Mason736

Good morning, fellow UnRaiders! Recently, as early as this morning, I've had 3 drive failures, all of the same model drive. Seagate Ironwolf ST8000VN04. I've been using Seagate drives religiously for years, and have never had an issue. My array currently has over 10 8TB drives of the Ironwolf ST8000VN002 model. Seagate has since replaced the ST8000VN002 with the ST8000VN04 model, and all three drives i've purchased and put in service in the last year, have failed. Has anyone else been having issues with these new drives from Seagate? Thanks

I was able to get a working copy of Catalina 15.5 up and running recently. However, when I benchmark the performance of the VM, with the same number of CPUs (E5-2690v2) against a similar Catalina 15.5 VM I have running in ESXi (same number of CPUs and same processor model), I'm getting significant performance deficiency vs the ESXi VM. Using Cinebench, the Unraid VM gets around 1500-1600 Cinebench score, vs 3300 with the ESXi VM. Any thoughts on what could be causing such a a large performance difference?

Can you recommend a software? I'm not familiar with one. The HDMI 2K is a device called a "Headless Ghost" HDMI adapter. It essentially fakes the GPU into thinking a monitor is attached. Its the only way I could get the GPU to initialize with a display.

First I want to thank everyone here for their work and contributions. I was FINALLY able to get my Catalina up and running stable. Everything works, including passed through GPU, except the audio. I have Lilu and WhateverGreen kexts installed, which helped with passing through the GPU. I have also set the code correctly, i think to allow multifuction to work, and changed the bus for the GPU/Audio to be on the same bus: <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x04' slot='0x00' function='0x0'/> </source> <alias name='hostdev0'/> <address type='pci' domain='0x0000' bus='0x04' slot='0x00' function='0x0' multifunction='on'/> </hostdev> <hostdev mode='subsystem' type='pci' managed='yes'> <driver name='vfio'/> <source> <address domain='0x0000' bus='0x04' slot='0x00' function='0x1'/> </source> <alias name='hostdev1'/> <address type='pci' domain='0x0000' bus='0x04' slot='0x00' function='0x1'/> </hostdev> I also installed the AppleALC kext, but it seemed to have no effect. I am seeing the HDMI audio out in the sound preferences, but I am not getting any sound coming through the VM via NoMachine. The server running the VM is headless, so there is no monitor or HDMI hookup directly. I'm using a headless ghost HDMI device to initiate the graphics adapter to work. The GPU I'm using is an AMD WX 4100 4 GB, and is correctly seen by the system. No matter which audio output choice I select, HDMI2K, or NoMachine Audio, I do not get any audio.

It has been working fine for me. Not sure why it failed that initial time, but VEEAM has been working flawlessly since then.

Update: So I just manually ran it, and as before it worked, as with Windows Server Backup. But when it runs overnight, it fails. I'm lost.

I'm having an issue, and not sure how to resolve it. I recently bought a new Hyper-V server, and wanted to use my Unraid server as a backup location for the Hyper-V backups. I created an SMB share on the Unraid server, and initially setup Windows Server Backup to perform a Bare Metal Recovery backup, as well as full backup of my Hyper-V VMs. When I ran an initial test after I created the backup job, it succeed. That night it succeeded as well. However, since the success the first night, it has failed every night with a network error. I switched to the community edition of Veeam, as I thought maybe the issue was Windows Server Backup. I set the backup repository to my SMB share on the Unraid server. Same thing occurred last night, the backup failed with errors of: "Processing VM Error: An unexpected network error occurred.....". Actual error from on of the VMs, they are all the same: 4/15/2020 4:20:42 AM :: Processing SophosXG2 Error: An unexpected network error occurred. Failed to flush file buffers. File: [\\192.168.1.125\Windows Server Backup\Veeam\Main VM Job\Main VM JobD2020-04-15T030038_6197.vbk]. Failed to download disk 'SophosXG2_old.vhdx'. Shared memory connection was closed. Failed to upload disk. Agent failed to process method {DataTransfer.SyncDisk}. Could it be because the drives are spun down, and not initially available? However I don't think that is the case, as when I tested why the Windows Server Backup would fail, it would fail as it was over 100GB into the backup. Forgot to mention, I have the Share on Unraid set to not use the Cache, (no), and write directly to the array. I didn't want to waste SSD writes on the backup of VMs. Any help is appreciated. Thanks

Update: I'm able to see my Radeon Pro WX 4100 in system information. But it doesn't show in the "About this Mac" page, and I'm not able to change the graphics to anything other than 1280x1024. Any Suggestions?

