ich777

Added you to the conversation...

This is something that @limetech has to answer but I think it is maybe SOON™er than you might think, maybe, or not...

Wouldn't this also work with OpenRGB, see:

I've also turned it off... my main PC is enough RGB for my whole life I think...

I would wait a little longer, proper iGPU support for Alder Lake is maybe introduced in 6.11 RC series... The issue with Linux and bleeding edge hardware is always that it works properly because Intel is always a bit behind submitting patches to the Kernel...

A really oversimplified answer would be that this is a VM in a Container with shared CPU and Memory and set up pretty quick, basically a real dedicated Linux environment somewhat like a VM. I run this for some time on my server now for testing reasons but was not quite sure if others are interested in this. Previously I ran a VM to build my Docker containers now I do everything in the LXC container with Docker installed there without wasting HW resources.

But are you sure that the container caused this? I have no issues to run the container with my Intel iGPU with HW transcoding on my main server and with a Nvidia P400 on my test server with HW transcoding. Maybe also your GPU crashed but that's only a guess and I really can't tell what happened. Are you sure that it completely hung if so this it is also possible that after you've finished the setup that the initial media scan and processing from the thumbnails caused a crash because for whatever reason that I really can't tell. Maybe try to only add a single path to the container with only one file in it an see if this is working. As said above, I have no issue on my main Server and on my test Server running the container one with a Intel iGPU and one with a Nvidia P400 through NVENC.

What features would be needed to start with, would be command line only good enough to start with and maybe a overview page from the config files?

This is just the default behavior for PhotoPrism, you can change the quality and set a fixed size for thumbnails with this Variables in the Docker template (Source) : +----------------------------------------+ | KEY | VALUE | +----------------------------------------+ | PHOTOPRISM_THUMB_UNCACHED | true | | PHOTOPRISM_THUMB_SIZE | 720 | | PHOTOPRISM_THUMB_SIZE_UNCACHED | 720 | | PHOTOPRISM_JPEG_SIZE | 720 | | PHOTOPRISM_JPEG_QUALITY | 80 | +----------------------------------------+ Of course you can set the thumbnail sizes however you like and even lower the PHOTOPRISM_JPEG_QUALITY even more but would not recommend to go below about 70. For this I would recommend that you create a issue over on GitHub as a feature request: Click No, because PhotoPrism creates different size thumbnails for every picture that you have in your library, it's basically the same when you run Nextcloud... Nextcloud for example creates IIRC 4 or 5 different size thumbnails for saved pictures.

@Bolagnaise, @McTaurus13 & @WenzelComputing I've written a PM to you all with a test package and instructions.

Of course this (deactivation) is needed and already possible, especially for my game server containers… For example if someone runs multiple CS GO servers they have to delete the old port mapping, and create a new port mapping with the config for the game server changed too. The reason behind this is that you easily can‘t translate ports from one to another, for example container:27015/host:27016 won‘t work or better speaking if you do so you can query the server but you can‘t connect because some game servers send information on which port the server runs to make sure it wasn‘t tampered with… Same goes for Valheim and many more. Ultimately when I leave automatic updates turned on all container will break that have custom ports in it because they get readded when CA Update or whatever kicks in. From my perspective it shouldn‘t be updated automatically and the user should go to the forums/repo, look up what changed and fix it themselves. I really don‘t like it that something is fully updated/changed automatically when I changed it by intention, well yes, if something breaks or doesn‘t work properly anymore I have to look up what changed and fix it. I don‘t think that‘s laziness by the users… Just my 2 cents…

This is usually not caused by a container since it doesn‘t has access to the flash drive. Did you maybe upgrade to a newer Unraid version? Does the 970 even support h265? I have to look into this but keep in mind that Jellyfin also supports transcoding to h265 and if this is not supported by your card maybe that‘s the reason why it was not working. Can‘t tell without anything (Screenshots, Diagnostics,…) why ot wasn‘t working. What was the source that you tried transcoding (h264, h265, h265 HDR)?

