Squirreljester

I've already uninstalled and reinstalled the nvidia driver, but in the plugin it says: Nvidia Driver Version: 595.71.05 Open Source Kernel Module: No Installed GPU(s): NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running. My syslog says my p2000 needs a 580.xx driver, which I had installed originally (580.142), but my Plex wouldn't start after the update because of a GPU passthrough error: Apr 29 20:35:25 <Redacted> rpc.mountd[23432]: Version 2.8.4 starting Apr 29 20:35:26 <Redacted> rc.nfsd: NFS server daemon... Started. Apr 29 20:35:30 <Redacted> webgui: Successful login user root from 192.168.15.178 Apr 29 20:35:39 <Redacted> kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 239 Apr 29 20:35:39 <Redacted> kernel: NVRM: The NVIDIA Quadro P2000 GPU installed in this system is Apr 29 20:35:39 <Redacted> kernel: NVRM: supported through the NVIDIA 580.xx Legacy drivers. Please Apr 29 20:35:39 <Redacted> kernel: NVRM: visit http://www.nvidia.com/object/unix.html for more Apr 29 20:35:39 <Redacted> kernel: NVRM: information. The 595.71.05 NVIDIA driver will ignore Apr 29 20:35:39 <Redacted> kernel: NVRM: this GPU. Continuing probe... Apr 29 20:35:39 <Redacted> kernel: NVRM: No NVIDIA GPU found. Apr 29 20:35:39 <Redacted> kernel: nvidia-nvlink: Unregistered Nvlink Core, major device number 239 Apr 29 20:35:39 <Redacted> kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 239 Apr 29 20:35:39 <Redacted> kernel: NVRM: The NVIDIA Quadro P2000 GPU installed in this system is Apr 29 20:35:39 <Redacted> kernel: NVRM: supported through the NVIDIA 580.xx Legacy drivers. Please Apr 29 20:35:39 <Redacted> kernel: NVRM: visit http://www.nvidia.com/object/unix.html for more Apr 29 20:35:39 <Redacted> kernel: NVRM: information. The 595.71.05 NVIDIA driver will ignore Apr 29 20:35:39 <Redacted> kernel: NVRM: this GPU. Continuing probe... Apr 29 20:35:39 <Redacted> kernel: NVRM: No NVIDIA GPU found. Apr 29 20:35:39 <Redacted> kernel: nvidia-nvlink: Unregistered Nvlink Core, major device number 239 Apr 29 20:35:40 <Redacted> kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 239 Apr 29 20:35:40 <Redacted> kernel: NVRM: The NVIDIA Quadro P2000 GPU installed in this system is Apr 29 20:35:40 <Redacted> kernel: NVRM: supported through the NVIDIA 580.xx Legacy drivers. Please Apr 29 20:35:40 <Redacted> kernel: NVRM: visit http://www.nvidia.com/object/unix.html for more Apr 29 20:35:40 <Redacted> kernel: NVRM: information. The 595.71.05 NVIDIA driver will ignore Apr 29 20:35:40 <Redacted> kernel: NVRM: this GPU. Continuing probe... Apr 29 20:35:40 <Redacted> kernel: NVRM: No NVIDIA GPU found. Apr 29 20:35:40 <Redacted> kernel: nvidia-nvlink: Unregistered Nvlink Core, major device number 239 UPDATE: I just reinstalled the 580.142 driver and rebooted for the 4th or 5th time tonight, here's my Plex log: 026-04-29 20:50:44.232132 [info] Host is running unRAID 2026-04-29 20:50:44.249597 [info] System information: Linux f734ff530687 6.12.54-Unraid #1 SMP PREEMPT_DYNAMIC Tue Oct 21 15:58:46 PDT 2025 x86_64 GNU/Linux 2026-04-29 20:50:44.268255 [info] Image architecture: 'amd64' 2026-04-29 20:50:44.286609 [info] Application name: 'plexpass' 2026-04-29 20:50:44.306715 [info] Base image release tag: '2026032801' 2026-04-29 20:50:44.323069 [info] Application image release tag: '1.43.1.10611-1-01' 2026-04-29 20:50:44.345254 [info] PUID defined as '99' 2026-04-29 20:50:44.364620 [info] User 'nobody' already has UID '99', skipping usermod 2026-04-29 20:50:44.387828 [info] PGID defined as '100' 2026-04-29 20:50:44.408034 [info] Group 'users' already has GID '100', skipping groupmod 2026-04-29 20:50:44.424773 [info] UMASK defined as '000' 2026-04-29 20:50:44.445045 [info] Permissions file '/config/perms.txt' exists, skipping setting ownership and permissions on '/config' and '/data' 2026-04-29 20:50:44.467570 [info] Deleting files in /tmp (non recursive)... 2026-04-29 20:50:44.490980 [warn] TRANS_DIR not defined,(via -e TRANS_DIR), defaulting to '/config/tmp' 2026-04-29 20:50:44.515990 [info] Starting Supervisor... 2026-04-29 20:50:44,641 INFO Included extra file "/etc/supervisor/conf.d/plexmediaserver.conf" during parsing 2026-04-29 20:50:44,641 INFO Set uid to user 0 succeeded 2026-04-29 20:50:44,644 INFO RPC interface 'supervisor' initialized 2026-04-29 20:50:44,644 INFO supervisord started with pid 49 2026-04-29 20:50:45,647 INFO spawned: 'claim-script' with pid 111 2026-04-29 20:50:45,648 INFO spawned: 'start-script' with pid 112 2026-04-29 20:50:45,649 INFO reaped unknown pid 50 (exit status 0) 2026-04-29 20:50:45,653 DEBG 'claim-script' stdout output: [info] Env var 'PLEX_CLAIM' value not set and/or 'CLAIM_SERVER' not set to 'yes', skipping claim process. 2026-04-29 20:50:45,653 INFO success: claim-script entered RUNNING state, process has stayed up for > than 0 seconds (startsecs) 2026-04-29 20:50:45,653 DEBG fd 9 closed, stopped monitoring <POutputDispatcher at 22862189155328 for <Subprocess at 22862190410256 with name claim-script in state RUNNING> (stdout)> 2026-04-29 20:50:45,653 DEBG fd 11 closed, stopped monitoring <POutputDispatcher at 22862189091152 for <Subprocess at 22862190410256 with name claim-script in state RUNNING> (stderr)> 2026-04-29 20:50:45,653 INFO exited: claim-script (exit status 0; expected) 2026-04-29 20:50:45,653 DEBG received SIGCHLD indicating a child quit 2026-04-29 20:50:45,658 DEBG 'start-script' stdout output: [info] Env var 'PLEX_CLAIM' value not set and/or 'CLAIM_SERVER' not set to 'yes', skipping edit of Preferences.xml for claim process. 2026-04-29 20:50:46,658 INFO success: start-script entered RUNNING state, process has stayed up for > than 1 seconds (startsecs) 2026-04-29 20:50:59,561 DEBG 'start-script' stdout output: Critical: libusb_init failed

