yayitazale

This is my case, with Tapo Cameras with no AAC audio on original feed: First, you have to define the hw accel argumentsfor ffmpeg. In my case, I have this, so I record always copying the original audio of the input of the record role. ffmpeg: hwaccel_args: preset-vaapi input_args: preset-rtsp-restream-low-latency output_args: record: preset-record-generic-audio-copy Then I defined a restream of the camera with a copy of the stream which transcodes audio to AAC: go2rtc: streams: ofice: - rtsp://user:[email protected]:554/stream1 -"ffmpeg:ofice#audio=aac" Finally, I defined the camera with the roles, using the restreamed feed as input for record role, and a direct secondary low resolution feed from the camera for detection: cameras: ofice: ffmpeg: inputs: - path: rtsp://127.0.0.1:8554/ofice?video=copy&audio=aac roles: - record - path: rtsp://user:[email protected]:554/stream2 roles: - detect With this, I record events videos with audio AAC, and I have a RTC stream aviable fo use in Home Assistant with AAC audio as well.

First of all, please give format to your comments and don't paste all the config file in plain text, attach it as a file, same for long logs. Next: you should use presets in the hardware acceleration, please read the docs. Also, you configured birdseye only for objects, so you will only have a image there when any camera has a object detection happening, so that behavior seems can be normal. Last, are you using unraid docker app or home assistant plugin?

You can try to reopen this or create a new issue on github and reference to it https://github.com/blakeblackshear/frigate/issues/5712#event-9071685940 Multi nvidia GPU is not working properly at least on unraid but no one seems to know why.

If you want audio, you should configure the GO2RTC restream and transcode de audio from PCM to AAC for example.

No, you are not. In red all the ones you are not filling and they are still there: docker run -d --name='frigate' --net='bridge' -e TZ="America/Denver" -e HOST_OS="Unraid" -e HOST_HOSTNAME="Unraid" -e HOST_CONTAINERNAME="frigate" -e 'FRIGATE_RTSP_PASSWORD'='XXXXXX -e 'PLUS_API_KEY'='' -e 'LIBVA_DRIVER_NAME'='' -e 'NVIDIA_VISIBLE_DEVICES'='' -e 'NVIDIA_DRIVER_CAPABILITIES'='' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.webui='http://192.168.0.2:5000/' -l net.unraid.docker.icon='https://raw.githubusercontent.com/yayitazale/unraid-templates/main/frigate.png' -p '5000:5000/tcp' -p '8554:8554/tcp' -p '8555:8555/tcp' -p '8555:8555/udp' -p ':1984/tcp' -v '/mnt/user/appdata/frigate':'/config':'rw' -v '/mnt/user/frigate_data/':'/media/frigate':'rw' -v '':'/trt-models':'ro' -v '/etc/localtime':'/etc/localtime':'rw' --device='' --device='' --shm-size=5000mb --mount type=tmpfs,target=/tmp/cache,tmpfs-size=100000000 --restart unless-stopped 'ghcr.io/blakeblackshear/frigate:stable' docker: bad format for path: . Can you show a screenshot of the template before launching it?

Please read the docs and change the inputs of the variables according to your needs. Delete the variables if you are not using them.

You have same IP on both rtsp connection to the camera on the go2rtc part and the restream config inside the camera config. The first one must be the IP of the camera, the second one must be the IP of the unraid server. Please check out, one of them is not correctly configured.

This is not an issue, you can open a discussion in the frigate github page.

Hi, I don't think so. You can open a issue on github if you want to push on this to the main development team. Currently th only way to use the USB coral is to launch the container in privileged mode.

Please read the docs https://docs.frigate.video/configuration/record

The extra parameter field will show up if you enable Advanced view in the upper corner of the configuration page of the containers template. This is something common in unraid there is plenty info on internet about that. You have to add the "--runtime=nvidia" there and not in the config file.

In theory, using the GPUID variable and the extra --runtime=nvidia parameter should work. If there are problems with certain GPUs we can talk with @ich777 to see if the driver is loading everything OK or maybe we have to open a issue on github.

Open the terminal of the container and do nvidia-smi to check if the passthrought of the GPU is correct. There some issues on systems with 2 GPU dedicated cards so maybe this is what is happening to you. In any case, you can check the logs of the container or the logs inside the webUI to see what is wrong. About the HW accel presets, I suggest you to use this for nvidia: ffmpeg: hwaccel_args: preset-nvidia-h264 For case B (I suggest you to first have the case A running well with nvidia GPU), you must select the option 2 at the deployment (nvidia version), but first you must create the models using the same GPU that you are going to use in frigate and the APP from store called "tensorrt-models" (follow the instructions and requirements of the template). And for you info, I just maintain the frigate templates, I'm not a core developer so give the cheers to them on github not to me.

