JorgeB

The only error I see appear to be trim related, do you see any errors without trying to trim the pool?

These two files are corrupt and can't be fixed: Dec 15 09:48:12 Tower kernel: BTRFS warning (device sdf1): checksum error at logical 89372102656 on dev /dev/sdf1, physical 39979978752, root 5, inode 184569, offset 379801600, length 4096, links 1 (path: plex/Youtube/TinyDesk/Diane Coffee - NPR Music Tiny Desk Concert.mkv) Dec 15 09:48:13 Tower kernel: BTRFS warning (device sdf1): checksum error at logical 547320229888 on dev /dev/sdf1, physical 135003369472, root 5, inode 180906, offset 363618304, length 4096, links 1 (path: plex/Youtube/Cercle/HAUTE live @ Cafe A for Cercle.mkv) Delete them and restore from backups, you can then run the balance, also good idea to run memtest.

Macvlan call traces are usually the result of having dockers with a custom IP address, see here for more info:

It also auto disables turbo write once i/o to multiple disk is detected.

Looks like you're using the most free allocation method, don't use that for best performance as writes to parity will overlap and be much slower.

Even if it's fixed the first check after you change something might still correct errors, so you should always run it twice, if the 2nd one finds more errors it's not fixed yet.

Start with this: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173 If all is as it should and it still crashes enable the syslog mirror/server.

Adding --no-perms to the rsync command should get ride of those, like mentioned FAT doesn't support permissions.

Also good idea to run a single stream iperf test to rule out a network issue.

IIRC it's a problem with Smartmontools, it doesn't show all the info for some SAS devices, including temp, older (and possibly newer) releases should.

It passed so it's fine for now, recommend replacing/swapping cables just to rule them out and if the emulated disk contenrs look correct rebuild on top.

Try upgrading to rc1

There are checksum errors, i.e., data corruption, run a scrub and post news diags after.

This is a very weird one, can't see any reason for the NVMe devices not initialize, can you try booting with a new trial key just to make sure it's nothing with the config?

Like mentioned before this is not a bug, it's a general support issue, and I also did mentioned that most likely trim wouldn't work with that LSI, if you still need support you should start a thread in the general support forum.

See here: https://forums.unraid.net/topic/53433-ssds-as-array-drives-question/?do=findComment&comment=522482

Cache2 is already not part of the pool, despite what the GUI shows, you just need to stop the array, unassign cache2, need to click the check box to allow array start and start the array, no balance will be needed.

You should free up some space, then convert to single, or the balance might fail.

Using onboard SATA ports + HBA if needed. Should be OK, but some USB enclosures create a custom partition or use a custom ID that might not be recognized by Unraid on a SATA port, it should still work with UD if needed.

Cache is completely full, it's also using dual data profiles, you should get a notifications about that (if enable), you need to free up some space and balance it.

Yeah, not sure about this also.

The trim itself is the command, behavior after trim depends on the device.

Yes, all appears to be OK with the pool, don't know where the notification comes from.

From what I remember besides the SATA controller it was only passing-trhoug a USB controller and a GPU, so it should be fine, but you should check.

It's not up to the controller, if up to the device (for SATA): https://en.wikipedia.org/wiki/Trim_(computing)#ATA NVMe devices are different.