JorgeB

They are usually fine with Unraid, they should be avoided for any kind of RAID.

And because of that there's no point in doing a pre-read when pre-clearing a new one, just clear and post-read.

Yeah, they are weird like that, even the long SMART test completes in a few seconds before they are used for writes.

In the future please use the appropriate support thread:

That because those SMR drives don't read the platters when they know there's no data there, do a full write and then you'll get the actual read speed from the platters.

Start here: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

Please use the appropriate docker support thread:

PCIe link width, that slot is x16 physically but only x4 electrically.

Tools -> Diagnostics

Is a driver being loaded? You can post the diags to confirm.

Problem appears to have been the disk, so wait for the SMART test to finish. Correct procedure in this case would be to re-enable the drive (if SMART test is successful).

Look for one on ebay, this one is considered a reputable seller.

Yes, but careful with Chinese fakes. It should, link limited to x4, but that's OK for most HDDs.

The controller includes the SATA multiplier, replace it with any LSI with a SAS2008/2308/3008/3408 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, 9400-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

Either one will work.

Best bet, look for an LSI HBA.

Ahh, OK, Asmedia only makes 2 port controllers, so that controller uses 4 Asmedia chips, and looks like 4 separate controllers on the diags. By the photo I see they have an extra chip probably to link all those controllers, and likely that's the one that is causing is causing issues with IOMMU.

Good catch, I missed that, that explains it.

According to diags you have 4: 06:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] (rev 02) Subsystem: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] Kernel driver in use: ahci Kernel modules: ahci 07:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] (rev 02) Subsystem: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] Kernel driver in use: ahci Kernel modules: ahci 08:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] (rev 02) Subsystem: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] Kernel driver in use: ahci Kernel modules: ahci 09:00.0 SATA controller [0106]: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] (rev 02) Subsystem: ASMedia Technology Inc. ASM1062 Serial ATA Controller [1b21:0612] Kernel driver in use: ahci Kernel modules: ahci You you really only have one this could be part of the problem.

Strange, Asmedia controllers aren't usually affect by IOMMU, does the same happen if you use another of the Asmedia controllers you have installed?

Stop all VMs and the VM service, change Domains share to cache="prefer" and run the mover, then post new diags.

The 6 or 10 ports controller you're using is in fact a 2 port controller with SATA port multipliers, they are a known problem, you'll need to replace it to get rid of those timeout errors.

Like mentioned above the correct procedure here would have been a new config, no need to re-sync parity. (Tools -> New config), but note:

Next time please post diags instead, wait for the extended SMART test to finish, it will be the best indicator of the current disk health.

Where are the vdisks located? You're using /mnt/user path and the domains shares exists on both cache and disk1.