deepstream-app crashes jetson nano

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

I’ve installed deepstream SDK and its dependencies on Jetson Nano with sdk manager. From /opt/nvidia/deepstream/deepstream-4.0/bin I run

./deepstream-app -c ../../deepstream-4.0/samples/configs/deepstream-app/source30_1080p_dec_infer-resnet_tiled_display_int8.txt

it starts, writes this:

Using winsys: x11 
Creating LL OSD context new
0:00:02.701522305  9270     0x38998f90 WARN                 nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:useEngineFile(): Failed to read from model engine file
0:00:02.701616525  9270     0x38998f90 INFO                 nvinfer gstnvinfer.cpp:519:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:initialize(): Trying to create engine from model files
0:00:02.701902467  9270     0x38998f90 WARN                 nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger:<primary_gie_classifier> NvDsInferContext[UID 1]:generateTRTModel(): INT8 not supported by platform. Trying FP16 mode.

and then device dies (no longer avail through net, keyboard, display is off), I have to reboot it. How to diagnose and fix this?

2

Hi,
You can see the log /var/log/syslog, it may have some useful info,
How many input sources you use, can you lower down the number of input source to see if it can solve your problem?

3

One of the first step in troubleshooting is to check your power supply.
It is recommended that you power the Jetson module through the DC power connector when running DeepStream app. Lot of these apps are very compute intensive and USB adapters may not be able to handle the transients.

4

Resolved by following these instructions (Jetson Nano - Use More Power! - JetsonHacks), not only watch but also read.