JorgeB

You can the NVMe device as cache, but as single device only.

It's not, it's missing: Data Metadata System Id Path RAID1 RAID1 RAID1 Unallocated -- ---------- --------- --------- -------- ----------- 1 /dev/sdad1 229.00GiB 2.00GiB 32.00MiB 216.10GiB 2 /dev/sdag1 83.00GiB 2.00GiB 32.00MiB 138.54GiB 3 missing 146.00GiB - - -146.00GiB -- ---------- --------- --------- -------- ----------- Total 229.00GiB 2.00GiB 32.00MiB 208.64GiB Used 197.49GiB 486.09MiB 48.00KiB It likely dropped offline at some point in the past but it's not being deleted possibly because of checksum errors, please run a scrub on the pool and post the results.

If you already replaced cable it could be a controller issue, if you still have free onboard SATA ports try using those for as many devices as possible, other than that try a new HBA if available.

See the FAQ.

We can now see SMART for disk3 at it really appears to be failing, since you already have a disabled disk and only single parity it can't be rebuilt, xfs_repair can't fix a disk with bad sectors, best bet is to use ddrescue to clone the disk then run xfs_repair.

SAS expander don't have drivers, and if rolling back didn't fix it it's not a HBA driver/Unraid issue, besides you have two identical HBAs and both are being detected and initialized correctly, just one of them is not detecting any connected devices, this is a hardware issue, check connections from the HBA to the second expander, also check power on that expander.

No, but if errors were/are found you'll need to run a correcting check, in that case you can cancel that one and start a correcting one before or after fixing the filesystem.

Check filesystem on disk1: https://wiki.unraid.net/Check_Disk_Filesystems#Checking_and_fixing_drives_in_the_webGui

Also, like mentioned above, it's not always true with this NIC, first google result: https://community.intel.com/t5/Ethernet-Products/Intel-PRO1000-PT-network-card-not-showing-up-on-PCIe-bus-one/td-p/277091

Never tried but should be OK.

I guess the mover goes in alphabetical order, and appdata comes first.

It's not, it's trying to move some data from disk2 to cache.

Samba is more chatty with -beta30, and don't know exactly what that error means but it appears to be related to the enhanced macOS interoperability, do you have that enable? If yes and it's not needed try disabling it.

@trurlmight know better why both appear on the diags, don't remember the exact circumstances that cause that. For the mover issue, please enable mover logging, run the mover and post new diags.

Disks can fail at any time, you can write zeros to the entire disk, it might fix it, but possibly only temporarily, use for example pre-clear, then run a read test or and extended SMART test.

What share is using cache that you want moved? You also appear to have shares downloads and Downloads, a setting for both exists on flash and it will only work for one, delete the one that shouldn't be there, also make sure only one of those exists on all the disks, Linux is case sensitive.

That disk is failing: 197 Current_Pending_Sector -O--CK 200 200 000 - 1

Can't download diags.

Please post the diagnostics after a format attempt: Tools -> Diagnostics

It really looks like a connection problem, but if you already replaced cables disk3 might be failing, and in that case there's no option to recover either disk, but I would still try another cable, swap with another disk.

Disk3 dropped offline, and since there's already a disable disk contents can't be emulated, power down, check/replace cables on disk3, power back up and post new diags.

Driver is almost certainly included, but without the NIC being detected nothing that can be done on Unraid side, this is a hardware problem. You can try setting PCIe to 1.0/1.1 on the board BIOS so see if it makes any difference.

Any partial/lost files will be moved to the lost+found folder, you can then analyze them, again parity can't help with filesystem corruption.

You can mount each disk with the UD plugin (use the read-only mode), if there's just one parity there should be only one disk without a filesystem, that would be parity, do a new config and re-assign all disks (data disk order is not important with single parity) and check parity is already valid before array start, if you have dual parity post back since there are a few differences.

I’ve been ruminating on this SAMBA aio issue because the very large read performance difference first reported by @trypowercyclereminded me of an issue I’ve seen before, but I was having trouble finding that post, now I know why, because those forums are gone? I did finally find it in my content: And this is the comparison I posted at the time: So I believe I noticed this issue at around the same time aio was introduced in Samba, and at the time disabling smb3 fixed it, now I wonder if it was already the same issue and disabling smb3 was also disabling aio, symptoms are very similar, and the problem wasn’t controller related but device related, some brands/models perform worse than others, so I now did some more tests with –beta30 and different disks. Ignore the normal max speed difference from brand to brand, I used whatever disks I had at hand, so some disks are older and slower than others, the important part is the aio on/off difference, tested with disk shares so no shfs interference, all connected to the same Intel SATA controller, each test was repeated 3 times to make sure results are consistent, read speed reported by robocopy after transferring the same large file. I think the results are very clear, and by luck (or bad luck) the tests this past weekend were done with the only disk that now doesn’t show a significant difference, note that I don’t think this is a disk brand issue, but a disk model issue, likely firmware related, possibly worse in older disks? I know you already plan to leave aio disable, but still one more data point that I believe really confirms it should be left disable.