Just to follow up with this, using this driver made the Steam Headless work perfectly, including adding the options to the docker for it to use the GPU. Also my Plex server is transcoding super super fast, faster than I can even GPU-Status can even catch the status of most of the time.

I ran the chown -R 1000:1000 /mnt/user/FoundryVTT command already, and it didn't help, the docker still starts and then stops with the same error. I also don't know what "change the docker user container to match the data." means. I've updated the docker settings with "--user 99:100" in the extra parameters and changes the repository to "felddy/foundryvtt:release" per your suggestions. I'm not sure how to proceed to resolve this issue.

I'm getting this at the end of the log and the container won't stay running: Warning: Failed to open the file downloading.zip: Permission denied * Failure writing output to destination 0 220M 0 1360 0 0 10249 0 6:16:41 --:--:-- 6:16:41 10225 * Connection #0 to host r2.foundryvtt.com left intact curl: (23) Failure writing output to destination Entrypoint | 2025-08-28 17:26:18 | [warn] Download from presigned URL failed with exit code 23. Entrypoint | 2025-08-28 17:26:18 | [error] No valid cached release file found. Unable to proceed with installation.

ok I downgraded my nvidia driver to 570.172.08 (I was on 575.64.03) and it appears to be working now with the GPU runtime added. I'm not sure what effect this will have on my Plex server transcoding. I will monitor.

I'm having an issue adding my nvidia gpu. I'm adding it just as it shows in the documentations, and it is identical to how I have Plex set up to use my GPU, which works fine. When I add --runtime=nvidia it loads the drivers per the log, but I get these errors: 2025-08-05 02:23:53,225 INFO spawned: 'xorg' with pid 426 2025-08-05 02:23:53,272 WARN exited: xorg (exit status 1; not expected) 2025-08-05 02:23:55,234 INFO spawned: 'xorg' with pid 457 2025-08-05 02:23:55,280 WARN exited: xorg (exit status 1; not expected) 2025-08-05 02:23:58,221 INFO spawned: 'xorg' with pid 496 2025-08-05 02:23:58,267 WARN exited: xorg (exit status 1; not expected) 2025-08-05 02:24:01,272 INFO spawned: 'xorg' with pid 538 2025-08-05 02:24:01,320 WARN exited: xorg (exit status 1; not expected) 2025-08-05 02:24:02,321 INFO gave up: xorg entered FATAL state, too many start retries too quickly When I take out the runtime entry, Steam-headless webUI connects and gives me a login. It has to be that entry, but for whatever reason it is just not working like it's supposed to, and I don't know why.

