July 17, 20242 yr 24 minutes ago, yayitazale said: I don't know how are you launching the container, but if it is found in unraid, you just need to pass-through it using the proper entry in the template. I have using this method for a long time , and have not had to pass-through before . I hav ejust noticed that i dont have the option to edit the template any more , just another problem. I may have to reinstall from an old backup?
July 18, 20242 yr Author On 7/17/2024 at 12:15 PM, mikey6283 said: I have using this method for a long time , and have not had to pass-through before . I hav ejust noticed that i dont have the option to edit the template any more , just another problem. I may have to reinstall from an old backup? With such little info I can't help you. I think maybe you need to reinstall the app completely from scratch
July 19, 20242 yr On 7/17/2024 at 3:15 AM, mikey6283 said: I have using this method for a long time , and have not had to pass-through before . I hav ejust noticed that i dont have the option to edit the template any more , just another problem. I may have to reinstall from an old backup? You have to install the drivers for the PCIe version I believe. For the template, you can see if it was saved under /boot/config/plugins/dockerman/user-templates
July 22, 20242 yr I tried to install frigate as docker in unraid bu always get that error message: Pulling image: ghcr.io/blakeblackshear/frigate:stable TOTAL DATA PULLED: 0 B Command:root@localhost:# /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker run -d --name='frigate' --net='bridge' --privileged=true -e TZ="America/New_York" -e HOST_OS="unRAID" -e 'FRIGATE_RTSP_PASSWORD'='XXXXXXX' -e 'PLUS_API_KEY'='' -e 'LIBVA_DRIVER_NAME'='iHD' -e 'NVIDIA_VISIBLE_DEVICES'='' -e 'NVIDIA_DRIVER_CAPABILITIES'='' -e 'YOLO_MODELS'='' -e 'USE_FP16'='' -e 'TRT_MODEL_PREP_DEVICE'='' -p '5000:5000/tcp' -p '8554:8554/tcp' -p '8555:8555/tcp' -p '8555:8555/udp' -p '1984:1984/tcp' -v '/mnt/user/appdata/frigate':'/config':'rw' -v '/mnt/user/Media/frigate':'/media/frigate':'rw' -v '/etc/localtime':'/etc/localtime':'rw' --device='/dev/bus/usb' --device='' --shm-size=256mb --mount type=tmpfs,target=/tmp/cache,tmpfs-size=1000000000 --restart unless-stopped 'ghcr.io/blakeblackshear/frigate:stable' invalid argument "" for --device=: bad format for path: See 'docker run --help'. The command failed.
July 22, 20242 yr I Tried this command: root@SERVER-DAN:~# docker pull ghcr.io/blakeblackshear/frigate:stable Error response from daemon: manifest for ghcr.io/blakeblackshear/frigate:stable not found I have internet access as I can ping ghcr.io root@SERVER-DAN:~# ping ghcr.io PING ghcr.io (140.82.114.34) 56(84) bytes of data. 64 bytes from lb-140-82-114-34-iad.github.com (140.82.114.34): icmp_seq=1 ttl=44 time=33.4 ms 64 bytes from lb-140-82-114-34-iad.github.com (140.82.114.34): icmp_seq=2 ttl=44 time=33.5 ms 64 bytes from lb-140-82-114-34-iad.github.com (140.82.114.34): icmp_seq=3 ttl=44 time=33.3 ms ^C --- ghcr.io ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 2003ms rtt min/avg/max/mdev = 33.391/33.453/33.530/0.160 ms
July 22, 20242 yr This argument is malformed: --device='' See if you have either extra parameters that are incorrect. or extra container devices that are incorrect:
July 22, 20242 yr Author 3 hours ago, lecrote said: I tried to install frigate as docker in unraid bu always get that error message: Pulling image: ghcr.io/blakeblackshear/frigate:stable TOTAL DATA PULLED: 0 B Command:root@localhost:# /usr/local/emhttp/plugins/dynamix.docker.manager/scripts/docker run -d --name='frigate' --net='bridge' --privileged=true -e TZ="America/New_York" -e HOST_OS="unRAID" -e 'FRIGATE_RTSP_PASSWORD'='XXXXXXX' -e 'PLUS_API_KEY'='' -e 'LIBVA_DRIVER_NAME'='iHD' -e 'NVIDIA_VISIBLE_DEVICES'='' -e 'NVIDIA_DRIVER_CAPABILITIES'='' -e 'YOLO_MODELS'='' -e 'USE_FP16'='' -e 'TRT_MODEL_PREP_DEVICE'='' -p '5000:5000/tcp' -p '8554:8554/tcp' -p '8555:8555/tcp' -p '8555:8555/udp' -p '1984:1984/tcp' -v '/mnt/user/appdata/frigate':'/config':'rw' -v '/mnt/user/Media/frigate':'/media/frigate':'rw' -v '/etc/localtime':'/etc/localtime':'rw' --device='/dev/bus/usb' --device='' --shm-size=256mb --mount type=tmpfs,target=/tmp/cache,tmpfs-size=1000000000 --restart unless-stopped 'ghcr.io/blakeblackshear/frigate:stable' invalid argument "" for --device=: bad format for path: See 'docker run --help'. The command failed. Hi, I don't mean to sound disrespectful but please, if you have had the time to post on the forum, you also had the time to read the instructions that I have added to the template with my time and effort. The problem is not that you don't pull from ghcr, the problem is that you have empty template entries which can't be there empty. The combinations to use frigate are many, so the only way to have a global template that works for everyone is to use only the entries you need and delete the rest.
July 22, 20242 yr On 7/18/2024 at 7:14 PM, yayitazale said: With such little info I can't help you. I think maybe you need to reinstall the app completely from scratch I am still have problems with the TPU. It is found by unraid and HA app and bond. Howerver now it is not found by the Coral drivers any ideaa Logs showing error in container
July 22, 20242 yr 5 hours ago, yayitazale said: Hi, I don't mean to sound disrespectful but please, if you have had the time to post on the forum, you also had the time to read the instructions that I have added to the template with my time and effort. The problem is not that you don't pull from ghcr, the problem is that you have empty template entries which can't be there empty. The combinations to use frigate are many, so the only way to have a global template that works for everyone is to use only the entries you need and delete the rest. Hi, there is no disrespectful here, I'm not an expert in docker and these time I found myself like a stupid guys. I really respect and enjoy the work you have done and all others people that do things like you for us. to answer your question yes I read the instruction. In the instruction the only thing I found is that. "Unraid Many people have powerful enough NAS devices or home servers to also run docker. There is a Unraid Community App. To install make sure you have the community app plugin here. Then search for "Frigate" in the apps section within Unraid - you can see the online store here" Did I miss something? I did install it on a second unraid and I did not had that problem. not finished to understand all the setup but at least the docker is running. This is on my main Unraid that is giving me headache. Any help is always really appreciate Regards
July 22, 20242 yr 11 hours ago, lecrote said: Did I miss something? The instructions are literally in the template as pointed out by foo_fighter. Not always apparent especially for people where english is a second language but they are there. If you are not populating the parameters, remove/delete them or it will cause an error.
July 23, 20242 yr Author 22 hours ago, mikey6283 said: I am still have problems with the TPU. It is found by unraid and HA app and bond. Howerver now it is not found by the Coral drivers any ideaa Logs showing error in container 6.97 kB · 3 downloads I don't know what are you doing with HA here. Witch TPU model do you have? How and where are you bonding it?
July 23, 20242 yr 21 minutes ago, yayitazale said: I don't know what are you doing with HA here. Witch TPU model do you have? How and where are you bonding it? I have a m2 pcie coral card which has been working fine for over 2 years. I reinstalled Frigate due to so many errors ffmpeg. The Coral is detected in unRaid . So i had to bind the pcie card so it can be seen in HA to work. Now the Coral drivers cannot find it. So i am not sure is it is faulty or not.
