fakesink

More information, please refer to rtspsrc: GStreamer Good Plugins 1.0 Plugins Reference Manual
RTSP spec
RFC 2326: Real Time Streaming Protocol (RTSP)

ok sure will do it

thank you