JonathanM

The drives in Unraid are identified by serial number, so as long as the controllers they are attached to doesn't mangle that number, you can move all drives including the licensed USB to another system entirely and the drives will automatically show up. Unraid also won't arbitrarily add drives without you assigning them to disk slots, so no worries there either.

As long as the machine will only be running when someone is physically near the machine that will be fine. Liquid cooling is risky unattended. The most common failure mode is pump failure, where the liquid no longer circulates, causing almost immediate overheating with possible CPU damage. If the worst case of a liquid leak happens, you could fry multiple parts. Air cooling uses relatively massive metal heatsinks, which still dissipate a good amount of heat even when the fan dies, due to convection and air movement from other fans in the system. That gives you plenty of time to find and investigate any failures, where a liquid cooling solution gives very little grace period before things get very warm.

You may use the array as normal, including writes, but it will drastically slow down both operations.

That's a term that used to be used to describe a wireless ethernet transceiver, when gaming consoles didn't all come with wifi. It's a device that joins your wifi network and has an ethernet jack that sends the network traffic to the wired device(s) connected to it. A more current term is a wireless bridge, but you only need one, not a pair. Here is an example. https://iogear.com/products/gwu637

Unraid itself can't use WIFI, but you can certainly set up a gaming adapter to connect to the WIFI and convert it to ethernet so Unraid can use it.

Unraid doesn't check and act on UPS status until way too late in the startup process. You need to make sure there is enough battery left to fully boot up and cleanly shut down before you start Unraid.

What is "it"?

SATA to SATA splitters are not good, to say the least. Practically anything is better, assuming it's compatible with your PSU. My ideal solution without making custom cables would be to source 4 pin "molex" cables that are compatible with the PSU, then using as few splitters as possible convert them to SATA power. If the PSU has 3 connectors for hard drive power, then 1 molex from each cable split into 2 SATA using a high quality crimped (not molded) splitter should work well. If you are short on space you could cut off the extra cable after the one molex connector on each cable that you are actually using. Obviously don't do that until you verify that you have the correct cables. WHATEVER you end up doing, your first priority BEFORE plugging in any drives is to verify the voltage levels are correct at the SATA power connectors. You are playing a high stakes game of russian roulette with very poor odds if you power it up first time with your drives connected. There are many posts here on this forum, and countless posts elsewhere on the internet about people frying their drives by using cables that physically fit, and were described as compatible. DON'T be another casualty.

Try a different browser on the problem computer.

Have you done this? Memtest is only definitive when it fails the RAM, passing just means it didn't hit the failure conditions. Last week I had to replace a stick of DDR5 that consistently blue screen rebooted windows 11, yet passed 24 hours of memtest with no issues. I trusted memtest and replaced both the CPU and the motherboard, only to continue to get the same reboots. Replaced the RAM, boom, stable system.

There is currently no way to restore a file from Unraid parity. Parity doesn't know which drive contains a bit (or more to the issue, possibly bits on multiple drives) that doesn't match, it only knows that a mismatch has occurred. It may be possible to write code that would iteratively "fail" one drive at a time virtually, and run a scrub through each iteration to see if one of the possibilities is more "correct" than the others, but ultimately there is no way without backup file comparison to be mathematically assured that you made the correct choice. If corruption is found on a ZFS volume (or any file system) in the parity array, the only viable current option is to restore that file from backup. Parity only deals with 1 specific failure mode, that is, the loss of the entire drive. If parity is NOT in perfect sync across all drives when that occurs, the rebuilt drive will be corrupt. Trying to use parity to deal with file system corruption is not a viable strategy.

This is exactly why it's recommended to shutdown based on time on battery first. Most of the time if the power is out for more than a minute or so, it's going to be down longer than the UPS can reasonably handle, so it's better to make the shutdown happen ASAP so there is the least battery drain possible. When the power does come back, a second outage is a distinct possibility, so waiting to boot until the battery backup has time to recharge is a good idea, and recharge rates are SLOW. Typical would be 10X, so if the UPS is running for 5 minutes, it's probably going to be close to an hour to recharge back to full. Also, batteries don't have to be replaced as often if you don't drain them below 50%. I have all my desktops and VM's set to shutdown after 1 minute of power out, they get their cue with local apcupsd in slave mode looking at the apcupsd master on Unraid. That gets the ball rolling, hopefully all the VM's are down by the time the 3 minute Unraid master shutdown is commanded. All the network infrastructure stays powered until manually shut down or their dedicated UPS is dead.

If it were me, I'd not do any reconfiguration until the data was backed up. Since that is the case, easiest way forward may be to copy from the emulated failed disk directly to the new empty disk. rsync -arv /mnt/disk5/ /mnt/disk6 rsync -narcv /mnt/disk5/ /mnt/disk6 First command does the copy, second command verifies the copy was bit perfect, it only outputs files with differences. After that is complete, you could do a new config and re-order the drives where you want them and rebuild parity. Short of obtaining another new disk to rebuild the failed one, my method gets you protected the fastest while recovering the data.

Set the shutdown to occur 3 or so minutes after power is lost. If the UPS is appropriately sized and the battery is in good condition that should allow for delays in shutdown, and keep the battery above 50% charge for best life. You should never run a UPS until the last possible moment, it's hard on equipment, and doesn't gain you anything.

Typically that's because file transfers often create and populate folders before starting actual file transfer.

@binhex, I'm seeing pretty much the exact same error now, any clues for me? The app starts ok, and I can use the GUI, right up until one of the connected apps sends a query, and prowlarr just dies. Well, apparently something about the connection wasn't right, I reestablished the VPN and no more crashes.

All that looks correct to me. If you browse the media share in that view, do the missing files show up?

Click the little blue box with the arrow at the left side of each entry.

In the dynamix file manager post screenshots of what shows in both folders, like this.

Probably, but it depends on which containers you end up using. No way for me to know what you will end up getting into, but 2TB is plenty for most people.

Be prepared for serious performance issues if you do that. The way parity works means that the background stuff ZFS does to try to increase performance causes slowdowns instead of increases. For now, I'd keep array drives XFS, and only use ZFS in pools.

Is /mnt/user and /mnt/user0 behaving correctly now?

Data safety, wide compatibility and ease of use have been the primary drivers for many, many years. Raw speed and specialization to specific server tasks have never been at the top of the list. There are plenty of distro's that cater to other niches, Unraid's focus has been an easy to use safe home NAS. The fact that it CAN be used in many other roles is largely coincidental.

Since you know for sure which drive is parity, and your pool config, you can simply reassign all the drives as they were, check the box "Parity is valid" (it's not totally valid, but pretty close) and the array should come back with all your data. Your shares are simply the root folders on all the drives, so they will come back up, with default settings. You will need to edit each one to export it via SMB and / or NFS and set users. The parity check that runs will find a few errors, hopefully not that many.