CaptainMorganCrunch

Well, I copied over two cfg files from the pools dir and the super.dat from my 7.2.5 backup zip and rebooted. It came back up with assignments! All good! Thanks for the help, @trurl !

I copied over my config folder but didn't have disk assignments. I checked my last backup from 7.3.1 and the pools dir is just garbage. I checked the 7.2.5 backup zip and I see zfspool.cfg and cache.cfg and none of the extra junk. So the question is, should I just copy over the 7.2.5 backup pool and super.dat file ontop of the 7.3.1 flash disk, or just format and write the 7.2.5 backup to the usb flash? (Feature request: sanity check/validate backups so there isn't false sense of success.)

I had unraid 7.3.1 running and the webGUI wouldn't load but I could still login over SSH. It looked like the USB boot drive was corrupted. I took a command line backup zip file. I had thought I had taken another backup after successful upgrade to v7.3.1, but I suspect I only downloaded it to a ramdisk that got wiped on PC reboot since I can't find it in my usual storage locations. When I made a replacement USB flash disk and copied over the config dir and booted up, I did the license key replacement for the new USB key but then saw none of my drives were assigned. At this point, I'm not sure what to do to move forward and get it operational again. I have an older backup of v7.2.5 before the upgrade to 7.3.1 I could try. I believe I could make the USB flash as 7.3.1 and just need to copy the latest .key file from the current usb drive to be the current licensed key for this USB flash along with the config dir. Use the backup of v7.2.5 as is, upon first boot, select the latest key file matched to USB and then upgrade to 7.3.1. Give up on importing previous config, reconfigure the assigned drives. I know which two drives are parity, but does it matter which drives get assigned to which slot? Thanks support.zip

Server has been rebuilt. Thanks again for everyone's help.

What do you mean by "rebuild on top"? Thanks!

Back to showing the 8TB and 500GB disks as disabled. Thanks unraid-diagnostics-20260321-0124.zip

The drive bays were not using a common power connection and it was tricky to change cables for the time being so I just swapped the bays the drives were in and precleared one of the 500GB disks that is reported as disabled. I stopped the array to try and use the 500GB disk again, but it still shows as disabled. In the meantime, another 3TB is now disabled. So the problem follows the drive bay, which is good to know for when I can swap SATA cables tomorrow. But I do need to get at least one more disk to work to start the array with 2 disabled/rebuilding disks. Thanks unraid-diagnostics-20260320-2238.zip

The other day, I got two drives reporting disabled, disk 15 (4TB) and disk 17 (500GB). I replaced them with an 8TB and another 500GB. They were precleared and then added and were rebuilding and the rebuild finished only to get those two drives disabled again shortly after. I tried rebooting and it came back up disk 15 and 17 still disabled. I also have disk 18 that reported 116068861 errors but didn't get disabled and doesn't have SMART issues reported. Motherboard is Supermicro H11SSL-i. I ran HD Sentinel on the 4TB and 500GB removed and they tested fine and no SMART issues. The replacement 8TB and 500GB report no SMART issues so I'm taking this to be a motherboard issue or a correct filesystem issue. Does it look like a filesystem or hardware issue on the motherboard side? Thanks in advance. unraid-diagnostics-20260319-2321.zip

It does appear to be a defective motherboard. Swapped it out with another of the same model and its been running for three days and no issues with the HBA. Thanks!

I recently replaced the CPU and motherboard from some older Z97 Intel motherboard to an Intel 265K with MSI Z890 Gaming Plus Wifi motherboard. I have an 8-port LSI HBA installed that was initially working, but now it isn't being found by unraid and so my array won't start with the missing drives. I do not know if this is related to me installing i915 related plugins to show iGPU or if this is completely unrelated and a hardware problem. I just swapped out the HBA and no difference. So next I'll be swapping out the motherboard (to rule out a defective PCIe slot) if I'm unable to solve this in BIOS or unraid driver detect changes. But I really hope it isn't the motherboard. Attached diagnostics. Thanks in advance! tower-diagnostics-20251213-1836.zip

