[Bug]: /dev/shm grows rapidly

[wes1993]

Describe the problem you are having

Hello,
I have found a BUG that cause the /dev/shm to grows rapidly and use all the space allocated (12 GB).

Steps to reproduce

Restart the container with command
docker compose -f /home/stefano/Containers/Frigate/docker-compose.yml restart

Version

Frigate (0.14.0-4e5a6eb)

Frigate config file

mqtt:
  enabled: true #Needed For HomeAssistant Control
  host: 192.168.0.22
  user: mqtt
  password: HiddenByME
database:
  path: /config/Database/frigate.db
#logger:
# Optional: default log level (default: shown below)
#  default: debug
ffmpeg:
  # Optional: global hwaccel args (default: shown below)
  # NOTE: See hardware acceleration docs for your specific device
  #hwaccel_args: []
  #  hwaccel_args:
  #    - -hwaccel
  #    - vaapi
  #    - -hwaccel_device
  #    - /dev/dri/renderD128
  #    - -hwaccel_output_format
  #    - yuv420p
  #hwaccel_args: -c:v h264_cuvid
  #hwaccel_args: -c:v hevc_cuvid
  hwaccel_args: preset-nvidia-h265
  output_args:
    #    record: -an
    record: preset-record-generic-audio-aac
#    record: preset-record-generic-audio-copy
#  output_args:
# Optional: output args for record streams (default: shown below)
#    record: -f segment -segment_time 60 -segment_format mp4 -reset_timestamps 1 -strftime 1 -c copy -an
# use Nvidia GPU for h264 decoding
#  input_args:
#    - -c:v
#    - h264_cuvid
# try to prevent smearing of the video from packetloss
#    - -rtsp_transport
#    - tcp

# Optional: Audio Events Configuration
# NOTE: Can be overridden at the camera level
audio:
  # Optional: Enable audio events (default: shown below)
  enabled: true
  # Optional: Configure the amount of seconds without detected audio to end the event (default: shown below)
  #max_not_heard: 30
  # Optional: Configure the min rms volume required to run audio detection (default: shown below)
  # As a rule of thumb:
  #  - 200 - high sensitivity
  #  - 500 - medium sensitivity
  #  - 1000 - low sensitivity
  #min_volume: 500
  # Optional: Types of audio to listen for (default: shown below)
  listen:
    - bark
    - fire_alarm
    - scream
    - speech
    - yell
    # Optional: Filters to configure detection.
#  filters:
# Label that matches label in listen config.
#    speech:
# Minimum score that triggers an audio event (default: shown below)
#      threshold: 0.8

# Optional: Object configuration
# NOTE: Can be overridden at the camera level
objects:
  # Optional: list of objects to track from labelmap.txt (default: shown below)
  track:
    - person
    - car
    - motorcycle
    - dog
    - cat
  # Optional: filters to reduce false positives for specific object types
  filters:
    person:
      # Optional: minimum width*height of the bounding box for the detected object (default: 0)
      #min_area: 5000
      # Optional: maximum width*height of the bounding box for the detected object (default: 24000000)
      #max_area: 100000
      # Optional: minimum width/height of the bounding box for the detected object (default: 0)
      #min_ratio: 0.5
      # Optional: maximum width/height of the bounding box for the detected object (default: 24000000)
      #max_ratio: 2.0
      # Optional: minimum score for the object to initiate tracking (default: shown below)
      #min_score: 0.5
      # Optional: minimum decimal percentage for tracked object's computed score to be considered a true positive (default: 0.7)
      threshold: 0.88
      # Optional: mask to prevent this object type from being detected in certain areas (default: no mask)
      # Checks based on the bottom center of the bounding box of the object
      #mask: 0,0,1000,0,1000,200,0,200
detectors:
  tensorrt:
    type: tensorrt
    device: 0 #This is the default, select the first GPU
detect:
  enabled: true
model:
  path: /config/model_cache/tensorrt/yolov7-320.trt
  #path: /config/model_cache/tensorrt/yolov7-640.trt
  input_tensor: nchw
  input_pixel_format: rgb
  #  width: 416
  width: 320
  #  height: 416
  height: 320
  # Optional: Label name modifications. These are merged into the standard labelmap.
  labelmap:
    16: dog
# Include all cameras by default in Birdseye view
birdseye:
  enabled: true
  mode: motion
# Optional: Record configuration
# NOTE: Can be overridden at the camera level
#record:
# Optional: Enable recording (default: shown below)
# WARNING: If recording is disabled in the config, turning it on via
#          the UI or MQTT later will have no effect.
# WARNING: Frigate does not currently support limiting recordings based
#          on available disk space automatically. If using recordings,
#          you must specify retention settings for a number of days that
#          will fit within the available disk space of your drive or Frigate
#          will crash.
#enabled: false
# Optional: Number of minutes to wait between cleanup runs (default: shown below)
# This can be used to reduce the frequency of deleting recording segments from disk if you want to minimize i/o
#  expire_interval: 60
# Optional: Retention settings for recording
#  retain:
# Optional: Number of days to retain recordings regardless of events (default: shown below)
# NOTE: This should be set to 0 and retention should be defined in events section below
#       if you only want to retain recordings of events.
#    days: 2
# Optional: Mode for retention. Available options are: all, motion, and active_objects
#   all - save all recording segments regardless of activity
#   motion - save all recordings segments with any detected motion
#   active_objects - save all recording segments with active/moving objects
# NOTE: this mode only applies when the days setting above is greater than 0
#    mode: all