I removed the container, removed the folder, re-installed, and didn't touch anything, left everything defaults, and now it appears to be working. This means some setting I changed broke it, but I'm not sure what did it. Does the password variable in the config need to be default or should I change that to something specific to the container?

I'm trying to get this to work. I can connect to the WebUI but when I click the connect button I get "Failed to Connect to server" message. I've looked through the thread but a lot of the solutions are years old, even the extra step of installing uinput on the project page is outdated. I've installed the app through Unraid, gave it a game library which it has installed it's files to so there's not a permission issue there. I have GPU passthough set up with the variable populated with my GPUID, --runtime=nvidia at the beginning in extra parameters. Is there additional steps I need to take to make this work outside of the instructions on the project page? I'm wondering about the password piece and possibly the timezone. I've tried to change the timezone to my own and I've tried leaving it default. I've also tried to put my own password in there, and I've kept the default, but I always get the same result.

Any ideas what's going on? I had this docker up and running for over 6 months without an issue, then I noticed it was off and wouldn't start. I've even tried doing the extra parameters "--user 99:100" that I found earlier in the thread, but no luck. What can I do?

dockerMan/templates-user.xml <?xml version="1.0"?> <Container version="2"> <Name>FoundryVTT</Name> <Repository>felddy/foundryvtt:release</Repository> <Registry>https://hub.docker.com/r/felddy/foundryvtt</Registry> <Network>bridge</Network> <MyIP/> <Shell>sh</Shell> <Privileged>false</Privileged> <Support>https://forums.unraid.net/topic/121065-support-silman-foundryvtt-felddy-docker-image/</Support> <Project>https://github.com/felddy/foundryvtt-docker</Project> <Overview>A secure, reliable, compact, and simple-to-use container for the Foundry Virtual Tabletop using felddy's Dockerhub images.&#xD; Felddy has the most robust docker image for FoundryVTT their image has over 10M downloads.&#xD; &#xD; You can get a Foundry Virtual Tabletop instance up and running in minutes using this container. This Docker container is designed to be secure, re liable, compact, and simple to use. It only requires that you provide the credentials or URL needed to download a Foundry Virtual Tabletop distrib ution.</Overview> <Category>GameServers:</Category> <WebUI>http://[IP]:[PORT:30000]</WebUI> <TemplateURL>https://raw.githubusercontent.com/silman/unraid_templates/master/foundry/foundry.xml</TemplateURL> <Icon>https://www.foundryvtt.wiki/fvtt-solid-512.png</Icon> <ExtraParams/> <PostArgs/> <CPUset/> <DateInstalled>1752594040</DateInstalled> <DonateText/> <DonateLink/> <Requires/> <Config Name="Game Data Path" Target="/data" Default="" Mode="rw" Description="Maps to container path /data. Path to persistent directory for st oring Config, Data, and Logs directories for Foundry to use." Type="Path" Display="always" Required="true" Mask="false">/mnt/user/FoundryVTT/</Con fig>

I apologize, where do I find that?

I've done all the updates that have come up when it says there's one available. How do I check if it's v12 or v13 when it won't stay running anymore?

I just started having a strange problem where the app isn't searching on it's own, and I can't click on the "automatic search" icon in the GUI. I also can't click the "Refresh and Scan" or "Search for Missing". What does work is clicking "Interactive Search", which brings up results, and it will let me select something to download. I just noticed that I also can't add any new shows. I hit the button to add the show, the button flashes, but then nothing happens. I can click on the button multiple times and nothing happens, and the show doesn't add. Any ideas what's going on? It worked up until a couple days ago.

I just started getting this error out of the blue: FoundryVTT | 2025-07-15 15:40:50 | [error] A fatal error occurred while trying to start the Foundry Virtual Tabletop server: EACCES: permission denied, open '/data/Data/systems/README.txt' Error: A fatal error occurred while trying to start the Foundry Virtual Tabletop server: EACCES: permission denied, open '/data/Data/systems/README.txt'

Support was able to re-issue my a new key and it worked. Strangely I had to go into the key management and re-link the key too. I didn't see that step in any of the instructions I read online.

