Aer

Getting Jellyfin VA-API to work with my APU again... I thought it'd be handy to offload some of the transcoding work so tried using a dedicated nvidia GPU (970gtx) So changed the container fields to have my GPU ID and the startup flags --runtime=nvidia to "extra parameters" input my device id etcc but found it didn't have enough hardware acceleration for modern codecs. Now I want to return to using my APU, i have the same flags as before: --device=/dev/dri I have a AMD Ryzen 3 3200G with Radeon Vega Graphics this was working fine before but jellyfin seems to fail transcoding, if I run root@Tower:/# /usr/lib/jellyfin-ffmpeg/vainfo --display drm --device /dev/dri/renderD128 Trying display: drm libva info: VA-API version 1.22.0 libva info: Trying to open /usr/lib/jellyfin-ffmpeg/lib/dri/nvidia_drv_video.so libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/nvidia_drv_video.so libva info: Trying to open /usr/lib/dri/nvidia_drv_video.so libva info: Trying to open /usr/local/lib/dri/nvidia_drv_video.so libva info: va_openDriver() returns -1 vaInitialize failed with error code -1 (unknown libva error),exit it shows nvidia_drv_video, do you tihnk this is whats causing it? Does anyone know how to get VA-API to use the AMD APU?

There is on invokeAI (option 3) under settings enable nsfw checker.

Update has worked. Old parameters.txt has been renamed and the yaml file has transferred with the settings. Thank you

Yes mine automatically updated after restarting the container too. If you have any luck getting in to Invoke-AI let me know.

I have been using invoke-ai It seemed it updated to version 4.0 upon startup. Now it's not launching and keep looping, i tried to find the command that is "not found" and edit the parameters.txt but am not sure what it's referring to. I've included a screenshot, anyone else had this?

Thanks for the suggestion. I've been trying to configure through nginx reverse proxy, it is proving difficult though, I'm using a cloud flare tunnel unfortunately now I do not have access to port forwarding. I did have nginx setup before the tunnel working fine. I've just got open public access to Invoke AI right now. 🤦🏻‍♂️

Is there a way to add a HTTP Basic Login, maybe with a bit of modifying? Currently using Invoke AI UI.

Thank you! Working with 970 GTX 4GB (3.5GB) so far, I doubt all features will. Output:

If I uninstall the plugin will it fully revert the changes so I can use the gpu in a vm?

Sorry I should have explained better. The share tab was showing 2 shares appdata and media for with 7.04tb free which is disk1 (in the first shares screenshot 8tb drive). Since then I've rebooted and all the shares have reappeared and on the new screenshot added to the previous post the right shares and drives are being shown. I'm not sure why this happened upon the first reboot after the update. So no problem now. Thanks

Just updated, can not see files in shares Also no files visible in the cache as well as appdata so docker images can not find their data. Did I do something before the update to cause this (files still in disk1) removed a directory in (cache/media/data) How can I setup the shares and get the files from disk1 to appear in the cache shares? Edit (if it happens to anyone else): I'm not sure what caused this, after seeing all files still there in disk1. I ended up rebooting once more and everything reappeared.

Wrote this last year as I didn't see any guides. http://www.rediscussed.com/transcoding-on-jellyfin-docker-image-on-an-unraid-os-using-an-amd-3200g-or-3400g-apu/

I finally got vaapi transcoding working for the AMD 3200G on unraid for this docker image of jellyfin. Thanks for the image. If anyone needs to know how I can write up what I did. APU usage on vaapi transcoding a 4k video between 10-20%. This is very good for a small processor. Previously without hardware accelerated transcoding 4 threads on 90% (software)

+1 for this, just set up my first unraid system. Couldn't believe there wasn't an easy way to take/manage snapshots. I heard good things about unraid so decided to buy and setup, I thought this was standard for most NAS. 😬 My fault for not checking. Luckily at work I've used zfs with snapshots - on another well known platform. For 40+ users using one share there's always someone that'll accidentally delete a file or two.