I have successfully setup a Catalina VM using Macinabox, however I want to pass though my graphics card, AMD Radeon WX 4100 to the VM. Whenever I add the GPU as a second graphics card, the VM freezes about halfway though bootup on the apple logo. If I change the GPU to the primary, the GPU fails to boot doing the same thing, freezing on boot. I have installed Lilu and Whatevergreen Kexts via Clover, and the issue persists. Any thoughts?

Update, its working. My mistake, I deleted the docker, for some reason the image file was still there. After I removed the image fine, and reinstalled the docker, its working.

Its still broken, just tried to install a new Catalina image.

I was finally able to get my GPU working (switched to AMD Radeon Pro WX 4100 from an nVidia card). It works great, however the resolution is locked at 1280x1024, and I'm not able to change it. I've installed Lilu and Whatevergreen kexts, as well as emuvariableuefi-64 via Clover, however it didn't seem to help the graphics. Any suggestions?

Wonderful, thanks for the clarification.

Im sorry, I don't think you're understanding the question. I am speaking specifically about a VM, not the metal unraid runs on itself. With Hyper-V, regardless of whether you pass through a GPU, same with V-Sphere, you can access the VM, and have full KVM usage with keyboard and mouse support. In unraid, if you pass-through a GPU, and don't have a monitor connected to the GPU for the VM, it appears there is no way to obtain a KVM like usage as a Type 1 Hypervisor. I've tried virt-manager as well, but when you pass through a GPU, without connecting to a monitor, you can't access the display of the VM until it boots to change the boot order (access Clover for instance on a Catalina VM), or do other administrative tasks.

Yes, but when you pass the GPU through, without hooking up to a monitor, how to you get to the post screen, enter BIOS, etc...?

I'm not sure if this question has been asked before, I didn't see it when I searched. I've been running Unraid for a couple years, but just started getting into the VM side. I've predominantly used Hyper-V or ESXI for VMs in the past, so the Unraid side is new to me. I currently run my server headless, in a server closet in my basement. If I pass-through a GPU to accelerate the graphics in a gaming VM or Mac Catalina VM, how do I administer the VM, like a Type 1 Hypervisor (Hyper-V, ESXI), and perform boot functions and other administrative tasks? I am not able to connect a monitor directly to the VM.

Any thoughts on this? I'd really love to be able to change the display resolution.

I got my nvidia GT730 to be natively seen by Catalina. However, I can't change the display resolution. Its locked in at 1280x1024. How do I change this?

I tried using VNC Viewer on Elementary OS. Installed VNC Server. I can connect just fine with VNC Viewer without passing through the GPU. After I pass it through, VNC Viewer will no longer connect, giving the error "Cannont currently show the desktop.".

Question about GPU passthrough and acceleration. My Unraid server is currently running in a rack in my basement. I access it headlessly, where it runs plex, some windows VMs, and other assorting things. I have a macOS Catalina VM, and a few linux VMs as well, that frankly, don't have that good of performance, vs when I run them in VMWare Workstation on my laptop. I bought a couple of nvidia GPUs to use for the VMs, but now I'm realizing I think I made a mistake. I use RealVNC Server/Viewer to access the macOS and Linux VMs today, and RDP for the Windows VMs. I'd like to use GPU acceleration for the macOS and Linux VMs, but still access them through VNC Viewer or xRDP, not directly connect a monitor to the VM, as this isn't physically possible. Is it possibly to hardware accelerate the linux and macOS VMs, and still remotely access the VMs, or do you have to use a phycially connected monitor for access with GPU passthrough. When I pass the GPU and sound through to the linux VM, and then try and connect to VNC viewer, i get an error saying it "Cannot currently show the desktop." Thank you.