How to fix in testing peoplenet? ERROR: <gst_nvinfer_parse_config_file:1242>: failed ERROR from primary_gie: Configuration file parsing failed

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) RTX 1060
• DeepStream VersionDeepstram 5.0
• JetPack Version (valid for Jetson only)
• TensorRT Version7.0.0.11
**• NVIDIA GPU Driver Version (valid for GPU only)**450.51
• Issue Type( questions, new requirements, bugs)
The errors I met:
(gst-plugin-scanner:12632): GStreamer-WARNING **: 09:30:47.081: Failed to load plugin ‘/usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_inferserver.so’: libtrtserver.so: cannot open shared object file: No such file or directory
Failed to load config file: No such file or directory
** ERROR: <gst_nvinfer_parse_config_file:1242>: failed
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream-5.0/lib/libnvds_mot_klt.so
gstnvtracker: Optional NvMOT_RemoveStreams not implemented
gstnvtracker: Batch processing is OFF
gstnvtracker: Past frame output is OFF
0:00:02.026577298 12631 0x55fa7de116f0 WARN nvinfer gstnvinfer.cpp:766:gst_nvinfer_start:<primary_gie> error: Configuration file parsing failed
0:00:02.026591276 12631 0x55fa7de116f0 WARN nvinfer gstnvinfer.cpp:766:gst_nvinfer_start:<primary_gie> error: Config file path: /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt
** ERROR: main:655: Failed to set pipeline to PAUSED
Quitting
ERROR from primary_gie: Configuration file parsing failed
Debug info: gstnvinfer.cpp(766): gst_nvinfer_start (): /GstPipeline:pipeline/GstBin:primary_gie_bin/GstNvInfer:primary_gie:
Config file path: /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt
App run failed

• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
Errors in tesing people in using deepstream SDK. I found one similar topic, and reset the setting in the config as the original setting in deepstream of people. But for my case, it doesn’t work.
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
I uploaded my config files and need some helps.config_infer_primary_peoplenet.txt (2.0 KB) deepstream_app_source1_peoplenet.txt (3.4 KB)

Can you try to comment out “model-engine-file=/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/peoplenet/resnet34_peoplenet_pruned.etlt_b1_gpu0_fp16.engine” in config_infer_primary_peoplenet.txt for the first time runing?

I tried to run in the root, and comment out. Still have:
Failed to load config file: No such file or directory
** ERROR: <gst_nvinfer_parse_config_file:1242>: failed
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream-5.0/lib/libnvds_mot_klt.so
gstnvtracker: Optional NvMOT_RemoveStreams not implemented
gstnvtracker: Batch processing is OFF
gstnvtracker: Past frame output is OFF
0:00:00.265447881 13907 0x55ab588c40f0 WARN nvinfer gstnvinfer.cpp:766:gst_nvinfer_start:<primary_gie> error: Configuration file parsing failed
0:00:00.265461863 13907 0x55ab588c40f0 WARN nvinfer gstnvinfer.cpp:766:gst_nvinfer_start:<primary_gie> error: Config file path: /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt
** ERROR: main:655: Failed to set pipeline to PAUSED
Quitting
ERROR from primary_gie: Configuration file parsing failed
Debug info: gstnvinfer.cpp(766): gst_nvinfer_start (): /GstPipeline:pipeline/GstBin:primary_gie_bin/GstNvInfer:primary_gie:
Config file path: /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt
App run failed

This error means some file is not found or can not be read.

Can you check whether there is resnet34_peoplenet_pruned.etlt file in your /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/peoplenet folder?

root@ai-System-Product-Name:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/peoplenet# ls
labels.txt resnet34_peoplenet_pruned.etlt

I double-checked, the etlt file is in this folder.
And for the config file:
root@ai-System-Product-Name:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models# ls
config_infer_primary_dashcamnet.txt config_infer_primary_retinanet.txt deepstream_app_source1_detection_models.txt frcnn_labels.txt mrcnn_labels.txt
config_infer_primary_detectnet_v2.txt config_infer_primary_ssd.txt deepstream_app_source1_facedetectir.txt labels_dashcamnet.txt peoplenet
config_infer_primary_dssd.txt config_infer_primary_trafficcamnet.txt deepstream_app_source1_mrcnn.txt labels_facedetectir.txt README
config_infer_primary_facedetectir.txt config_infer_primary_yolov3.txt deepstream_app_source1_peoplenet.txt labels_peoplenet.txt retinanet_labels.txt
config_infer_primary_frcnn.txt config_infer_secondary_vehiclemakenet.txt deepstream_app_source1_trafficcamnet.txt labels_trafficnet.txt ssd_labels.txt
config_infer_primary_mrcnn.txt config_infer_secondary_vehicletypenet.txt detectnet_v2_labels.txt labels_vehiclemakenet.txt yolov3_labels.txt
config_infer_primary_peoplenet.txt deepstream_app_source1_dashcamnet_vehiclemakenet_vehicletypenet.txt dssd_labels.txt labels_vehicletypenet.txt

Its is better to remove “model-engine-file=peoplenet/resnet34_peoplenet_pruned.etlt_b1_gpu0_fp16.engine” in deepstream_app_source1_peoplenet.txt too.

Before I reset the original config, I found one FP32.engine in that peoplenet folder when I set FP32 setting.
But that time came out this exactly same errors.

I have removed that line and back with the same errors.
That is confusing that I saw the engine in this folder so it is not the tlt-export problem, right?

The error is clear, there is something wrong in your config_infer_primary_peoplenet.txt.

Is the “labelfile-path=labels_peoplenet.txt” the correct setting for your case? You must be careful for using relative path.

I tried many ways for path settings already, relative or the complete paths
,for now all full paths

config-file=/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt

root@ai-System-Product-Name:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models
/peoplenet# cat /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/config_infer_primary_peoplenet.txt
################################################################################

Copyright © 2020, NVIDIA CORPORATION. All rights reserved.

Permission is hereby granted, free of charge, to any person obtaining a

tlt-encoded-model=/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/peoplenet/resnet34_peoplenet_pruned.etlt

root@ai-System-Product-Name:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/peoplenet# ls
labels.txt resnet34_peoplenet_pruned.etlt

labelfile-path=/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/labels_peoplenet.txt

root@ai-System-Product-Name:/opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models#
cat /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/labels_peoplenet.txt
Person
Bag
Face

and same errors, that’s why I feel sad.

This error shows the config file is not found.
Are you runing “deepstream-app -c deepstream_app_source1_peoplenet.txt” in /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/? Please check your deepstream_app_source1_peoplenet.txt setting and “config-file” parameter in config_infer_primary_peoplenet.txt. Since you changed the original config file, we can not know the actual path and file used.

I rewrote the whole config and finally it is working!
And your help about comment out engine is important!
My platform just support for FP32!
Very kind of you! Thank you so much!