I have 2 pools, cache with only 1 1TB SSD as the write cache and nvme with 2 256GB SSD to run containers from.

I think all of them might be giving up the ghost and I would like some confirmation before I find replacements.

Please check the attached diagnostics.

Thank you in advance for any advice.

tower-diagnostics-20251017-0944.zip

Btrfs is detecting data corruption on both pools, start by running memtest.

Thank you @JorgeB, will do and report back later.

I got bad RAM indeed, memtest86+ did the trick. Replacements are on the way.

What about the SSDs though, is it too early to say if they are also bad?

SSDs should be fine; at least I don't see anything for now suggesting they are the problem. Once you get the RAM issue resolved, scrub both pools, and if any corrupt files are found, you should delete/restore them from a backup until no more errors are found.

I swapped in another pair of old but good DDR4 sticks, and the server booted up fine. Scrubbed the pools, deleted the corrupted files (1 was an Appdata backup tar file, kinda ironic). Parity check is still running but so far so good. Containers are back up for now. Thanks for the help as always, @JorgeB !