After updating to Viseron - 3.3.0 - ebf0c21 (docker roflcoopter/amd64-cuda-viseron:latest) - errors

[motogon]

hi

After updating to Viseron - 3.3.0 - ebf0c21 (docker roflcoopter/amd64-cuda-viseron:latest)

• started giving errors:

2025-10-11 18:09:57.636 [DEBUG ] [viseron.components.ffmpeg.stream.camera_1] - Getting stream information for rtsp://*****:*****@192.168.3.57:554/Sms=48.unicast 2025-10-11 18:09:57.637 [DEBUG ] [viseron.components.ffmpeg.stream.camera_3] - Getting stream information for rtsp://*****:*****@192.168.3.57:554/Sms=38.unicast 2025-10-11 18:09:57.637 [DEBUG ] [viseron.components.ffmpeg.stream.camera_2] - Getting stream information for rtsp://*****:*****@192.168.3.57:554/Sms=50.unicast 2025-10-11 18:09:57.637 [DEBUG ] [viseron.components.ffmpeg.stream.camera_1] - FFprobe command: ffprobe -hide_banner -loglevel error -print_format json -show_error -show_entries stream=codec_type,codec_name,width,height,avg_frame_rate -rtsp_transport tcp rtsp://*****:*****@192.168.3.57:554/Sms=48.unicast 2025-10-11 18:09:57.638 [DEBUG ] [viseron.components.ffmpeg.stream.camera_3] - FFprobe command: ffprobe -hide_banner -loglevel error -print_format json -show_error -show_entries stream=codec_type,codec_name,width,height,avg_frame_rate -rtsp_transport tcp rtsp://*****:*****@192.168.3.57:554/Sms=38.unicast 2025-10-11 18:09:57.638 [DEBUG ] [viseron.components.ffmpeg.stream.camera_2] - FFprobe command: ffprobe -hide_banner -loglevel error -print_format json -show_error -show_entries stream=codec_type,codec_name,width,height,avg_frame_rate -rtsp_transport tcp rtsp://*****:*****@192.168.3.57:554/Sms=50.unicast 2025-10-11 18:10:00.418 [DEBUG ] [viseron.components.ffmpeg.stream.camera_3] - Closing LogPipe 2025-10-11 18:10:00.418 [DEBUG ] [viseron.components.ffmpeg.stream.camera_3] - Stream information from FFprobe: Width: 2592 Height: 1944 FPS: 10 Video Codec: hevc Audio Codec: pcm_alaw 2025-10-11 18:10:00.419 [DEBUG ] [viseron.components.ffmpeg.stream.camera_3] - LogPipe thread ended 2025-10-11 18:10:00.876 [DEBUG ] [viseron.components.ffmpeg.stream.camera_1] - Closing LogPipe 2025-10-11 18:10:00.876 [DEBUG ] [viseron.components.ffmpeg.stream.camera_1] - Stream information from FFprobe: Width: 2880 Height: 1616 FPS: 10 Video Codec: hevc Audio Codec: pcm_mulaw 2025-10-11 18:10:00.877 [DEBUG ] [viseron.components.ffmpeg.stream.camera_1] - LogPipe thread ended 2025-10-11 18:10:01.649 [DEBUG ] [viseron.components.ffmpeg.stream.camera_2] - Closing LogPipe 2025-10-11 18:10:01.650 [DEBUG ] [viseron.components.ffmpeg.stream.camera_2] - Stream information from FFprobe: Width: 704 Height: 576 FPS: 15 Video Codec: h264 Audio Codec: None 2025-10-11 18:10:01.650 [DEBUG ] [viseron.components.ffmpeg.stream.camera_2] - LogPipe thread ended 2025-10-11 18:10:02.509 [ERROR ] [viseron.components] - Uncaught exception setting up domain camera for component ffmpeg: [Errno 28] inotify watch limit reached

Please advise how to fix them.