#record:
#  enabled: False
#  retain:
# Optional: Number of days to retain recordings regardless of events (default: shown below)
# NOTE: This should be set to 0 and retention should be defined in events section below
#       if you only want to retain recordings of events.
#    days: 3
# Optional: Mode for retention. Available options are: all, motion, and active_objects
#   all - save all recording segments regardless of activity
#   motion - save all recordings segments with any detected motion
#   active_objects - save all recording segments with active/moving objects
# NOTE: this mode only applies when the days setting above is greater than 0
#    mode: all
# Optional: Number of minutes to wait between cleanup runs (default: shown below)
# This can be used to reduce the frequency of deleting recording segments from disk if you want to minimize i/o
#  expire_interval: 60
#  events:
#    retain:
#      default: 10
#      mode: active_objects
record:
  enabled: true
  retain:
    days: 7
    mode: all
    #mode: active_objects
  events:
    # Optional: Number of seconds before the event to include (default: shown below)
    pre_capture: 15
    # Optional: Number of seconds after the event to include (default: shown below)
    post_capture: 15
    # Optional: Objects to save recordings for. (default: all tracked objects)
    retain:
      default: 14
      mode: active_objects
# Optional: Configuration for the jpg snapshots written to the clips directory for each event
# NOTE: Can be overridden at the camera level
snapshots:
  # Optional: Enable writing jpg snapshot to /media/frigate/clips (default: shown below)
  enabled: true
  # Optional: save a clean PNG copy of the snapshot image (default: shown below)
  #clean_copy: True
  # Optional: print a timestamp on the snapshots (default: shown below)
  #timestamp: True
  # Optional: draw bounding box on the snapshots (default: shown below)
  #bounding_box: False
  # Optional: crop the snapshot (default: shown below)
  #crop: False
  # Optional: height to resize the snapshot to (default: original size)
  #height: 175
  # Optional: Restrict snapshots to objects that entered any of the listed zones (default: no required zones)
  #required_zones: []
  # Optional: Camera override for retention settings (default: global values)
  #retain:
  # Required: Default retention days (default: shown below)
  #default: 10
  # Optional: Per object retention days
  #objects:
  #person: 15
# Optional: in-feed timestamp style configuration
# NOTE: Can be overridden at the camera level
timestamp_style:
  # Optional: Position of the timestamp (default: shown below)
  #           "tl" (top left), "tr" (top right), "bl" (bottom left), "br" (bottom right)
  position: tl
  # Optional: Format specifier conform to the Python package "datetime" (default: shown below)
  #           Additional Examples:
  #             german: "%d.%m.%Y %H:%M:%S"
  format: '%d.%m.%Y %H:%M:%S'
  # Optional: Color of font
  #color:
  # All Required when color is specified (default: shown below)
  #red: 255
  #green: 255
  #blue: 255
  # Optional: Line thickness of font (default: shown below)
  #thickness: 2
  # Optional: Effect of lettering (default: shown below)
  #           None (No effect),
  #           "solid" (solid background in inverse color of font)
  #           "shadow" (shadow for font)
  #effect: None
go2rtc:
  streams:
    Parcheggio: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=0&stream=0
      #      - ffmpeg:Parcheggio#audio=aac
    Parcheggio_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=0&stream=1
    BagnoStudio: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=1&stream=0
    BagnoStudio_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=1&stream=1
    CameraOspiti: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=2&stream=0
    CameraOspiti_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=2&stream=1
    GiardinoPosteriore: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=3&stream=0
    GiardinoPosteriore_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=3&stream=1
    Ripostiglio: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=4&stream=0
    Ripostiglio_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=4&stream=1
    ParcheggioMoto: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=5&stream=0
    ParcheggioMoto_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=5&stream=1
    Laghetto: # Risoluzione video: 2304x1296
      - bubble:https://admin:[email protected]:80/bubble/live?ch=6&stream=0
    Laghetto_SD: # Risoluzione video: 640x360
      - bubble:https://admin:[email protected]:80/bubble/live?ch=6&stream=1
