ich777

Einfach die Daten des alten USB Sticks auf den neuen kopieren, den neuen USB Stick umbennenen das er UNRAID heißt, den alten abziehen, die Datei "make_bootable.bat" als Administrator ausführen und den Anweisungen am Bildschirm folgen. Danach den neuen Stick an den Server, starten und dann musst du eine Lizenz kaufen da er sich beschweren wird das deine GUID nicht zu deiner Testlizenz passt.

Maybe try to disable "Allow encoding in HEVC format" in the Playback section and also try to disable Tone Mapping.

Oh, I forgot that you have to blacklist the iGPU first because you are on 6.10.0-rc4, please execute this from the command line and reboot afterwards: echo "blacklist i915" > /boot/config/modprobe.d/i915.conf Do you've mounted the transcoding directory from Emby to your /tmp directory? If so please don't do this, this can cause OOM errors, I would rather recommend doing it like this:

I find it a bit suspicious that your 5600g doesn't work where a 5700PRO is reported as working fine... If you really going to buy a Nvidia card look out for a Nvidia T400 you can get them brand new for about $130,- and it as a maximum power consumption from 35W and doesn't need external power connection.

Please remove any option or better speaking force_probe, blacklist the i915 module, install only the Intel GPU TOP plugin, the plugin will handle everything for you including the force_probe and reboot. Do you have a monitor connected to the iGPU? If not temporarily connect a monitor to it and try it again.

I see nothing suspicious in there, I only see that it uses this path in the container to put the transcoding files to: /config/transcodes/... Where it should be /cache/transcodes/... But that is only a minor thing since both paths should exist.

Do you have those options put somewhere? Please post your Diagnostics. I have no issues with my 10th gen i5-10600 on 6.8.3, 6.9.0, 6.9.1, 6.9.2, 6.10.0-RC series. Also my test system with a Asrock J3160-ITX has no issue on those versions using the iGPU for transcoding.

What do you mean with this? Why do you want to get rid of Emby's transcoding limit? I thought you mean Jellyfin? If you want to disable throttling in Emby then go to the server settings, at Transcoding disable throttling and click Save. In Jellyfin you can't limit the usage because the develeopers from Jellyfin disabled throttling and there is nothing I can do about that. As said above it would not harm your system when you GPU goes up to 100% here is it what it looks like on my system: Of course I'm using a Intel iGPU and not a APU from AMD but that should make no difference (please look at the Video load which is also at 100%). If you get a OOM error (please tell me where you got that message) there is most certain a different reason for this. Other users are using my container with a AMD APU without any issues, regardless of the GPU usage.

These are two different things. Your GPU is at full load becaus the throttling doesn‘t work in Jellyfin when you are transcoding, there is a bug in it and the developers disabled throttling entierely even if you‘ve enabled it in the settings. This has also the benefit that the whole movie is transcoded in one go and not in little chunks but there are more downsides to this the benefits from my perspective. The Out of Memory error, may I ask where you saw that it ran OOM, is usually caused because the movie is transcoded in one go, at full blast and maybe to some location on RAM. Can you double check if the directory /mnt/user/appdata/Jellyfin-AMD-Intel-Nvidia/cache exists on Unraid? It looks a bit suspicious to me why you are are running out of memory when you are transcoding. You can try to start a transcode with a way lower resolution and at a lower bitrate, in the web player click on the gear Icon on the bottom right, at Quality select something really low like 480p 1.5Mbit/s and look if your RAM fills up again… As said abive throttling is not working in Jellyfin so your GPU usage should be 100% again but as said above Jellyfin transcodes at full blast without any throttling.

Yes, transcoding can consume easily that ammount of RAM. I have 64GB of ram and if I transcode something that is 4K (about 20GB) to a compatible format the transcoding cache can be about 18 GB Can you go to the Jellyfin settings at Playback and can you look what transcoding path is set? If you get a Out of Memory error something must be written to RAM and fill up your entire RAM.

Looks like you are transcoding to RAM. Can you send a screenshot from your template? Have you changed anything in the config or the template? Because in Emby throttling is working where in Jellyfin it isn‘t, so to speak, in Emby only little chunks are encoded where in Jellyfin on the other hand the the full movie is transocded at full blast. Do you have a cache drive installed in your server?

No, but you can try to run it from a terminal from Unraid: HOSTNAME="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "adguard_hostname=" | cut -d '=' -f2-)" USERNAME="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "adguard_username=" | cut -d '=' -f2-)" PASSWORD="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "adguard_password=" | cut -d '=' -f2-)" PROTO="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "adguard_protocol=" | cut -d '=' -f2-)" PORT="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "adguard_port=" | cut -d '=' -f2-)" EXP_PORT="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "exporter_port=" | cut -d '=' -f2-)" INTERV="$(cat /boot/config/plugins/prometheus_adguard_exporter/settings.cfg | grep "exporter_interval=" | cut -d '=' -f2-)" prometheus_adguard_exporter -adguard_hostname ${HOSTNAME} -adguard_port ${PORT} -adguard_protocol ${PROTO} -adguard_username ${USERNAME} -adguard_password ${PASSWORD} -server_port ${EXP_PORT} -interval ${INERV}s Simply copy and paste this and it should start and you should get an error message if something went wrong, make sure that you've stop the exporter on the plugin page first.