Actually, I found an option to restore from my zip file backup, so I did that. I got my server back up and it looks right, but now I'm trying to replace the license key, and when I hit confirm key replacement, I get a little "Error attempting to replace key" I had to remove the Trial.key and Plus.key because it was complaining there were multiple key files. I left the Unleashed.key file in there, and it then I was given the option to replace the key. Now I'm logged into Unraid trying to replace my key, and I'm getting an error, but I'm not sure what the issue is. I paid for Unleashed when I added my last 2 drives less than a year ago.

I'm replacing a drive that's messed up and Windows can't repair. When I go through the creator to create the new flash drive, there's different options for OS versions. I was still on 7.0.1 when my USB died. I have a recent backup of the USB. My question is, should I install 7.0.1 back onto the replacement drive and then upgrade it when everything is back up and running? Will it mess anything up if I install the newest version onto the new USB and then restore my backup to it?

FoundryVTT recommended specs for dedicated server is 2vCPU and 4gb ram Dungeondraft minimum specs are i3 CPU, 4gb ram, and Intel integrated graphics 6000 I have 8vCPU's, 16GB ram, and my p2000 GPU assigned to the VM. I should be more than fine specs wise, better even if I figure out a second GPU option. I've tried using the virtual graphics but Dungeondraft didn't like it and wouldn't even start up.

I still need to update my signature with this info: AMD Ryzen 9 5950X 16-Core @ 3400 MHz ASUSTeK COMPUTER INC. ROG STRIX B550-A GAMING Memory 64gb Drives 10+2 Parity (140tb) + 2 m.2 nvme (500gb cache and 1tb app) LSI Logic Controller Card LSI00301 SAS 9207-8i 8Port GPU: Nvidia p2000

I've been wanting to add a second video card to my system but I only have 2 x16 slots. One of them has my p2000 and the other has my LSI hard drive controller card. I've been trying to figure out an option for using the 4x slot for it. The goal is to keep my p2000 for Plex transcoding and use a second video card to dedicate to a VM running Foundry VTT and Dungeondraft. My options currently are: 1. Move the LSI card to a 4x slot using a riser to free up an x16 slot (not preferred) 2. Use a 4x to 16x riser or cable. According to this it appears that I wouldn't lose that much performance: https://www.techpowerup.com/review/nvidia-geforce-gtx-1080-pci-express-scaling/24.html 3. Find a video card that slots into a 4x slot. This is the best one I've found and the option I'm leaning toward unless someone else has a better option: https://www.amazon.com/SRhonyra-GeForce-Graphics-Profile-Compatible/dp/B0CHDZTW6P 4. Use some sort of external video card. I've seen these incased with their own power supplies, but they are very expensive.

It sounds and looks like I can't display it in GPU Statistics if it's bound to a VM, even if I check that box to bind it to vfio, right?

I found this in System Devices. Is this what you're talking about?

The VM is definitely using the card, I'm running an application it the VM that wouldn't work until I gave the VM access to the gpu. Per the images, the gpu is attached to the vm, and is checked in the gpu statistics plugin. Doesn't that mean it should be in vfio and should be working? Where is "system settings"? GPU Statistics plugin: VM settings:

I read a little bit about the vfio binding. I don't know what that is, but it sounds like I need to install a plugin for it. How can I get this showing the correct information? Do I need to install that plugin? Run some commands on the console? Are you saying that this plugin won't work if the card is assigned to a VM like I have it? I also edited my post to include a little more info.

I had my p2000 bound to my dockers and specifically my Plex ever since I set this Unraid server up, and the dashboard GPU box always gave me the utilization of the GPU. I recently unbound it from my dockers and Plex server so that I can use it in a VM. I set it in the VM settings, and an application that requires a graphics card that wasn't working with the VM default is now working with the p2000 attached. But now the GPU info box on the dashboard says "GPU not available bound to VFIO or inuse in a VM." It absolutely is "in use" as verified by the application on my VM that wouldn't run without it, and in the box it does show p2000 as the device, so it seems like it's at least detecting its existence. I have the Nvidia driver installed, and I've pointed the "GPU Statistics" plugin I have installed which shows up in the device dropdown, so it does know about the card and is pointed to it. Is there something in Unraid that I'm missing that I need to set up to get the dashboard GPU data working? Something in the plugin? When I refresh the dashboard it shows what's in the attached image for a second, but that disappears and shows the above error after a second.