    Capanno: # Risoluzione video: 1920x1080
      - rtsp:https://192.168.0.57:554/realmonitor?channel=0&stream=0.sdp
    Capanno_SD: # Risoluzione video: 640x360
      - rtsp:https://192.168.0.57:554/realmonitor?channel=0&stream=1.sdp
    Veranda: # Risoluzione video: 1920x1080
      - rtsp:https://192.168.0.59:554/realmonitor?channel=0&stream=0.sdp
    Veranda_SD: # Risoluzione video: 640x360
      - rtsp:https://192.168.0.59:554/realmonitor?channel=0&stream=1.sdp
    Studio: # Risoluzione video: 1920x1080
      - rtsp:https://192.168.0.60:554/realmonitor?channel=0&stream=0.sdp
    Studio_SD: # Risoluzione video: 640x360
      - rtsp:https://192.168.0.60:554/realmonitor?channel=0&stream=1.sdp
    Cucina: # Risoluzione video: 1280X720
      - rtsp:https://admin:[email protected]:554/Streaming/Channels/101
    Cucina_SD: # Risoluzione video: 768x432
      - rtsp:https://admin:[email protected]:554/Streaming/Channels/102
cameras:
  Parcheggio:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Parcheggio?video=all&audio=all # <--- the name here must match the name of the camera in restream
          #- path: http:https://127.0.0.1:1984/api/stream.mp4?src=Parcheggio&mp4=all
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Parcheggio_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 0
    motion:
      mask:
        - 0,0.09,0.164,0.097,0.165,0.045,0.305,0.045,0.307,0,0,0
        - 0.331,0.347,0.638,0.351,0.635,0.905,0.431,0.908,0.431,0.773,0.336,0.755
      threshold: 80
      contour_area: 10
      improve_contrast: 'true'
    zones:
      Vialetto2:
        coordinates: 0.326,0.988,0.363,0.659,0.337,0.252,0.347,0.232,0.347,0.229,0.624,0.238,0.71,0.293,0.836,0.342,0.863,0.09,0.932,0.11,1,0.144,1,0.322,0.969,0.529,0.931,0.713,0.979,0.826,1,1
        inertia: 3
      Vialetto1:
        coordinates: 0.332,0.222,0.332,0,0.52,0,0.57,0,0.746,0,0.839,0.171,0.58,0.145,0.63,0.237
    review:
      alerts:
        required_zones: Vialetto2
      detections:
        required_zones: Vialetto2
    objects:
      filters:
        car:
          mask:
            - 0.344,0.519,0.367,0.036,0.558,0.051,0.626,0.517
            - 0.663,0,0.663,0.147,0.851,0.183,0.876,0
  BagnoStudio:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/BagnoStudio?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/BagnoStudio_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 1
    motion:
      mask:
        - 0,0.09,0.18,0.089,0.18,0.048,0.305,0.045,0.307,0,0,0
        - 0.708,1,0.671,0.813,0.543,0.673,0.562,0.195,0.536,0.199,0.201,0.197,0.202,0,1,0,1,1
      threshold: 60
      contour_area: 20
      improve_contrast: 'true'
    objects:
      filters:
        person:
          mask: 559,261,533,109
  CameraOspiti:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/CameraOspiti?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/CameraOspiti_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 2
    motion:
      mask:
        - 0,0.09,0.196,0.085,0.196,0.05,0.305,0.045,0.307,0,0,0
        - 0,0.482,0.232,0.492,0.217,0.127,0.273,0.207,0.423,0.205,0.469,0.102,0.469,0,0,0
      threshold: 70
      contour_area: 10
      improve_contrast: 'true'
  GiardinoPosteriore:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/GiardinoPosteriore?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/GiardinoPosteriore_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 3
    motion:
      mask:
        - 0,0.09,0.29,0.087,0.291,0.045,0.305,0.045,0.307,0,0,0
        - 0,0.742,0.255,0.745,0.322,0.728,0.353,0.683,0.367,0.651,0.43,0.589,0.466,0.519,0.479,0.441,0.421,0.306,0.428,0.217,0.433,0.175,0.42,0.098,0.395,0,0.181,0,0,0,0,0.104,0,0.482,0.143,0.44,0.181,0.317,0.113,0.532,0,0.634
        - 0.652,0.76,0.642,0.606,0.647,0.509,0.666,0.441,0.746,0.416,0.816,0.474,0.837,0.516,0.852,0.574,0.856,0.663,0.83,0.736,0.752,0.793,0.684,0.801
        - 0.454,0.277,0.491,0.277,0.5,0.22,0.465,0.174,0.438,0.192,0.432,0.265
        - 0.648,0.287,0.623,0.275,0.621,0.22,0.629,0.162,0.647,0.132,0.67,0.104,0.698,0.092,0.853,0.115,0.869,0.022,0.957,0,1,0.035,1,0.454,0.788,0.376,0.669,0.302
        - 0.435,0.045,0.694,0.088,0.623,0.187,0.438,0.164
        - 0.016,0.856,0.027,0.705,0.1,0.684,0.1,1,0,1
      threshold: 50
      contour_area: 20
      improve_contrast: 'true'
    objects:
      filters:
        motorcycle:
          mask: 0.43,1,0.956,0.703,0.911,0.638,0.9,0.583,1,0.509,1,1
        car:
          mask: 0.423,1,0.948,0.698,0.91,0.629,0.903,0.581,1,0.513,1,1
  Ripostiglio:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Ripostiglio?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Ripostiglio_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 4