July 23, 20242 yr Author 8 minutes ago, mikey6283 said: I have a m2 pcie coral card which has been working fine for over 2 years. I reinstalled Frigate due to so many errors ffmpeg. The Coral is detected in unRaid . So i had to bind the pcie card so it can be seen in HA to work. Now the Coral drivers cannot find it. So i am not sure is it is faulty or not. If you bind a device to a VM, of course the host is not going to be able to use it anymore so that's why the driver is not finding it. Why are you binding it? To use it with Frigate container template you don't need to bind it. If you are using it inside a VM then I can't help you, this is out of my knowledge and is not the scope of this threat.
July 23, 20242 yr 5 minutes ago, yayitazale said: If you bind a device to a VM, of course the host is not going to be able to use it anymore so that's why the driver is not finding it. Why are you binding it? To use it with Frigate container template you don't need to bind it. If you are using it inside a VM then I can't help you, this is out of my knowledge and is not the scope of this threat. Ok thanks , I will unbind it and try again. I appreciate your help.
July 23, 20242 yr 2 hours ago, yayitazale said: If you bind a device to a VM, of course the host is not going to be able to use it anymore so that's why the driver is not finding it. Why are you binding it? To use it with Frigate container template you don't need to bind it. If you are using it inside a VM then I can't help you, this is out of my knowledge and is not the scope of this threat. Canyou point me in the way forward to solve this error , this why i went down the wrong road by binding the coral. 024-07-23 09:47:45.910646578 [2024-07-23 10:47:45] frigate.app INFO : Output process started: 245 2024-07-23 09:47:45.910650735 [2024-07-23 10:47:45] detector.cora1 INFO : Starting detection process: 243 2024-07-23 09:47:45.917718804 [2024-07-23 10:47:45] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-23 09:47:45.918440266 [2024-07-23 10:47:45] frigate.detectors.plugins.edgetpu_tfl ERROR : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors. 2024-07-23 09:47:45.918631113 Process detector:cora1: 2024-07-23 09:47:45.923071054 Traceback (most recent call last):
July 24, 20242 yr Author 21 hours ago, mikey6283 said: Canyou point me in the way forward to solve this error , this why i went down the wrong road by binding the coral. 024-07-23 09:47:45.910646578 [2024-07-23 10:47:45] frigate.app INFO : Output process started: 245 2024-07-23 09:47:45.910650735 [2024-07-23 10:47:45] detector.cora1 INFO : Starting detection process: 243 2024-07-23 09:47:45.917718804 [2024-07-23 10:47:45] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-23 09:47:45.918440266 [2024-07-23 10:47:45] frigate.detectors.plugins.edgetpu_tfl ERROR : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors. 2024-07-23 09:47:45.918631113 Process detector:cora1: 2024-07-23 09:47:45.923071054 Traceback (most recent call last): First of all, is the coral driver finding the coral? Then, can you show how are filling out the template?
July 24, 20242 yr 5 minutes ago, yayitazale said: First of all, is the coral driver finding the coral? Then, can you show how are filling out the template? thanks for reply Yes drivers are found . I think that thet the Frigate addon may be at faullt as it has to rbe reactivated when you click on it.
July 24, 20242 yr Author 55 minutes ago, mikey6283 said: thanks for reply Yes drivers are found . I think that thet the Frigate addon may be at faullt as it has to rbe reactivated when you click on it. Can you post the image of the "Coral Accelerator Module Drivers" plugin? Edited July 24, 20242 yr by yayitazale
July 24, 20242 yr 21 minutes ago, yayitazale said: Can you post the image of the "Coral Accelerator Module Drivers" plugin?
July 24, 20242 yr I know that you all are talking abut a docker version. But does anyone know how to get the Google Coral TPU pci version to show up on a VM i like running it as a VM so what or HOW. Can anyone help. I know that i might have been talked about here on UNraid forums but i have not found it yet 🙃 Thanks John G.
