Clips: The media could not be loaded

1.- Clips worked fine previous version 0.9.0-8fd12f0-amd64, after updating 0.9.0-beta2-amd64, adapting settings for events recording (clips settings deprecated): https://github.com/blakeblackshear/frigate/blob/master/docs/docs/configuration/cameras.md

Clips display snapshot correctly but event recording does not play in browser.

Error : “The media could not be loaded, either because the server or network failed or because the format is not supported.”

I have intentionally cropped the image after error msg, looks good in the browser.

2.- Can I only activate event recording without activating 24/7 recording?

Thanks!

Version of frigate 0.9.0-3340952

Config file

  host: 192.168.10.90
  user: user
  password: pass
  topic_prefix: frigate
cameras:
  jardin:
    ffmpeg:
      inputs:
        - path: rtsp://user:pass@192.168.10.103:554/Streaming/Channels/101
          roles:
            - detect
            - clips
            - rtmp
    snapshots:
      enabled: true
      crop: false
      bounding_box: true
    objects:
      track:
        - person
    record:
      enabled: True
      retain_days: 0
      events:
        enabled: true
        max_seconds: 300
        pre_capture: 5
        post_capture: 10
        retain:
          default: 10
    detect:
      width: 2560
      height: 1440
      fps: 5
detectors:
  coral:
    type: edgetpu
    device: usb
birdseye:
  enabled: True
  width: 1920
  height: 1080
  quality: 1
  mode: continuous
ffmpeg:
  hwaccel_args:
    - -hwaccel
    - vaapi
    - -hwaccel_device
    - /dev/dri/renderD128
    - -hwaccel_output_format
    - yuv420p

Frigate container logs

No errors found

Frigate stats

{
  "coches": {
    "camera_fps": 5.1,
    "capture_pid": 240,
    "detection_fps": 2,
    "pid": 230,
    "process_fps": 5.1,
    "skipped_fps": 0
  },
  "detection_fps": 10.5,
  "detectors": {
    "coral": {
      "detection_start": 0,
      "inference_speed": 15.79,
      "pid": 215
    }
  },
  "dormitorios": {
    "camera_fps": 5,
    "capture_pid": 242,
    "detection_fps": 0,
    "pid": 231,
    "process_fps": 5,
    "skipped_fps": 0
  },
  "entrada": {
    "camera_fps": 5,
    "capture_pid": 238,
    "detection_fps": 0,
    "pid": 228,
    "process_fps": 5,
    "skipped_fps": 0
  },
  "hall": {
    "camera_fps": 5.1,
    "capture_pid": 236,
    "detection_fps": 2.9,
    "pid": 226,
    "process_fps": 5.1,
    "skipped_fps": 0
  },
  "jardin": {
    "camera_fps": 5.1,
    "capture_pid": 232,
    "detection_fps": 0,
    "pid": 221,
    "process_fps": 5.1,
    "skipped_fps": 0
  },
  "piscina": {
    "camera_fps": 5.1,
    "capture_pid": 235,
    "detection_fps": 5.6,
    "pid": 222,
    "process_fps": 5,
    "skipped_fps": 0
  },
  "service": {
    "storage": {
      "/dev/shm": {
        "free": 2123,
        "mount_type": "tmpfs",
        "total": 2147.5,
        "used": 24.5
      },
      "/media/frigate/clips": {
        "free": 32623.6,
        "mount_type": "ext4",
        "total": 58011.6,
        "used": 22410.7
      },
      "/media/frigate/recordings": {
        "free": 32623.6,
        "mount_type": "ext4",
        "total": 58011.6,
        "used": 22410.7
      },
      "/tmp/cache": {
        "free": 2000,
        "mount_type": "tmpfs",
        "total": 2000,
        "used": 0
      }
    },
    "uptime": 459,
    "version": "0.9.0-3340952"
  }
}

Computer Hardware

• OS: Ubuntu 20.04
• Install method: Docker Compose
• Coral Version: USB

[HSB-collab]

I have a similar but different problem: my events are not always being recorded. I am running Supervised HA in a Virtualbox VM on an Ubuntu 20.04 host. I have Frigate version 0.9.1-800F33E running in a separate docker container. I have 10 cameras in my setup but to make it simpler I show below the config.yml file for just 1 camera. When an event occurs, I get a snapshot and the Frigate events menu thinks there is a video recording but when I try and play it I get a messasge saying “The media could not be loaded, either because the server or network failed or because the format is not supported.”.

Any ideas as to what I am doing wrong? Also, how do I get the logs out of the Docker container? Also, how do I run ffprobe in the container?

Sorry about the questions. I am guessing I am doing something dumb but what???

My config.yml file is:

mqtt:
  host: 10.0.0.251
  user: mqtt_user
  password: *****
ffmpeg:
  hwaccel_args:
    - -hwaccel
    - vaapi
    - -hwaccel_device
    - /dev/dri/renderD128
    - -hwaccel_output_format
    - yuv420p
  global_args: -hide_banner -loglevel debug    
cameras:
 # Name of your camera
  Camera1:
    ffmpeg:
      inputs:
        - path: rtsp://****:*****@10.0.0.120/img/video.sav
          roles:
            - detect
            - rtmp
            - record
    detect:    
      width: 1280
      height: 720
      fps: 5  
    objects:
      track:
        - person
        - car
      filters:
        person:
           min_score: 0.6
          threshold: 0.66
    record:
      enabled: true
      retain_days: 0
      events:
        max_seconds: 60
        retain:
          default: 10
    snapshots:
      enabled: true

 detectors:
  coral:
    type: edgetpu
    device: pci

[mspinolo]

I have the same issue, 6 camera and happening on all of them by time to time (10% of the clips…more or less)