    motion:
      mask:
        - 0,0.09,0.184,0.085,0.183,0.043,0.305,0.045,0.307,0,0,0
        - 0.388,1,0.412,0.832,0.438,0.636,0.438,0.526,0.412,0.342,0.4,0.232,0.353,0.227,0.377,0.115,0.388,0.085,0.409,0,0.359,0,0,0,0,1
        - 0.559,0,0.735,0,0.721,0.2,0.555,0.195
      threshold: 40
      contour_area: 10
      improve_contrast: 'true'
  ParcheggioMoto:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/ParcheggioMoto?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/ParcheggioMoto_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 5
    motion:
      mask:
        - 0,0.09,0.164,0.097,0.165,0.045,0.305,0.045,0.307,0,0,0
        - 0.468,1,0.468,0.344,0.551,0.073,0.75,0.003,1,0,1,1
      threshold: 80
      contour_area: 20
      improve_contrast: 'true'
    objects:
      filters:
        motorcycle:
          mask: 0.25,1,0.252,0.234,0.467,0.226,0.468,1
        car:
          mask: 0.459,0.05,0.608,0.083,0.611,0.215,0.446,0.209
  Laghetto:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Laghetto?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Laghetto_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 6
    motion:
      mask:
        - 0,0.09,0.136,0.089,0.134,0.047,0.305,0.045,0.307,0,0,0
        - 0.249,0.229,0.353,0.285,0.403,0.245,0.416,0.205,0.444,0.212,0.52,0.232,0.562,0.105,0.55,0.052,0.463,0,0.218,0
        - 0.688,0.344,0.707,0.165,0.771,0.187,0.825,0.063,0.852,0.01,0.98,0,0.992,0.047,1,0.145,0.998,0.394,0.861,0.342
      threshold: 40
      contour_area: 25
      improve_contrast: 'true'
    zones:
      Prato:
        coordinates: 0.255,0.261,0.263,0.247,0.408,0.245,0.418,0.215,0.547,0.159,0.562,0.12,0.623,0.13,0.701,0.182,0.678,0.339,0.847,0.369,1,0.461,1,0.578,0.892,0.629,0.638,0.287
        inertia: 3
      Vialetto:
        coordinates: 0.004,0.098,0.139,0.097,0.14,0.057,0.219,0.053,0.265,0.234,0.088,0.656,0.061,0.472,0.004,0.466
        objects:
          - car
      Piazzale:
        coordinates: 0.089,0.679,0.251,0.263,0.636,0.292,0.92,0.673,0.121,0.718,0.086,0.735
        inertia: 3
    review:
      alerts:
        required_zones: Prato
      detections:
        required_zones: Prato
    objects:
      mask:
        - 0,0.888,0.128,0.88,0.15,0.765,0.085,0.74,0.075,0.594,0.057,0.482,0.007,0.486
        - 0.26,1,0.267,0.684,0.813,0.694,0.823,1
  Capanno:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Capanno?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Capanno_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
      hwaccel_args: preset-nvidia-h264
    snapshots: # <----- Enable snapshots
      timestamp: true
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 7
  Veranda:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Veranda?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Veranda_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
      hwaccel_args: preset-nvidia-h264
      #    detect:
      #      enabled: False # <---- disable detection until you have a working camera feed
    snapshots: # <----- Enable snapshots
      timestamp: true
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 8
  Studio:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Studio?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Studio_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
      hwaccel_args: preset-nvidia-h264
      #    detect:
      #      enabled: False # <---- disable detection until you have a working camera feed
    snapshots: # <----- Enable snapshots
      timestamp: true
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 9
    objects:
      filters:
        car:
          mask: 0.509,0.203,0.514,0.015,0.66,0.038,0.652,0.242
  Cucina:
    enabled: true # <------ True by default
    ffmpeg:
      inputs:
        - path: rtsp:https://127.0.0.1:8554/Cucina?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - record
        - path: rtsp:https://127.0.0.1:8554/Cucina_SD?video=all&audio=all # <--- the name here must match the name of the camera in restream
          roles:
            - detect
            - audio
      hwaccel_args: preset-nvidia-h264
      #    detect:
      #      enabled: False # <---- disable detection until you have a working camera feed
    ui:
      # Optional: Adjust sort order of cameras in the UI. Larger numbers come later (default: shown below)
      # By default the cameras are sorted alphabetically.
      order: 10
camera_groups:
  Interne:
    order: 1
    icon: LuWebcam
    cameras:
      - Veranda
      - Studio
      - Cucina
version: 0.14

