Can't Run Deepstream on Nano

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hello,

I have a Jetson Nano Devkit and want to run deepstream samples on it. I am using a SanDisk Ultra 200GB SD card.

I created a new image file using create-jetson-nano-sd-card-image.sh script in JetPack_4.2.3_Linux_GA_P3448. Wrote it over SD card with Etcher app. Nano booted, root size was 181GB.

I re-run sdkmanager to install other packages to Nano without flashing. Didn’t put Nano in recover mode, sdkmanager connected nano over network and installed all packages. Done successfully.

Tried to run 8 video deepstream sample:

deepstream-app -c source8_1080p_dec_infer-resnet_tracker_tiled_display_fp16_nano.txt

Unfortunately, it couldn’t run:

root@nano:/opt/nvidia/deepstream/deepstream-4.0/samples/configs/deepstream-app# deepstream-app -c source8_1080p_dec_infer-resnet_tracker_tiled_display_fp16_nano.txt
** ERROR: <create_multi_source_bin:682>: Failed to create element 'src_bin_muxer'
** ERROR: <create_multi_source_bin:745>: create_multi_source_bin failed
** ERROR: <create_pipeline:1045>: create_pipeline failed
** ERROR: <main:632>: Failed to create pipeline
Quitting
App run failed

This is the output of “deepstream-app --version-all”:

deepstream-app version 4.0.1
DeepStreamSDK 4.0.1
CUDA Driver Version: 10.0
CUDA Runtime Version: 10.0
TensorRT Version: 5.1
cuDNN Version: 7.5
libNVWarp360 Version: 2.0.0d5

How can I fix this issue? Any help will be appreciated.

Thanks.

Solution is removing:

/home/<user>/.cache/gstreamer-1.0/registry.aarch64.bin
2

Hi r1ch13r1ch,

Confirmed the deepstream is working on JetPack-4.2.3 (r32.2.3) with DS-4.0.1 on Jetson-Nano.
Please make sure all the sdk components are installed. (include CUDA, cuDNN and MMAPI…)

3

Hi carolyuu,

Thank you for your answer.

Problem resolved after removing:

/home/<user>/.cache/gstreamer-1.0/registry.aarch64.bin
4

by rm /home//.cache/gstreamer-1.0/registry.aarch64.bin does not work for me.