[Detector Support]: Cameras not working after rebuild

[aacostarua]

Describe the problem you are having

No cameras work right now. Was working before, had to re-install everything now it doesn't seem to work.

Version

0.13.1

Frigate config file

mqtt:
  host: 192.168.1.10
  port: 1883
  user: admin
  password: xx



detectors:
  tensorrt:
    type: tensorrt
    device: 0 # This is the default, select the first GPU


model:
  path: /config/yolov7-320.trt
  input_tensor: nchw
  input_pixel_format: rgb
  width: 320
  height: 320

ffmpeg:
  hwaccel_args: preset-nvidia-h265


record:
  enabled: False
  retain:
    days: 3
    mode: motion
  events:
    retain:
      default: 30
      mode: motion

cameras:
  safe_closet:
    ffmpeg:
      inputs:
        - path: rtsp:https://admin:[email protected]:554/cam/realmonitor?channel=1&subtype=0 
          roles:
            - detect
            - rtmp
    objects:
      track:
        - person
        - dog
        - cat
    snapshots:
      enabled: True
      timestamp: False
      bounding_box: False
      crop: False
      required_zones: []
      retain:
        default: 10
        objects:
          person: 15

  living_room:
    ffmpeg:
      inputs:
        - path: rtsp:https://admin:[email protected]:554/cam/realmonitor?channel=1&subtype=0 
          roles:
            - detect
            - rtmp
    objects:
      track:
        - person
        - dog
        - cat
    snapshots:
      enabled: True
      timestamp: False
      bounding_box: False
      crop: False
      required_zones: []
      retain:
        default: 10
        objects:
          person: 15

docker-compose file or Docker CLI command

version: "3.9"
services:
  frigate:
    environment:
    - YOLO_MODELS=yolov7-320
    container_name: frigate
    privileged: true
    restart: unless-stopped
    image: ghcr.io/blakeblackshear/frigate:stable-tensorrt
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              device_ids: ['0']
              capabilities: [gpu]
    shm_size: "256mb"
    devices:
      - /dev/bus/usb:/dev/bus/usb
      - /dev/dri/renderD128
      - /dev/dri/renderD129
      - /dev/bus/usb/001/004 # Assuming this is the Google Coral USB device location
    volumes:
      - /etc/localtime:/etc/localtime:ro
      - ./config:/config
      - /data/frigate:/media/frigate
      - type: tmpfs
        target: /tmp/cache
        tmpfs:
          size: 1000000000
    ports:
      - "5000:5000"
      - "1935:1935"
      - "8554:8554"
      - "8555:8555/tcp"
      - "8555:8555/udp"


  mqtt:
    container_name: mqtt
    image: eclipse-mosquitto:1.6
    ports:
      - "1883:1883"
    environment:
      - MQTT_USERNAME=admin
      - MQTT_PASSWORD=xxx