Relevant log output

2024-06-17 10:24:21.844145441  [INFO] Preparing Frigate...
2024-06-17 10:24:21.869627860  [INFO] Starting Frigate...
2024-06-17 10:24:23.978553624  [2024-06-17 10:24:23] frigate.app                    INFO    : Starting Frigate (0.14.0-4e5a6eb)
2024-06-17 10:24:23.979157879  [2024-06-17 10:24:23] frigate.util.config            INFO    : Checking if frigate config needs migration...
2024-06-17 10:24:24.094822717  [2024-06-17 10:24:24] frigate.util.config            INFO    : frigate config does not need migration...
2024-06-17 10:24:42.985207905  [2024-06-17 10:24:42] peewee_migrate.logs            INFO    : Starting migrations
2024-06-17 10:24:42.985279181  [2024-06-17 10:24:42] peewee_migrate.logs            INFO    : There is nothing to migrate
2024-06-17 10:24:43.017669123  [2024-06-17 10:24:43] frigate.app                    INFO    : Recording process started: 462
2024-06-17 10:24:43.037000838  [2024-06-17 10:24:43] frigate.app                    INFO    : Recording process started: 466
2024-06-17 10:24:43.044144148  [2024-06-17 10:24:43] frigate.app                    INFO    : go2rtc process pid: 107
2024-06-17 10:24:43.110379790  [2024-06-17 10:24:43] frigate.app                    INFO    : Output process started: 517
2024-06-17 10:24:43.141578901  Process output_processor:
2024-06-17 10:24:43.149103513  Traceback (most recent call last):
2024-06-17 10:24:43.149133371    File "/usr/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2024-06-17 10:24:43.149134939      self.run()
2024-06-17 10:24:43.149139486    File "/usr/lib/python3.9/multiprocessing/process.py", line 108, in run
2024-06-17 10:24:43.149141752      self._target(*self._args, **self._kwargs)
2024-06-17 10:24:43.149149612    File "/opt/frigate/frigate/output/output.py", line 68, in output_frames
2024-06-17 10:24:43.149150929      move_preview_frames("cache")
2024-06-17 10:24:43.149152190    File "/opt/frigate/frigate/output/output.py", line 183, in move_preview_frames
2024-06-17 10:24:43.149153233      shutil.move(preview_holdover, preview_cache)
2024-06-17 10:24:43.149154307    File "/usr/lib/python3.9/shutil.py", line 804, in move
2024-06-17 10:24:43.149172726      raise Error("Destination path '%s' already exists" % real_dst)
2024-06-17 10:24:43.149181828  shutil.Error: Destination path '/tmp/cache/preview_frames/preview_restart_cache' already exists
2024-06-17 10:24:43.161671522  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Parcheggio: 564
2024-06-17 10:24:43.179649771  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for BagnoStudio: 573
2024-06-17 10:24:43.185640468  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for CameraOspiti: 580
2024-06-17 10:24:43.200110576  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for GiardinoPosteriore: 602
2024-06-17 10:24:43.214535974  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Ripostiglio: 611
2024-06-17 10:24:43.230343750  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for ParcheggioMoto: 635
2024-06-17 10:24:43.244787264  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Laghetto: 649
2024-06-17 10:24:43.258607164  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Capanno: 659
2024-06-17 10:24:43.270289761  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Veranda: 680
2024-06-17 10:24:43.283618182  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Studio: 696
2024-06-17 10:24:43.287074035  [2024-06-17 10:24:43] detector.tensorrt              INFO    : Starting detection process: 507