Exactly.

I have to report that I had to deprecate the container for AssettoCorsa Competizione because I can't get it to run anymore. It kind of works but crashes after a few minutes: SessionManager::randomizeGreenFlagTriggers: s:0.890000 e:0.950000 r:0.939746 Track monza was set and updated Event changed Reading custom event rules FileToStruct cfg/eventRules.json SessionManager::randomizeGreenFlagTriggers: s:0.890000 e:0.950000 r:0.909884 Significant change detected, updating lobby (1|1) Resetting weekend to friday night Reset time to friday night: 0 -> 0 0100:fixme:msvcp:_Locinfo__Locinfo_ctor_cat_cstr (00000000019AF760 1 C) semi-stub 0100:fixme:msvcp:_Locinfo__Locinfo_ctor_cat_cstr (00000000019AF760 1 C) semi-stub 0100:fixme:msvcp:_Locinfo__Locinfo_ctor_cat_cstr (00000000019AF760 1 C) semi-stub Reset time to first session: 0 -> 61200 SessionManager::randomizeGreenFlagTriggers: s:0.890000 e:0.950000 r:0.927759 ==ERR: Couldn't setup UDP broadcasting socket (SO_BROADCAST failed) ==ERR: Couldn't setup UDP broadcasting socket (SO_REUSEADDR failed) Listening to TCP 9232 | UDP 9231 Trying to connect to lobby (0 times, interval 10000 s) Session changed: Practice -> Practice 0 Detected sessionPhase <waiting for drivers> -> <waiting for drivers> (Practice) TCP connect returns: -1 ==ERR: RegisterToLobby TCP connection failed, couldn't connect to the lobby server Trying to connect to lobby (1 times, interval 13000 s) Server was running late for 1 step(s), not enough CPU power TCP connect returns: 0 RegisterToLobby succeeded ==ERR: This server is only supported on Windows operating systems. Running WINE may work dependent on your settings, but there are differences that affect the user's experience negatively. Sent lobby registration request for monza Please read the second last line... I think they broke something in a recent update so that it won't work anymore on Linux through WINE.

If you are on 6.9.2 please read this GitHub issue: Click

Are you already on RC5 and have tried if it is fixed? I think you have to wait a little longer until 6.11 RC series is released which will include a newer Kernel.

Ich glaube hier gibt es keinen richtigen Workaround weil OSX mit der Hardware einfach anders umgeht und wenn so eine Maschine runtergefahren wird glaubt die Karte das sie weiter läuft und resetted nicht mit dem AMD Vendor Reset.

Really curious if those adapters will work too...

What's the exact issue? This is only the containerd Daemon which is needed for Docker to run properly and manage all the containers that are installed on your system, even if you have no containers installed. What you see is only the command and with which parameters it was started and that the log level is set to error, but this is not an error...

This is more or less easy from what I see, you only need a MariaDB and then the container itself. My issue is that I don't think I can get the game here in Europe (only finding results for Lineage II on the used Market) and I don't create templates anymore when I don't actually own the game. From what I see in the compose file you only need to do a port mapping 2000:2000 and mount the /.config folder to a folder on your host and of course create a database in the MariaDB container and connect it afterwards.

I will look into this ASAP, give me a few days, my time is really limited right now.

I will look into this when I got home from work but from what I read in the logs it seems that the server can't connect with the master server, could be either a issue with the last update from the container (because I'm using WINE and things changed a bit) or with the port forwarding. I will report back.

Please read this post:

This is the official container, I've only made the template. I think this is the issue that you've created on Github. To change to the preview branch simply change the repository from "photoprism/photoprism" to "photoprism/photoprism:preview" (without double quotes):

Have you forced a video transcode? Is this maybe only the transcode for the audio stream? To force a transcode play a file, click on the gear icon on the bottom right and select a preset that is lower than the input file (if you play a file with 1080p 720p 1.5Mbit/s will force a transcode for example).