JonathanM

Unraid doesn't use ANY of the options shown at that calculator link. It's UNRAID. As long as all the remaining data drives are functioning 100%, Unraid can emulate and rebuild from ANY number of data drives. It's currently limited by choice to 28 data disks, so if you have all the same size disks you have less than 3% lost overhead. The probabilities of one of the remaining drives acting up during a rebuild is not negligible, so we usually recommend 2 parity drives any time you have more than 10 or so data disks, but it's all a matter of statistics, and you can choose to run all 28 disks with one parity if you like.

except Unraid itself

Data drives are the only place to store data, parity doesn't hold any sensible data. So in a sense, yes you are losing the entire 20TB parity drive of capacity to gain the ability to rebuild any single drive failure, assuming all the rest of the data drives are working seamlessly across their entire capacity. https://wiki.unraid.net/Parity

Why? There is no good reason to do that. Set Unraid's DNS to a reliable external source like 1.1.1.1 or 9.9.9.9, or google's DNS of 8.8.8.8.

yes, it's already emulating the pcie bus in either case.

Correct, it will just be emulating the PCI or PCIe slot, depending on which machine type the VM is using.

possibly The issue here is that the USB controller is being generated in software by the KVM emulation, so the more resources you can give to the host and remove from the VM the better. Only give the VM the bare minimum to run well, no more.

Easier to split out the audio from the HDMI connection. If acs override won't split them up, then it's probably a motherboard limitation.

You really need IPMI or the equivalent. https://pikvm.org/ could be a good option.

Preclear is not required, as the drives will be fully overwritten with the emulated data during the rebuild. Since you have 2 failed drives currently, you are not protected from another failure until you get at least 1 of the drives rebuilt successfully. I can't see a reason not to rebuild both at the same time.

You can easily set up a trial USB of Unraid and see if it will boot.

Your data will be fine. Any containers and VM's will have to be rebuilt, but as long as the drives are assigned in the correct slots, all your data will show back up just fine. Shares will no longer have any security customizations, so be sure to go into the shares page and set them to restrict access if desired. Do you have ANY files backed up from your old USB?

I don't believe Unraid has an M1 version, X86-64 only at this time.

If you can't ping the dns IP address, you won't get any farther until you fix that.

The Unraid host provided VNC server should only be used for management tasks that can't be accomplished with remote access software in the VM guest itself. For a remote desktop, try installing nomachine on the VM and your desktop, and use that. For remote terminal, enable SSH in the VM and connect directly using putty or similar.

Perhaps here? or here? You can find the correct support thread for your container by clicking on "Support" in the Unraid app icon drop down menu.

The UUID is part of the filesystem, and since rebuilding a drive clones everything including all the filesystem stuff, the original and rebuilt drive have the exact same UUID, and only one will get mounted until you change it to a unique UUID. The UUID is copied to the rebuilt disk, so it still belongs to the array. Only needed if you want to mount both the rebuilt and original drive at the same time in the same OS.

You would need to look at the diagnostics to see what UD is seeing when you try to attach the drives. If you can't figure it out by yourself, post in the Unassigned Devices support thread and attach your diagnostics covering the period where you have the drive attached.

A single Ryzen Threadripper 1920X is no where near the raw processor power of dual E5-2680 V4's, using passmark as a gauge. Power is not as clear cut either, and I don't know a good way to compare that, as TDP is only a 100% CPU measure, and doesn't translate to idle or low CPU usage. This assumes the same HBA and drive load for both systems. Based on my observations of prior performance, I think enterprise supermicro stuff will be more reliable long term.

See here. https://forums.unraid.net/topic/44142-support-binhex-plex-pass/?do=findComment&comment=767160

Host network type doesn't remap ports, it passes them all directly. Doesn't matter whose version.

The issue with pools not being redundant despite appearing to work properly was prevalent a year or two ago, now that JorgeB mentioned it. If you can't determine that yourself, attach diagnostics to your reply to this post.

That is exactly what the parity system does, it uses all the rest of the data drives plus the parity drives to emulate the missing drives, precisely as if they were still there. If you can't get the physical drives to respond, you have to assume they failed, and must be replaced. You really need to remove the drives and check in another system if the SATA and power leads in your server have been checked.

Before you follow through, you may want to wait on @JorgeB to comment on whether the procedure is working correctly on the version of Unraid you are running. You didn't post diagnostics, so it's impossible to say whether it will work for you.