Yes, but I won't create a template for this or better speaking a AiO container since I really don't want that Blizzard/Activision (now Microsoft) come after me...

Achte wie gesagt nur darauf das du das Windows 11 template nimst damit TPM funktioniert, dafür musst du aktuell wie gesagt auf der Unraid version 6.10.0-rc4 sein.

Oh, one thing that I maybe can recommend is that if you have a backup from your appdata (CA Backup) that you can try to restore the database from that backup.

To me it seems the Database is corrupt but I really can't tell why this happened... I've now tried a few times to update the container image itself and it always worked perfectly fine both on my test and on my productive machine. Try to create an Issue over on the Jellyfin Github here: Click

Can you try to rename this file to something completely different (don't delete it) and see if this makes a difference? Otherwise I can only recommend that you create a Issue on Github: Click

Can you first try to switch the branch to the RC series like in the attached screenshot: and see if this fixes the issue? I know this error but there are different solutions to resolve this: Click Is there a file called "library.db-journal" in the data subfolder and if yes, how big is it? Is it possible that your Cache is maybe full and the mover moved the file off of the Cache?

Vergiss bitte auch nicht den Access Mode umzustellen wie oben im Screesnhot ersichtlich.

Please navigate to Jellfin the directory located in your appdata folder and do the following form within the jellyfin folder: chown -R 99:100 * chmod -R 777 * (of course when the container is stopped)

Bitte vergiss auch nicht das du auf R/W Slave umstellst:

Ja ist es... Teilweise, willst du die bestehende Installation verwenden? Wenn ja, hast du Bitlocker aktiv? Es könnte evtl. zu komplikationen kommen wegen unterschiedlichem BIOS Typ usw. Du kannst auf jeden Fall die komplette Festplatte durchreichen. Hier mal ein Thread zur Installation: Durchreichen würd ich die Platte ca. so: Du kannst die ID deiner Platte finden indem du eine Terminal Fenster öffenst und: ls /dev/disk/by-id/ eingibst und dann Enter drückst. Dann solltest du eine komplette Liste mit deinen Disks bekommen und dann nimst du die erste ohne irgendwelche -part und irgnorierst das @ am Schluss: und gibst dann im VM template bei der Primary vDisk Location Manuel, gibst den kompletten Pfad an (so wie oben bei mir im Screenshot also /dev/disk/by-id/PLATTENNAME) und Primary vDisk Bus stellst du auf VirtIO. Du musst aktuell auf jeden Fall auf der "next" version von Unraid sein um Windows 11 zu betreiben, keine Angst, die sind wirklich stabil! Ich hab deinen Thread in den Deutschen Bereich verschoben, offe das ist in Ordnung für dich.

Do you have a Cache drive installed in your server if yes do you have a SSD or do you have a HDD as Cache drive? Are you familiar with the Terminal and how you navigate to your Jellyfin appdata folder from the command line? If not, I can help, bu please first post a screenshot from your Docker template from Jellyfin. Greetings from Austria, Christoph

Das ist Meiner Meinung nach viel zu viel, meiner läuft Quartalsmäßig also alle 3 Monate und startet Sonntag in der Nacht. Ja ist aber auch vollkommen Klar alle Bits auf den Datenträgern werden überprüft, ist so als wenn du die Festplatte komplett befüllst (wir reden hier von einer) bzw. ein bisschen länger dauert es meistens aber ist nicht wirklich trivial. Das macht keinen Unterschied ob Daten auf den Festplatten sind oder nicht, wie schon bei 1. geschrieben werden dei Bits und nicht die Dateien auf Fehler überprüft und hat keine auswirkung wieviel oder welche Daten auf den Platten sind. Puh, das ist teilweise sogar schnell, ist normalerweise die hälfte der tatsächlichen Leistung der Platten, wie immer, aber vermutlich ein bisschen weniger. Wenn die SSD grad sonst nichts anderes zu tuhen hatte dann solltest du volle Geschwindigkeit sehen sprich so um die 110-115MB/s, kommt aber auch auf deinen Switch an und ob dien PC von dem du kopiert hast evtl. grad im hintergrund was anderes gemacht hat usw... Hast du TurboWrite an oder kopierst du auf einen Share der gecached ist?

I would recommend to create a issue over on their Github here: Click