JackSafari

See this thread. The utility is a zip file in one of the messages. https://forums.unraid.net/topic/76035-help-struggling-with-nvidia-audio-passthrough/#comment-1077965

Can I easily passthrough new a AMD GPU to Windows 11 VM? I am planning on upgrading the current NVidia GPU and want to know if can replace the NVidia GPU with an AMD GPU, or should I stick with NVidia. I have been using unraid with the NVidia GPU and it has been working without error for 3+ years. I am passing through the GPU to Windows 11. To get it to work correctly, I had use a custom utility to configure Windows 11, otherwise the audio was a problem.

I created a second user account on the unraid server for personal use. What steps need to be taken to give that account telnet access within my private local network. No access via internet is required. I am able to telnet using root, thus telnet service is working as expected. I understand that the unraid webUI does not have telnet config options, thus it must be done via cli.

Thanks, I will continue to investigate. It is odd that one mod works, but all others do not. According to what I read so far, there should be a Mods directory somewhere on the game server, but I cannot find it. I can't even find the one mod that does work. Overall no big deal. Already completed the game twice without mods.

I have installed ich777's Satisfactory Server and it works perfectly. I have only been able to install one mod "Infinite Nudge" and it works fine. It was the first mod I installed. SMM appears to down and install the mods (no errors while using SMM). When I attempt to join the server, I get this error message "Server missing Mods" for all other mods other than Infinite Nudge. When I disable the mods that will not load, I can join the server and play, otherwise I am blocked. Using SFTP, SMM connects to the server and appears to install mods (no errors are displayed) Any ideas as to what I can try or where I can find related error logs?

It does look like the problem is being caused by an expanding VM.

tower-diagnostics-20240930-0501.zip Uploading diagnostics. I do not recall setting up anything specific to save to mnt/user. However, VMs are being saved to cache/domains. The only one that is active 24/7 is a Windows 10 vm, and as shown is significantly large. Other than this specific cache problem, the unraid server is not having any other problems, running for nearly 3 years.

Trying to figure out why the cache drive keeps dropping extremely low (< 1MB) When too low, running VMs with either pause or crash. Suggestions?

Thanks. This solved it. 👍 qemu-img resize vdisk.qcow2 --shrink 1500G The unraid UI reports 1500G as being 1TB. I checked using qemu-img info and it reports the correct size of 1500G, and Windows 10 recognizes it at 1500G

I did the exact same thing. I immediately discovered the error before starting the VM and attempted to resize it from 15T to 1500G The unraid UI will not allow me resize it smaller even though I have not restarted the Windows VM. The size remains at 15T My entire unraid server is only 12T. The VM is important. I can't wipe it and start all over. Looking for suggestions on how to reduce the vdisk from 15T --> 1500G or revert back to the original size of 1000G

The best that I can determine is that the cache drive dropped to no free space and that caused configuration corruption with the only active VM (Windows). I was able to create new VMs using the existing vDisks and so far everything is back to normal.

Is there a way to recover missing VMs configurations? The VM drives still exist I have no idea what could have gone wrong. The VMs that are now suddenly missing were not running when I rebooted unraid. There appears to be nothing wrong with unraid. The three remaining VMs work without error. EDIT: I did notice that the 2TB cache drive had dropped to 16MBs tower-diagnostics-20230802-1104.zip tower-syslog-20230802-1803.zip

Thanks... That explains things. It has been there so long I completely forgot when it got installed. During the install of 6.12.2 there was a notification that a tool pack was removed, but it appears it the USB Tab was not removed. EDIT: I have removed Nerd pack tools from plugins. This removed "USB" from the menu tab.

Below is all that I get for the USB tab. All other tabs work as expected. Tried - rebooting unraid - access unraid from other computers and devices (Windows 11, iPhone, iPad) - clear browser cache. - Waited 1+ hours for the page to load. - chrome and edge usbtabpage.txt

Currently my windows 10 VM works nearly 100% except I can't get Bluetooth enabled. Not sure what config steps need to be taken in unraid so that the Windows VM can see access the Bluetooth hardware. Windows Device manager is reporting a driver error, but I can't find Bluetooth errors in the system event log. The VM has Bluetooth selected tower-diagnostics-20230712-0226.zip tower-syslog-20230712-0924.zip

After 24 hours no further lockups. It appears reinstalling virtio drivers and\or switching Q35-7.1 has solved the problem.