Config:

` Thanks for trying out Viseron!

This is a small walkthrough of the configuration to get you started.

There are far more components and options available than what is listed here.

See the documentation for the full list of configuration options.

logger:
logs:
viseron.components.ffmpeg: debug

Start by adding some cameras

ffmpeg:
camera:
camera_1: # This value has to be unique across all cameras
name: Gate
host: 192.168.3.57
port: 554
stream_format: rtsp
path: /Sms=48.unicast
username: syno
password: 7961b5e0f2e758f8f6ca04c075f853d7

  #rtsp://syno:7961b5e0f2e758f8f6ca04c075f853d7@192.168.3.57:554/Sms=48.unicast

camera_2:  # This value has to be unique across all cameras
  name: Home
  host: 192.168.3.57
  port: 554
  stream_format: rtsp
  path: /Sms=50.unicast
  username: syno
  password: aa2b7b0b9be14c53aacad32bb7bc96a2

camera_3:  # This value has to be unique across all cameras
  name: Street
  host: 192.168.3.57
  port: 554
  stream_format: rtsp
  path: /Sms=38.unicast
  username: syno
  password: 475bc591e26d98c83e93de0b37180529

Then add an object detector

darknet:
object_detector:
model_path: /detectors/models/darknet/yolov7-tiny.weights
model_config: /detectors/models/darknet/yolov7-tiny.cfg

require_motion: true

cameras:
  camera_1:  # Attach detector to the configured camera_1 above
    fps: 10
    scan_on_motion_only: false  # Scan for objects even when there is no motion
    log_all_objects: true
    labels:
      - label: person
        confidence: 0.80
        trigger_recorder: true
      - label: vehicle
        confidence: 0.70
        trigger_recorder: true
      - label: dog
        confidence: 0.7
        trigger_event_recording: true
      - label: cat
        confidence: 0.8
        trigger_event_recording: true  


  camera_2:  # Attach detector to the configured camera_2 above
    fps: 10
    scan_on_motion_only: false  # Scan for objects even when there is no motion
    log_all_objects: true
    labels:
      - label: person
        confidence: 0.80
        trigger_recorder: true
      - label: dog
        confidence: 0.7
        trigger_event_recording: true
      - label: cat
        confidence: 0.8
        trigger_event_recording: true  

  camera_3:  # Attach detector to the configured camera_1 above
    fps: 10
    scan_on_motion_only: false  # Scan for objects even when there is no motion
    log_all_objects: true
    labels:
      - label: person
        confidence: 0.80
        trigger_recorder: true
      - label: vehicle
        confidence: 0.70
        trigger_recorder: true
      - label: dog
        confidence: 0.7
        trigger_event_recording: true
      - label: cat
        confidence: 0.8
        trigger_event_recording: true  

You can also use motion detection

mog2:
motion_detector:

cameras:

  camera_2:  # Attach detector to the configured camera_2 above
    fps: 10
    recorder_keepalive: true
    threshold: 30
  camera_1:  # Attach detector to the configured camera_2 above
    fps: 10
  camera_3:  # Attach detector to the configured camera_2 above
    fps: 10

To tie everything together we need to configure one more component.

nvr:
camera_1: # Run NVR for camera_1
camera_2: # Run NVR for camera_2
camera_3: # Run NVR for camera_2

telegram:
telegram_bot_token: xxxxxx:xxxxxxxxxx-ew2Z_QIMEnWg9tak
telegram_chat_ids: [xxxxxxxxx]

telegram_user_ids:

- xxxxxxx

detection_label:
- person
- car
- truck
- dog
send_detection_thumbnail: true
send_detection_video: true
send_detection_message: true
cameras:
camera_1:
camera_2:
camera_3:

mqtt:
broker: 192.168.3.45
port: 1883
username: mqtt1
password: mqtt
home_assistant:
discovery_prefix: Viseron

Now you can restart Viseron and you should be good to go!`