2024-06-17 10:24:43.300018309  [2024-06-17 10:24:43] frigate.app                    INFO    : Camera processor started for Cucina: 709
2024-06-17 10:24:43.316374873  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Parcheggio: 717
2024-06-17 10:24:43.331947659  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for BagnoStudio: 722
2024-06-17 10:24:43.343042550  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for CameraOspiti: 726
2024-06-17 10:24:43.368749998  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for GiardinoPosteriore: 733
2024-06-17 10:24:43.401780099  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Ripostiglio: 739
2024-06-17 10:24:43.404773893  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for ParcheggioMoto: 745
2024-06-17 10:24:43.417245373  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Laghetto: 750
2024-06-17 10:24:43.462824593  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Capanno: 759
2024-06-17 10:24:43.462830808  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Veranda: 763
2024-06-17 10:24:43.518818779  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Studio: 773
2024-06-17 10:24:43.518825354  [2024-06-17 10:24:43] frigate.app                    INFO    : Capture process started for Cucina: 779
2024-06-17 10:24:43.519387936  [2024-06-17 10:24:43] frigate.app                    INFO    : Audio process started: 788
2024-06-17 10:24:43.604250930  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : Loaded engine size: 73 MiB
2024-06-17 10:24:43.785173953  [2024-06-17 10:24:43] 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.
2024-06-17 10:24:43.900039171  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +6, GPU +8, now: CPU 229, GPU 208 (MiB)
2024-06-17 10:24:43.948557513  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] Init cuDNN: CPU +2, GPU +16, now: CPU 231, GPU 228 (MiB)
2024-06-17 10:24:43.967172965  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +72, now: CPU 0, GPU 72 (MiB)
2024-06-17 10:24:43.981601654  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +8, now: CPU 158, GPU 220 (MiB)
2024-06-17 10:24:43.987311150  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] Init cuDNN: CPU +0, GPU +8, now: CPU 158, GPU 228 (MiB)
2024-06-17 10:24:43.996415650  [2024-06-17 10:24:43] frigate.detectors.plugins.tensorrt INFO    : [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +16, now: CPU 0, GPU 88 (MiB)

Operating system

Proxmox

Install method

Docker Compose

Network connection

Wired

Camera make and model

Misecu

Any other information that may be helpful

Seems that restart command won't clear the directory /tmp/cache/preview_frames/preview_restart_cache, but i'm not sure if /tmp/cache/preview_frames/preview_restart_cache is related with /dev/shm

[blakeblackshear]

It is. This is happening because the output process has failed and isn't able to clear the frames in /dev/shm.

[wes1993]

Perfect, I hope you could reproduce the issue :-D

[h2rSys]

I was having same problems when updating to 14 beta... I was restarting frigate every hour...using about 8 gb of shm.. But I think I have a clue... I disabled birdseye and now I only need to restart every day... Maybe is something related...

[Samywamy10]

Am having this same issue. I disabled birdseye and still getting the issue. I can fix it by clearing the cache within container eg

docker exec -it frigate bash
rm -rf /tmp/cache