RAID 5 slows down after a few weeks

A little disappointed. Guess I expected more information from dmesg. Unless I'm missing something.

How do I get to place where I can even try these things? Guides are telling me that I need to find a Shell and that in order to find a Shell I need to get to the Terminal, which appears to be on some kind of desktop. But all I've ever seen is the console pictured above.

I don't know how to get there.

Found that a bunch of my movies on the NAS are corrupted after the outages, one of them so severe that the system spent about three hours recovering itself. That was about two months ago. Now I'm remuxing them all to see any error messages in MKVToolNix, and ripping or restoring from other backups the faulty ones. I have my NAS connected to an UPS now. But in the process of remuxing them and moving them to and from the NAS, the server became inaccessible again. Is it safe to simply shut the NAS off with the power button on the case? Because Windows/Putty can't even find the IP address. I don't want to turn it off that way, considering sudden power losses probably led to all this, but I have no other way to turn it off.

The data speeds are probably fine. I overreacted. They don't always stay at 113 Mbps, but they don't drop to 0 after replacing the power supply and the movies haven't stuttered or stopped during playback since.

Forcefully restarted.

"[ 16.111199] FAT-fs (sdb1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck."

When I type in fsck, it only says...

[ 17.445639] r8169 0000:06:00.0 enp6s0: Link is Up - 1Gbps/Full - flow control off

[ 17.445652] IPv6: ADDRCONF(NETDEV_CHANGE): enp6s0: link becomes ready

[ 81.620934] EXT4-fs (md0): mounted filesystem with ordered data mode. Opts: u ser_xattr,usrjquota=aquota.user,grpjquota=aquota.group,jqfmt=vfsv0,acl

[ 82.439250] softdog: initialized. soft_noboot=0 soft_margin=60 sec soft_panic =0 (nowayout=0)

[ 82.439253] softdog: soft_reboot_cmd=<not set> soft_active_on_bo ot=0

root@openmediavault:~# ^C

root@openmediavault:~# fsck

fsck from util-linux 2.33.1

e2fsck 1.46.1 (9-Feb-2021)

/dev/sdb2 is mounted.

e2fsck: Cannot continue, aborting.

root@openmediavault:~# ^C

root@openmediavault:~# fsck

fsck from util-linux 2.33.1

e2fsck 1.46.1 (9-Feb-2021)

/dev/sdb2 is mounted.

e2fsck: Cannot continue, aborting.

Been dealing with the lack of on-board graphics and this other shit for so long that I stupidly forgot I can just use a monitor. Was using this better SATA expansion card made for PCIe 3 and intended to use a cheap PCIe 2 to 3 adapter for the video card (GT 1030 doesn't need the full port.), but the speeds are the same as with the old SATA card made for PCIe 2. Because that new SATA card arrived weeks before the PCI adapter and the motherboard only has one PCIe 3 lane, I had no GPU plugged in. Interesting thing is that when I used the old SATA card for my PC, most of the SATA devices became inaccessible, but when I use the PCIe 3 SATA card on my PC, everything shows. Was worried I'd never be able to use an NVMe SSD with this motherboard, since that deactivates the two SATA ports on the same lane. So I'm happy to know that with this SATA card intended for my NAS I will be able to install an NVMe on my PC. While I keep all the hard drives on the NAS connected to PCIe 2 and plug the video card right into the motherboard rather through through the adapter cable.

Wish I knew of an easier way to see what movies are corrupted. Worried about it happening again.

Never mind. When the NAS is unreachable on the network, nothing shows up on the monitor connected to it either. So having a monitor there is completely useless to me. Tried disabling some things in the BIOS, like XMP, but it still randomly becomes unavailable.

Had to forcefully restart again. Again the server was unreachable and nothing showed on the monitor. But whatever process prevented me from unmounting the array and running fsck on my last few attempts stopped running after the restart. fsck told me nothing useful.