JorgeB

It's the name of the device.

Like mentioned, that's not possible, you wouldn't be able to start the array. You could try this without much risk (but array will be unprotected): -stop the array -unassign disk15 (if this is the one supposedly rebuilt) -start the array See if the emulated disk 15 mounts, if it doesn't post new diags.

No, what I'm saying is that something is missing from your description, what you say happened is not possible, and without a complete description of what was done or the logs it's very difficult to help.

Something is missing here, Unraid wouldn't let you replace one disk and assign a new one one at the same time.

It's not, and you don't want ever to do that, formatting is never part of a rebuild, what I was asking was if it was ever formatted before replacing it, the one that was replace, assuming it was only one, since you have 2 unmountable disks.

Syslog doesn't show the rebuild, but both disks 15 and 16 are unmountable due to an invalid partition, that error is usually reserved for disks that were never formatted, since they are presumably the last disks you added to the array are you sure they were ever formatted?

It's a known issue with Samsung SSDs and some AMD chipsets, but Samsung SSDs are very picky with cable quality in general.

We need the diagnostics: tools -> diagnostics

Errors are on both ATA9 and ATA10, both ports of the same Marvell controller, could be cables, but since it's Marvell it could be the controller, replace cables and see if they go away.

When you first booted there was only one device for that filesystem, can't say what happened to the missing one(s) since whatever it was happened before current boot, but you then recreated the pool and it looks to be fine now, but to confirm post output of: btrfs fi usage -T /mnt/disks/test

Looks mostly fine, there a single Reported_Uncorrect error, unless it keeps increasing it's fine, but you should run regular parity checks, once a month is usually the recommendation, it will catch any issues since it does a complete surface scan on all disks, basically same as an extended SMART test.

These sync errors are unrelated to a disk problem. Which disk is it? You have too many for me to look at all SMART reports. Sync errors are always corrected on parity, never on data drives. No, it is possible is some rare cases to corrupt parity if there are read errors on a disk during a correcting check, hence why it's recommended that parity checks should always be non correct unless errors are expected, like after an unclean shutdown.

Apr 17 16:28:45 Tower emhttpd: unclean shutdown detected A few sync errors are normal, even expected, after an unclean shutdown.

Errors on parity also look like a power/connection issue, it dropped offline.

Apparent connections issues with cache2, replace cables.

One request if you don't mind, I would like to see the devices grouped more closely together, similar to the array devices, so they didn't take so much vertical real estate and more could fit on the page, e.g. 8 arrays devices vs 8 UDs: Also not a very big fan of the new color, maybe because they are links now? But not a big deal.

You can recover the space with trim, see here.

Already solved here:

Please post the diagnostics: Tools -> Diagnostics

Apr 13 22:59:55 CHASM kernel: usb 1-1-port4: disabled by hub (EMI?), re-enabling... Apr 13 22:59:55 CHASM kernel: usb 1-1.4: USB disconnect, device number 16 Apr 13 22:59:55 CHASM kernel: usb 1-1.4: new low-speed USB device number 17 using ehci-pci Apr 13 22:59:55 CHASM kernel: hid-generic 0003:0764:0501.000F: hiddev96,hidraw0: USB HID v1.10 Device [CPS Value2200E] on usb-0000:00:1a.0-1.4/input0 Apr 13 23:00:01 CHASM apcupsd[30039]: Communications with UPS restored. USB ports is getting disable and the device keeps disconnecting and reconnecting, try a different USB port.

You can mirror he syslog to the flash drive or to the array, AFAIK you can't use devices outside user shares.

Try this.

Cache filesystem is corrupt again, if it's happening multiple times without an apparent reason you likely have some hardware issue.

Those are system devices, not errors.

Please post the diagnostics: Tools -> Diagnostics