Please provide complete information as applicable to your setup.
• Hardware Platform (GPU) • DeepStream Version 6.0 • TensorRT Version 8.0.1 • NVIDIA GPU Driver Version (valid for GPU only) 470.63.01
Hi,
I am getting the following error while trying to run a pretrained model using DeepStream SDK. I am using an Azure VM with a NVIDIA T4 GPU.
Error:
cuGraphicsGLRegisterBuffer failed with error(219) gst_eglglessink_cuda_init texture = 1
**PERF: 357.48 (356.92)
**PERF: 638.46 (594.00)
**PERF: 642.81 (616.60)
ERROR from tiled_display_queue: Internal data stream error.
Debug info: gstqueue.c(988): gst_queue_handle_sink_event (): /GstPipeline:pipeline/GstBin:tiled_display_bin/GstQueue:tiled_display_queue:
streaming stopped, reason not-negotiated (-4)
Quitting
ERROR from sink_bin_queue: Internal data stream error.
Debug info: gstqueue.c(988): gst_queue_handle_sink_event (): /GstPipeline:pipeline/GstBin:processing_bin_0/GstBin:sink_bin/GstQueue:sink_bin_queue:
streaming stopped, reason not-negotiated (-4)
[NvMultiObjectTracker] De-initialized
App run failed
I have followed one of the suggestions from a forum (link given below) but still no use. I would like to see a video being inferred after running one of the pretrained models using DeepStream SDK.
Please give more logs about 5A not helps.
besides, you can use filesink like deepstream-app, also you can use rtspstreaming, view the output through player like vlc.
Hi @Amycao We are having a T4 NVIDIA GPU. I have heard that it has no display drivers. As per the requirement to install DeepStream 6.0 we have upgraded our driver to 470.63.01 by following the steps from the below link ( dGPU Setup for Ubuntu)