TS-431 feels unstable, is it expected behavior or should I replace it?

[jollino]

Hello all, I've been lurking for a while and figured I'd ask your opinion about my current QNAP setup, as you all seem much more knowledgeable in the field of QNAP devices than I am.

TL;DR: 5-year-old TS-431 with extremely lightweight usage feels unstable (crashes when copying files to it via SMB, always says the volume isn't clean, etc.). Is it normal or should I replace it?

I currently have a TS-431 that I bought in December 2015, together with two 5 TB WD Red (WD50EFRX) disks in a RAID1 configuration, and In 2018 I added two 6 TB WD Reds (WD60EFRX) — couldn't find fivers at the time — and upgraded to RAID5. To the best of my knowledge, they're all CMR but I'm not 100% sure about the 5 TB ones.

Note that I use my NAS mostly as a convenient repository of cold-ish data (with further copies elsewhere) rather than to access live data directly through the mounted shares. That also happens, but it's mostly a bucket for stuff to keep out of the way but within reach.

The point is that it just feels "unstable". Maybe it's because the TS-431 is admittedly quite limited with just half a gig of RAM, maybe it's just old and the newer firmware updates just weigh it down, but I often feel like I'm walking on eggshells with this setup. Performance is decent for my needs but the whole experience feels less than solid. Let me explain.

A while back it started rebooting randomly when doing RAID scrubbing, which I had set up to do once a month. A long thread on the official QNAP forum that I opened, and remote diagnostics by QNAP via the built-in app led to no odd findings and a clean bill of health. I eventually just disabled scrubbing and moved on with life; after all my usage is extremely light.

Then at some point it started rebooting randomly when moving lots of data to the NAS via SMB from macOS. I suspect this is an issue with Apple's own SMB implementation, but the fact remains that it's the NAS that for some reason panics and reboots. Oddly enough, using AFP there's no such issue, so I've reverted back to using AFP for the time being. It doesn't make me feel very safe to know that the most basic use case of a NAS using the most common protocol (as buggy as the implementation on macOS may be) makes it die. A server just shouldn't crash when a client is out of specs. I have made sure that the code page and charset settings are all properly set.

Also, I'm not sure if this is normal, but every boot takes about ten minutes. It just spends an insane amount of time doing something at the very beginning, during which it doesn't even respond to pings.

Moreover, lately it has begun complaining that the volume wasn't unmounted cleanly upon every reboot, despite only ever rebooting cleanly via the web interface. I read somewhere that QNAP claims it's a kernel issue and a full restore is needed to sort it out, which is appalling. I currently just don't have the extra storage to copy everything out of the NAS, restore it, and copy it back. I do have backups, but they're not as convenient right now for something that shouldn't even be needed.

The disks themselves appear to be perfectly fine. All disk have an acceptable temperature despite the current heatwave, show "Good" for both I/O and SMART, all SMART values are also good, and complete tests on all of them report no errors. The NAS, despite its limited power, shows 300+ MB free out of a total of 479 MB of RAM (no idea where the missing 33 have gone...). CPU usage stays well below 10% when not transferring data. I only keep the most basic apps and services on it as I frankly don't need the rest.

I understand that QNAP caters to users with more intensive use cases, so I'm clearly in the minority here. It's just that sometimes I feel like I'm walking on eggshells, especially as each reboot takes so long and the system then claims that the volume the disk wasn't unmounted properly, not to mention the random crashes via SMB.

As I'm on the edge of upgrading the disks as I need more space (probably two at a time given the current prices — going with 8 TB Ironwolves), I can't help but wonder: is something wrong with my NAS that I should focus on, or is it just the expected behavior from a low-end, five-year-old machine? Should I just replace the whole thing with something newer, or maybe even build my own mini-ITX rig with TrueNAS on it?
Or, if I were to physically remove the disks and restore the NAS to factory settings, would I be able to re-add the RAID 5 volume across those disks afterwards?

Any advice would be most welcome. Thank you in advance!