Here is most recent diagnostic. tower-diagnostics-20230620-1402.zip Edit-1: In effort to solve the problem, today I reinstalled the Virtio drivers and switched from i440fx-7.1 to Q35-7.1. Edit-2: It has been a couple of hours, the Windows 10 vm remains stable. I have done some moderate stress testing that has previously triggered the lockup (screen is frozen, vm is 'paused' by unraid and cannot be unpaused). I am going to let the Windows 10 vm run for at least 24 hours without any further config changes or reboots.

Unfortunately making the suggested changes did not resolve the problem. Overnight while sitting idle the vm locked up again while it was not actively being used; the desktop clock froze at 8:06am. At that point the VM had been running OK for over 12 hours and idle for about 6 hours. Any further suggestion to try?

Yes, it is likely that is the cause the problem because it was the only change to the VM's config that I have made recently. I needed to do an extra step to get unraid to accept the <memoryBacking> change back to nosharepages, I needed to remove: <filesystem type='mount' accessmode='passthrough'> .... </filesystem> Otherwise, unraid rejects the change as a configuration error. is it possible to config virtiofs to work with a windows 10\11 VM that has various other hardware passthroughs? (ie GPU).

I am having a similar or exact same thing occurring on a stable VM that has been running for over 1 year. I am still looking to find a solution. I posted my details here.

I am have the exact same issue. My Windows 10 (version 22H2, build 19045.3086) system ends exactly as the log posted above. The problem recently started for reasons unknown. I also have GPU pass through. The VM has been very stable for over a year. This is the first time I have experienced a fatal VM problem on unraid. Observations: Unraid "pauses" the vm, not stops it. "Resume" does not resume the vm. "Force Stop" kills the vm, but starting the vm causes the vm to be highly unstable and quickly unraid pauses the vm within 1-2 minutes. The only way I am able to clear the problem temporarily is to restart unraid. After start, I can run the VM again and it will be stable for a limited period of time; sometimes many hours. The problem with the VM is more likely to occur during video streaming, such as watching YT videos. This is where the problem appears to be happening. Unraid's Windows 10 vm log: -device '{"driver":"vfio-pci","host":"0000:2d:00.0","id":"hostdev0","bus":"pci.0","addr":"0x8","romfile":"/mnt/user/isos/vbios/GTX1060.rom"}' \ -device '{"driver":"vfio-pci","host":"0000:2d:00.1","id":"hostdev1","bus":"pci.0","addr":"0x9"}' \ -device '{"driver":"usb-host","hostdevice":"/dev/bus/usb/001/005","id":"hostdev2","bus":"usb.0","port":"3"}' \ -device '{"driver":"usb-host","hostdevice":"/dev/bus/usb/001/006","id":"hostdev3","bus":"usb.0","port":"4"}' \ -device '{"driver":"usb-host","hostdevice":"/dev/bus/usb/003/002","id":"hostdev4","bus":"usb.0","port":"5"}' \ -device '{"driver":"usb-host","hostdevice":"/dev/bus/usb/001/004","id":"hostdev5","bus":"usb.0","port":"6"}' \ -sandbox on,obsolete=deny,elevateprivileges=deny,spawn=deny,resourcecontrol=deny \ -msg timestamp=on char device redirected to /dev/pts/0 (label charserial0) 2023-06-19T08:10:19.827059Z qemu-system-x86_64: libusb_set_interface_alt_setting: -5 [NOT_FOUND] .... This is how the log ends every time the unraid forces the WIndows 10 to pause. tower-diagnostics-20230619-0137.zip tower-syslog-20230619-0825.zip WindowsLog.txt

Suggestion\Request. Make this more obvious in some manner. Not a complaint feedback. It took a bit longer than expected because the small "+" icon was off to the left side and only appears when the VM is shutdown. I was running the VM at the time, and was looking for how to add an additional drive, and I could not find the option. I understand it can only be added once the VM is shut down, but while running it there are no visual clues\hints. It help others in the future.

Thanks. You are quick to reply. I should done some research before posting the question. Chat AI is great for answering such questions. AI has already solved a few unraid tech problems.

Thanks. I logged on with a new account because I did not have access through my regular account at that time. I ended up rebuilding the VM settings from scratch. The Windows 10 vm drives still existed and were not damaged. It turned out to be a good thing because for reasons unknown, the previous settings were causing problems that caused Windows 10 vm to hang for 40 minutes when starting the VM. I think it might of been because I mapped one of Window's document folders to an unraid share, and Windows didn't like the mapped location for some reason and hung for 40 minutes.

What is the significance of this error\warning? Such as what impact does it have on unraid that they are not loaded? I am aware that I can temporally fix it with the following unraid shell commands, but the message returns after the server is rebooted. modprobe usbip_host modprobe vhci_hcd