Relevant log output

frigate.detectors.plugins.tensorrt INFO    : Loaded engine size: 181 MiB
2024-04-04 16:08:47.363005685  [2024-04-04 16:08:47] frigate.detectors.plugins.tensorrt WARNING : Using an engine plan file across different models of devices is not recommended and is likely to affect performance or even cause errors.
2024-04-04 16:08:47.416570289  Cuda failure in file 'yolo_layer.cu' line 70: 35
2024-04-04 16:08:47.416588829  Fatal Python error: Aborted
2024-04-04 16:08:47.416590144  
2024-04-04 16:08:47.416596547  Thread 0x0000713c09c226c0 (most recent call first):
2024-04-04 16:08:47.416631266    File "/usr/lib/python3.9/threading.py", line 312 in wait
2024-04-04 16:08:47.416701126    File "/usr/lib/python3.9/multiprocessing/queues.py", line 233 in _feed
2024-04-04 16:08:47.416731723    File "/usr/lib/python3.9/threading.py", line 892 in run
2024-04-04 16:08:47.416780209    File "/usr/lib/python3.9/threading.py", line 954 in _bootstrap_inner
2024-04-04 16:08:47.416844836    File "/usr/lib/python3.9/threading.py", line 912 in _bootstrap
2024-04-04 16:08:47.416851986  
2024-04-04 16:08:47.416853305  Current thread 0x0000713c0e743740 (most recent call first):
2024-04-04 16:08:47.416950078    File "/opt/frigate/frigate/detectors/plugins/tensorrt.py", line 89 in _load_engine
2024-04-04 16:08:47.416998128    File "/opt/frigate/frigate/detectors/plugins/tensorrt.py", line 220 in __init__
2024-04-04 16:08:47.417071493    File "/opt/frigate/frigate/detectors/__init__.py", line 18 in create_detector
2024-04-04 16:08:47.417116976    File "/opt/frigate/frigate/object_detection.py", line 53 in __init__
2024-04-04 16:08:47.417187214    File "/opt/frigate/frigate/object_detection.py", line 102 in run_detector
2024-04-04 16:08:47.417231891    File "/usr/lib/python3.9/multiprocessing/process.py", line 108 in run
2024-04-04 16:08:47.417302806    File "/usr/lib/python3.9/multiprocessing/process.py", line 315 in _bootstrap
2024-04-04 16:08:47.417360154    File "/usr/lib/python3.9/multiprocessing/popen_fork.py", line 71 in _launch
2024-04-04 16:08:47.417429088    File "/usr/lib/python3.9/multiprocessing/popen_fork.py", line 19 in __init__
2024-04-04 16:08:47.417478449    File "/usr/lib/python3.9/multiprocessing/context.py", line 277 in _Popen
2024-04-04 16:08:47.417546399    File "/usr/lib/python3.9/multiprocessing/context.py", line 224 in _Popen
2024-04-04 16:08:47.417595077    File "/usr/lib/python3.9/multiprocessing/process.py", line 121 in start
2024-04-04 16:08:47.417641771    File "/opt/frigate/frigate/object_detection.py", line 183 in start_or_restart
2024-04-04 16:08:47.417715084    File "/opt/frigate/frigate/object_detection.py", line 151 in __init__
2024-04-04 16:08:47.417754601    File "/opt/frigate/frigate/app.py", line 453 in start_detectors
2024-04-04 16:08:47.417822566    File "/opt/frigate/frigate/app.py", line 683 in start
2024-04-04 16:08:47.417867084    File "/opt/frigate/frigate/__main__.py", line 17 in <module>
2024-04-04 16:08:47.417937285    File "/usr/lib/python3.9/runpy.py", line 87 in _run_code
2024-04-04 16:08:47.417986934    File "/usr/lib/python3.9/runpy.py", line 197 in _run_module_as_main

Operating system

Debian

Install method

Docker Compose

Coral version

USB

Any other information that may be helpful

Running frigate from: /media/user1/Storage/frigate-0.13.1
Had it working with my CPU and GPU previously, now nothing currently works.

sudo docker run -it --rm --gpus all ubuntu nvidia-smi
[sudo] password for user1:
Thu Apr 4 20:45:48 2024
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.239.06 Driver Version: 470.239.06 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 NVIDIA GeForce ... Off | 00000000:02:00.0 On | N/A |
| 30% 45C P2 45W / 151W | 478MiB / 8116MiB | 2% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
+-----------------------------------------------------------------------------+

[aacostarua]

Any idea what causes this error @NickM-27 ? How should I try fixing this??

[NickM-27]

Not much to go off of. Driver looks really new so maybe that is not playing nice, or maybe the nvidia container toolkit wasn't installed correctly

[aacostarua]

Works now thanks! Reinstall NVIDIA drivers to 535.

Would like to know how to add the Google Coral to the setup above to work with the GPU if you don't mind showing me what's up with that.

[NickM-27]

You can't mix detector types. You can use coral for detection and GPU for decoding.