July 24, 20242 yr 3 hours ago, John Graham said: I know that you all are talking abut a docker version. Which is the scope of this thread. More specifically blakeblackshear - Frigate. I suggest you post in a VM discussion about passing through devices. Note, The Frigate documention advises against running Frigate in a VM
July 25, 20242 yr ok will i said i was looking for help ... ok then i stoped my VM and loaded the docker version(s) and i have not found one with Beta or even superviser installed any i missing some this thanks ( Dont yell i am trying ) John G
July 26, 20242 yr I have been using Frigate for a little while now and it is working ok but i think I have room for improvement. Can anyone see anything I could change / optimize? mqtt: enabled: true host: 192.168.XX.XX port: 1883 user: XXXXXXXX password: XXXXXXX topic_prefix: XXXXXXXXXX client_id: XXXXXXXXXXX stats_interval: 60 ##------------------------------------------------------------------------------------------------------------------------------------------------------ detectors: coral1: type: edgetpu device: pci:0 coral2: type: edgetpu device: pci:1 ##------------------------------------------------------------------------------------------------------------------------------------------------------ database: path: /config/frigate.db ##------------------------------------------------------------------------------------------------------------------------------------------------------ auth: enabled: true reset_admin_password: false #cookie_name: frigate_token cookie_secure: false session_length: 86400 # 24 hours refresh_time: 43200 # 12 hours failed_login_rate_limit: None # trusted_proxies: [https://XXXXXXX.XXXXXXXXXX.XXX] hash_iterations: 600000 ##------------------------------------------------------------------------------------------------------------------------------------------------------ model: path: /edgetpu_model.tflite labelmap_path: /labelmap.txt width: 320 height: 320 input_pixel_format: rgb input_tensor: nhwc model_type: ssd labelmap: 1: person # 2: car ##------------------------------------------------------------------------------------------------------------------------------------------------------ birdseye: enabled: true mode: continuous quality: 5 restream: true width: 1920 height: 1080 ##------------------------------------------------------------------------------------------------------------------------------------------------------ ffmpeg: global_args: -hide_banner -loglevel warning -threads 2 hwaccel_args: auto input_args: preset-rtsp-generic retry_interval: 10 ##------------------------------------------------------------------------------------------------------------------------------------------------------ detect: width: 1920 height: 1080 fps: 15 enabled: true min_initialized: 2 max_disappeared: 50 stationary: interval: 50 threshold: 50 ##------------------------------------------------------------------------------------------------------------------------------------------------------ objects: track: - person #- fox #- car #- dog #- cat filters: person: # Optional: minimum width*height of the bounding box for the detected object (default: 0) min_area: 2000 # 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 # Optional: maximum width/height of the bounding box for the detected object (default: 24000000) #max_ratio: 0.8 # Optional: minimum score for the object to initiate tracking (default: shown below) min_score: 0.50 # Optional: minimum decimal percentage for tracked object's computed score to be considered a true positive (default: shown below) threshold: 0.6 # car: # threshold: 0.7 # fox: # threshold: 0.5 # cat: # threshold: 0.5 # dog: # threshold: 0.5 ##------------------------------------------------------------------------------------------------------------------------------------------------------ review: alerts: labels: # - car - person # required_zones: # - Turret-Front # - PTZ-Front detections: labels: # - car - person # required_zones: # - Turret-Front # - PTZ-Front ##------------------------------------------------------------------------------------------------------------------------------------------------------ motion: threshold: 20 lightning_threshold: 0.8 contour_area: 10 frame_alpha: 0.01 frame_height: 100 improve_contrast: true ##------------------------------------------------------------------------------------------------------------------------------------------------------ record: #sync_recordings: True enabled: true retain: days: 7 mode: all preview: quality: medium export: timelapse_args: -vf setpts=0.04*PTS -r 30 events: pre_capture: 5 post_capture: 5 objects: - person retain: default: 7 mode: motion objects: person: 7 ##------------------------------------------------------------------------------------------------------------------------------------------------------ snapshots: enabled: true clean_copy: true timestamp: true bounding_box: true crop: false height: 1080 retain: default: 7 objects: person: 7 quality: 80 # required_zones: # - Turret-Front # - PTZ-Front ##------------------------------------------------------------------------------------------------------------------------------------------------------ go2rtc: streams: Turret: - ffmpeg:rtsp://192.168.XX.XX:554/11 PTZ: - ffmpeg:rtsp://192.168.XX.XX:554/12 ##------------------------------------------------------------------------------------------------------------------------------------------------------ cameras: Turret: ffmpeg: output_args: record: preset-record-generic-audio-aac inputs: - path: rtsp://192.168.XX.XX:554/11 input_args: preset-rtsp-restream roles: - detect - record - rtmp motion: threshold: 30 contour_area: 10 improve_contrast: 'true' mask: 0.752,0.04,1,0.044,1,0,0.752,0 objects: filters: car: {} mask: 0,0,0,0.28,0.271,0.173,0.684,0.118,0.905,0.203,1,0.346,1,0 zones: Turret-Front: coordinates: 0.284,0.178,0.093,0.235,0.001,0.286,0,1,1,1,1,0.354,0.903,0.209,0.685,0.127 mqtt: enabled: true timestamp: true bounding_box: true crop: true height: 270 quality: 70 ##------------------------------------------------------------------------------------------------------------------------------------------------------ PTZ: ffmpeg: output_args: record: preset-record-generic-audio-aac inputs: - path: rtsp://192.168.XX.XX:554/12 input_args: preset-rtsp-restream roles: - detect - record - rtmp motion: threshold: 30 contour_area: 10 improve_contrast: 'true' mask: 0.754,0,0.755,0.038,1,0.038,1,0 zones: PTZ-Front: coordinates: 0.074,0.099,0.085,0.305,0.122,0.636,0.177,1,1,1,1,0.124,0.735,0.087,0.728,0.036,0.564,0.024,0.352,0.036 inertia: 3 objects: mask: 0,0,0.072,0.096,0.349,0.034,0.561,0.021,0.729,0.034,0.738,0.084,1,0.121,1,0 mqtt: enabled: true timestamp: true bounding_box: true crop: true height: 270 quality: 70 ##------------------------------------------------------------------------------------------------------------------------------------------------------ live: #stream_name: Turret height: 1080 quality: 10 ##------------------------------------------------------------------------------------------------------------------------------------------------------ telemetry: network_interfaces: - eth - enp - eno - ens - wl - lo # Optional: Configure system stats stats: # Enable AMD GPU stats (default: shown below) #amd_gpu_stats: True # Enable Intel GPU stats (default: shown below) #intel_gpu_stats: True # Enable network bandwidth stats monitoring for camera ffmpeg processes, go2rtc, and object detectors. (default: shown below) # NOTE: The container must either be privileged or have cap_net_admin, cap_net_raw capabilities enabled. network_bandwidth: true # Optional: Enable the latest version outbound check (default: shown below) # NOTE: If you use the HomeAssistant integration, disabling this will prevent it from reporting new versions version_check: true version: 0.14 ##------------------------------------------------------------------------------------------------------------------------------------------------------ logger: # Optional: default log level (default: shown below) default: info # Optional: module by module log level configuration logs: frigate.mqtt: info ##------------------------------------------------------------------------------------------------------------------------------------------------------ camera_groups: All_Camera's: order: 1 icon: LuCamera cameras: - PTZ - Turret - birdseye
July 26, 20242 yr more qustions .. i have move Frigate out of Home Assistant and in to a docker. Looks like it's running very good thanks for the poke in the right direction. now for the question. startedd up and look at the log file and i see this: 05:26:07] detector.coral INFO : Starting detection process: 3051 05:26:07] frigate.app INFO : Output process started: 3053 05:26:07] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 05:26:07] frigate.detectors.plugins.edgetpu_tfl INFO : TPU found 2024-07-26 05:26:07.133535528 05:26:07] frigate.app INFO : Camera processor started for Frount_Door_Cam: 3073 so i see nothing after that .. i see if found it but does that mean it working ? and also i see prople talking about also haveing a GPU is that something i need to add to the system? I know a lot of questions .. sorry more to come 🤓 John G
July 26, 20242 yr Author On 7/24/2024 at 10:46 AM, mikey6283 said: Run the command "ls -l /dev/" in the unraid console and see if you have the apex_0
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.