JorgeB

Used to be WD and still have a lot of them, but after several similar failures now buy mostly Toshiba, as a bonus they are usually cheaper, at least in my country.

You can just go to Tools -> System Devices and search for "LSI", e.g.:

It's a normal rebuild, but to try and recover the partition on the old disk you'd need to use a new disk for this, not much point in doing a parity check after since the rebuilt disk will reflect current parity (plus the other disks). First see if testdisk can recover the old partition, can't help with that part as I've not used testdisk in a while, but there are instructions in author's site, as well as other tutorials you can find with Google, I can help with the UD part if the partition is recovered, just post back here.

OK, just tried this and it does work for the "display all" feed, looks like it doesn't respect the other setting, would still prefer how it was before but at least now I can get around it by clicking on the ball, just need to get used to that.

Yes, that's how it always was for me, not now, and I didn't change anything, AFAIK you can change the default "all activity" display, look for example this thread, there's your last post, but not Brit's, only his reaction. I didn't try to click on the ball for this because I don't usually need to because of this setting, it still works when browsing the forum by section:

This started recently, likely after some forum update, all posts from the same thread are now grouped together in the feed and only the last post is shown, meaning that when you click on it you'll go directly to the last post on that thread, not your first unread post in that thread, so if there were multiple posts I need to scroll up until I find the "Unread replies" line, before if there were multiple post in the same thread they all appeared in the feed, so going bottom to top it would take me directly to the first unread post in that thread, any chance this can go back to how it was?

First you need to rebuild the emulated disk, and parity will agree with that rebuild. You can still try to recover the partition on the actual disk with testdisk, and if successful you can mount it with UD after changing the XFS UUID, that can be done in the UD options.

Previous issue was with the cache, not the array drives, just downgrading shouldn't cause any issue with those but post new diags so we can't see what the problem is.

It's also logged as disk issues, wait for the SMART test to complete, note that these errors can sometimes be intermittent, also good to monitor this attribute for WD drives, if it keeps climbing there will likely be more read errors soon: ID# ATTRIBUTE_NAME FLAGS VALUE WORST THRESH FAIL RAW_VALUE 1 Raw_Read_Error_Rate POSR-K 200 200 051 - 3519

Next time please post the complete diagnostics instead, but both disks are reporting recent UNC@LBA errors, you should run an extended SMART test on both.

You should downgrade back to v6.9, looks like your hardware and the newer Linux kernel don't get along.

For now I only see a filesystem problem, not a device problem, but if the filesystem is getting frequently corrupt you likely have an underlying hardware issue.

I don't use SAS devices, we can ask @WardyBoi

If you have another PC try the HBA there to make sure it's working.

1 pass should be a few hours, less if errors are detected soon.

Cache filesystem is corrupt, you should backup and re-format.

The pool failed to remove the missing device due to data corruption, if you haven't yet you should run memtest, then see here to monitor the pool for future errors.

Docker image corrupted because it ran out of space, just recreate it.

Still think that most likely the the disks were never converted, filesystem for the array disks was set to reiser, changing hardware wouldn't change that, but if they don't mount xfs your only option is to convert (again).

Now that I think about it I believe a long time ago I could mount a disk reiserfs after formatting it with xfs, stop the array, click on all the disks and change the filesystem from reiserfs to XFS, see if they still mount, there could be issues with filesystem since mounting/unmounting will always generate some writes.

Same but without -n or nothing will be done.

Those don't seem very likely, but honestly except for spin down not sure what else could cause that, if you have another PC available you can try running it there.

Forgot to mention, if the filesystem in the emulated disk can't be fixed or there's a lot of lost+found files you can try using testdisk on the actual disk to recover the deleted partition, then try to mount it outside the array with UD.

That's impossible, you must be mistaken and they were never converted to XFS.

Try a different bay first, if still issues try a different port or controller.