JorgeB

Current 6TB disk is the correct size (for 6TB), but the disk that failed for some reason was larger than normal: [sizeSb] => 5860765964 This is the normal size for 6TB: [sizeSb] => 5860522532 You'll need to replace it with an 8TB drive, or manually copy the data and do a new config.

Unraid needs to format all the array/pool drives, so you'd need at least one free drive to assign first, then could use the UD plugin to copy the data, assuming they really are simple btrfs volumes.

You can buy RAM with JEDEC default settings for 3200, just an example: https://www.kingston.com/dataSheets/KVR32N22S6_8.pdf "Gamer" DIMMs usually default to 2133 or 2400 and you need to use the XMP profiles to set it higher, e.g.: https://www.kingston.com/dataSheets/HX432C16FB3_8.pdf I wouldn't really call that overclocking, though you can still I can do it since they usually use higher voltage because of the tighter timings, but worse would be using for example 3600 RAM on a CPU that officially only supports up to 3200.

Any progress on this? Still happening.

So it's working now? Also note that SATA port multipliers are not recommended with Unraid, you may run into more issues in the future.

Doubt the beta will help, but here it is if you want to try: https://we.tl/t-DIJOfT95VK

No, you'd need to have all the disks assigned to rebuild, with single parity data drive order is not important, but parity would need to be correctly assigned.

Please post the diagnostics: Tools -> Diagnostics

Thanks for the update, I've seen several issues caused by those AMD controllers in IDE mode before, unfortunately missed that this time.

du isn't accurate with btrfs, what the GUI shows is the correct space used, if you have Windows vdisks there see this.

You could downgrade temporarily to v6.8 to disable that VM from auto starting.

Yeah, that was true before v6.9, now with v6.9 they start anyway, which is IMHO a very bad idea.

You need to edit the configuration/XML, you can also ask for more help in the appropriate sub-forum: https://forums.unraid.net/forum/51-vm-engine-kvm/

That means the server is not getting an IP address from your router.

You can try a recovery tool like testdisk, if it was just the partition that was deleted it should recover it.

Reoccurring filesystem corruption without an apparent reason suggests an underlying hardware issue, start by running memtest.

Mar 24 21:45:54 Tower kernel: SQUASHFS error: Unable to read fragment cache entry [b19178] Mar 24 21:45:54 Tower kernel: SQUASHFS error: Unable to read page, block b19178, size 14a24 These suggest a flash drive problem, backup, re-do it, restore config folder.

Depending on the RAID controller disks might be unmountable after the move, if that happens it can be fixed but requires rebuilding each disk, e.g.:

There's no partition on that disk, somehow it got deleted.

Likely unrelated to this issue.

This should do it: mount -o remount -o nodiscard /mnt/cache

Do that. SMART not working is a sign there's a problem, no point in trying to bypass it.

Changed Status to Closed Changed Priority to Other

disk5 is failing SMART: device is NOT READY (e.g. spun down, busy) A mandatory SMART command failed: exiting. To continue, add one or more '-T permissive' options. Try replacing the cables, if it's the same it's likely a failing disk.

This is a very bad idea. See if the disk mounts with the UD plugin, unlike Unraid it doesn't required a specific partition layout, if it also doesn't mound post the output of: fdisk -l /dev/sdX Replace X with correct letter.