Please read the full documentation. You are using old format hardware accel presets. Are you planning to use nvidia GPU just to HW accel (case A) or/and you want to detect (case B)? If this is case A: Did you install the nvidia driver plugin? Can you see the GPU doing nvidia-smi in the unraid console? Did you passtrought correctly the GPU using the GPUID (taken from the nvidia driver plugin)? Did you see the GPU inside the docker container of frigate doing nvidia-smi? Just in case, are you using the GPU in any VM on unraid? If this is case B , all the questions of the case A plus: Did you select the nvidia branch at the frigate app install? Did you created the models? Please also recheck your config file to meet the the docs: https://docs.frigate.video/

Template updated to v12 This is a breaking change so you MUST uninstall the APP and reinstall it from the store adapting you config file and template inputs. You can read the full changelog here: https://github.com/blakeblackshear/frigate/releases/tag/v0.12.0 And the docs are also updated: https://docs.frigate.video Plase read them entirely before posting for any help here.

Yes, a lot. You can see them in the Github releases https://github.com/blakeblackshear/frigate/releases and documentation link for frigate 12 is already in the overview of the unraid app: https://deploy-preview-4055--frigate-docs.netlify.app/

Can you attatch a diagnostic file?

Try to launch the container in privileged mode.

No. If you define media math to /mnt/user/media/cctv/frigate and you have the /mnt/user/media share set cache:yes the clips and recordings will be saved on cache and then mover will move them to array. Thats the point of having two diferent paths, one for the media and another one for config. The only thing to keep in mind in that setup is that by default, frigate's database will be living on media folder so if you don't change the path in the config file to /config as I said, once the database is moved to array by the mover, the array will never spin down.

I don't have this issues. 32Gb of ram and only 4% of difference between frigate running or not running.

As I said, if you use /mnt/user paths and change the config file to /config you shouldn't see HDD drives to spin up until mover is invoked.

So you just can config the "/mnt/user/cctv" to cache=yes and "/mnt/user/appdata" to cache=prefer and your array will only spin up when mover is invoked or if your cache disk fills up. Another thing you have to do according to the frigate's docs is to define the database's path to the config folder doing this in the config.yml: database: path: /config/frigate.db With that configuration the files under cctv will live only in cache until mover is invoked and it will move them into the array. The files under /appdata will live in the cache foreverif there is enough space. Thats the point of using a cache disk/pool, you will speed up r/w of the disks and don't need to use de HDD array until mover is invoked. This is the standard method proposed in the link I sent you of the documentation on how to use shares in unraid to get the maximum performance out of your system, extend the lifespan of your hard disks and reduce power consumption. I don't know anything about "transfers were slow", I just use this standard configuration since 5 years ago without any issues at all, so I can't help you much more with that.

Use "user cache" path (for example /mnt/user/cctv) instead of disk/pool path, and same for config path. Your mover is moving files from the cache into the array as you have setup that share as "cache:yes" instead of "cache:prefer". I'm not sure that you can achieve that on pools shares so I suggest you to follow the best practices recommended on the unraid manual: https://wiki.unraid.net/Manual/Shares

Please open a issue in github to the main developers as in the docs they only "support" the --gpus=all extra paramateres. https://github.com/blakeblackshear/frigate/issues

It depens how did you configure the container template. By default, without changing the proposed directory, you must create a config.yml file in the "/mnt/user/appdata/antennas". You can do it in several ways, for example: If you have the appdata share shared externally, you can go to "\\IP_OF_YOUR_UNRAID\appdata" in the folder explorer of your pc, create the folder "antennas" and then create the config file. Using the unraid terminal: create the folder doing "mkdir /mnt/user/appdata/antennas" and then creating the file. To create the file, again: If you have the appdata share shared externally, you can go to "\\IP_OF_YOUR_UNRAID\appdata\antennas", right click, create new file, name it with "config.yml" and then edit it using notepad, notepad++, vscode, or whatever code editor you want. Using the unraid terminal: navigate to the created folder doing "cd /mnt/user/appdata/antennas" and then "nano config.yml", it will open a dialog to put the configuration inside de file and then you can save it and close it doing crtl+o > enter > ctrl+x Inside the file you need to insert: tvheadend_url: http://user:pass@IP:port antennas_url: http://IP:port tuner_count: number of tuner In my case for example: tvheadend_url: http://my_user:[email protected]:9981 antennas_url: http://192.168.1.20:5004 tuner_count: 4 I hope you can now understand it.