JorgeB

Then it's not correctly flashed, it's not even using the mpt3sas driver in RAID mode, it's using the MegaRAID driver, that's for RAID controllers only.

It's normal if there are a lot of small files, like Plex metadata for example.

This is after swapping old RAM with new one correct? Or just using different settings?

That's a RAID controller, see here for a list of recommended models:

Any new corruptions can only be new, scrub scan all the data and compares checksumns, if they fail later it's new. I would say no, but of course can't say it with 100% confidence, about 99.99%.

Same happened to a lot of users, e.g., there's info on the linked thread on how to change it.

Unlikely, and re-creating the flash your fix it anyway, use the USB tool. Yes, but first get the server booting.

If it's not RAM board and CPU would be the next suspects.

Only the last one was a correcting check, so it's expected to find exactly the same errors, next one should find 0 errors.

If you have any forwarded ports from your router to the server make sure you close them all, a number of users have been similarly hacked lately.

Most likely, a number of users have been similarly hacked lately.

There are no 8TB Samsung drives, you can post your diagnostics if you want more informed advice.

Cache is still corrupt, you're sure it was recently formatted? Need to do it again, and if it was just formatted and it's already corrupt it suggests an underlying issue.

Can't you access the WebGUI? By name or IP?

Several disks are missing the partitions, that suggest they were deleted by someone, probably when you where hacked. A number of user have been similarly hacked lately, did you have any forwarded router ports (or DMZ enable) to your server? Did you have a strong password for root?

You can try this and then post that log.

A number of user have been similarly hacked lately, did you have any forwarded router ports (or DMZ enable) to your server?

Looks to be Nvidia related.

I also have it so yes. If there are doubts about hacking I would backup current flash, recreat it then restore only super.dat (disk assignments) and the key, boot the server and start array and make sure all data is there, then reconfigure the server.

It shouldn't, but you can try enabling it.

There are macvlan call traces, these are usually the result of having dockers with a custom IP address, more info below. You can reboot form the console if needed, by typing "reboot"

Mar 18 17:14:12 Servo root: ERROR: system chunk array too small 34 < 97 Mar 18 17:14:12 Servo root: ERROR: superblock checksum matches but it has invalid members Mar 18 17:14:12 Servo root: ERROR: cannot scan /dev/sdc1: Input/output error This is about parity, the error itself can be ignore since parity doesn't have a filesystem, and I don't think it would cause any other issues. Mar 18 17:14:17 Servo kernel: BTRFS info (device md2): bdev /dev/md2 errs: wr 0, rd 0, flush 0, corrupt 2666, gen 0 Disk2 is showing data corruption, you should run a scrub. As for the LSI, it's using the MegaRAID driver, I would recommend flashing it to IT mode, note that it will require a new config since the disks IDs will change, then see if the related errors go way or not.

You didn't need to rebuild the disk to see the contents, does data on that disk look OK after the filesystem repair? Any lost+found folder(s)?

Yes, like jonathanm mentioned most boards use "fakeRAID", you can use something like an LSI MegaRAID controller to have raid volumes assigned as devices, I used one for some time mostly for testing and it was fast, but not having easy access to SMART data was a deal breaker for me.