Yes, it spins for a while and then stops. Hasn't synced since August. What? Are you referring to the Flash backup feature in the pic above? (Tip: try to avoid asking questions in the negative). I'm not trying to hijack this thread with a DIFFERENT problem, I have the same problem as listed in OP and the disabled Flash Backup in Unraid Connect cloud. 2025.11.21.0223, it says up to date. Not recently, but I believe so. If I did, it would have been months ago the first round of troubleshooting the issue. It's been upgraded twice with several reboots since this problem first happened. Perhaps if I KNEW what the problem that was supposedly resolved was, I'd maybe understand what was missed or different for me. I have two unraid servers and this issue only happened to one of them. Based on the image from chitalov showing that flash backup was working but then showing a 3 month old backup is not confirmation the bug is fixed. Nor is your post showing its activated but doesn't confirm whether the backup in Unraid Connect actually backed up recently. So are you just assuming the problem was fixed, or that since it resolved itself for one user that is must be fixed for everyone? Just trying to understand the scope of the fix since its not documented.

I don't understand what you are saying. Kindly point me to the release notes where this bug was resolved. Obviously I do still have an issue with flash backup or I wouldn't be frustrated by this continued issue, it makes people nervous. I am on 7.2.2 and the problem is NOT fixed.

Given that this feature is critical to proper and quick server restoration, its a very bad look for Limetech that they are not treating this bug with high priority. This is the type of bug that someone experiences a hardware failure (which we all know is inevitable with this flash drives eventually) and the user is initially thankful for having a flash backup option only to find out it has been broken for months with no notifications or warnings and their day becomes way harder and longer. @JorgeB Who are the representatives for Limetech that can comment on the lack of priority and communication to users for this critical bug? Regards

I have a Plus license with 12 drives. Can I purchase an Unleash key and replace the Plus key with the new Unleashed key through the webGUI so flash drive isn't changed), and then be able to use Plus key on a new USB flash/server? Thanks

Wow, I can't believe it. Looks like everything is working again! Thank you so much for your help!

Looks like /dev/nvme3n1p1 really is defective and I'll need to replace it. unraid-diagnostics-20250823-0111.zip

Label: none uuid: 9e7e00da-48c5-4dae-9371-25e83b91f33c Total devices 2 FS bytes used 466.13GiB devid 4 size 1.86TiB used 1.86TiB path /dev/nvme2n1p1 devid 5 size 1.86TiB used 1.77TiB path /dev/nvme3n1p1

I have two 2TB NVMe drives in btrfs raid 1 for cache. nvme2n1 and nvme3n1. nvme3n1 was reporting errors and a scrub couldn't fix it. I wanted to remove nvme3n1 from the pool so I stopped the array, removed the pool, added pool back with 1 slot and nvme2n1, but when I go to restart the array, I get: Cache INTEL_SSDPEKNU020TZ_BTKA312103AM2P0C - 2 TB (nvme2n1) 33 °C 0.0 B/s 0.0 B/s 0 btrfs Unmountable: wrong or no file system Is there any way to get the array to restart with the supposedly good nvme2n1, or is it also corrupted beyond repair and need to replace and restore from backup? Thanks in advance unraid-diagnostics-20250822-1639.zip

All good, thanks very much for walking me through all the steps. Much appreciated!

That seemed to go ok. Thank you again! unraid-diagnostics-20250309-2049.zip

# btrfs balance start -f -dconvert=single,soft -mconvert=single,soft /mnt/cache Done, had to relocate 1197 out of 1439 chunks Thank you, again. Note, it has a 2TB drive, so that 4.1TB is odd for only showing 1 drive. unraid-diagnostics-20250305-1351.zip

Sorry, got busy and only getting time to try now: # btrfs balance start -dconvert=raid1,soft -mconvert=raid1,soft /mnt/cache ERROR: error during balancing '/mnt/cache': Invalid argument There may be more info in syslog - try dmesg | tail

Thank you. unraid-diagnostics-20250218-0015.zip

So I did as you last stated, and after selecting 1 slot, (nvme2n1) device and start the array, I get: And the following errors: I shutdown, put the original 2TB Intel NVME back in and started up. The array was stopped and expecting cache slot 1 to be assigned. I assigned it the Intel nvme (taken offline and tested and cleared), and I get: So it looks like I'm back where I started. After removing the pool, maybe something else needs to be done to make it forget the old profiles before adding the single cache drive back? Thanks!

In the first step, "on main click on the first device for that pool and then "remove pool"", "REMOVE POOL" is greyed out and cannot be selected. Do I need to stop array first?