My apologies I tried looking through a lot of resources regarding, but I just didn't understand a lot of it. My cache drive is showing:

Cache Samsung_SSD_870_EVO_1TB_S6PTNM0RB07354A - 1 TB (sdb) * 0.0 B/s 0.0 B/s 0 auto Unmountable: unsupported or no file system

I assume it is dead, but some of the topics I saw indicate it may be revivable. Either way, when it died all my docker containers are back to default, they are missing all the settings/data. How do I go about fixing this from here? I attached the diags here as well.

Any help would be greatly appreciated.

blightednas-diagnostics-20251018-1311.zip

Edited October 20, 2025Oct 20 by samo_
added version prefix

There are constant ATA errors with the cache device; replace both cables and post new diags after array start.

Thanks for the reply! I replaced the cable for the cache drive with one that has been working in another pc, it still shows as unmountable. There are SMART errors on another drive (sdf part of the main array), I didn't replace that cable I believe the drive is legit dying, a replacement came in the mail this weekend but I don't want to change anything else before resolving the cache/docker/etc issues in fear of losing anything.

I attached the latest logs here. Thanks again for looking at this!

blightednas-diagnostics-20251020-1145.zip

Still constant errors; if both cables were replaced, it may be a bad device.

So from my understanding, all the cached stuff gets moved to the main array by mover at some interval right? Like the missing settings/configs for all my docker containers?

I don't have an ssd I can throw in there to replace that drive right now. How can I remove the drive and cache pool and have all the containers point back to the configs in the array or whatever?

Then I will order some ssds for the cache, how would I set them up that if one of the drives die I won't run into the container issues again? And are there any recommendations for specific ssd brands, sizes or how many for redundancy?

Last question then, what beer do you drink?

18 minutes ago, samo_ said:

So from my understanding, all the cached stuff gets moved to the main array by mover at some interval right?

That will depend on the share settings for the shares that are going to the pool, and typically, Docker related files remain on the pool.

19 minutes ago, samo_ said:

How can I remove the drive and cache pool and have all the containers point back to the configs in the array or whatever?

You can remove the pool and see if the container comes up without it; it, probably won't, but if they are pointing to /mnt/user they would if they really existed on the array.

Well shoot, yeah it looks like everything is gone. After removing the cache pool I checked /mnt/user and there are configs but they are all blank or default.

Is there anything I can do at this point to try and recover anything from that cache drive?

I would try to clone the SSD with ddrescue:

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-525075

Thanks for your help! ddrescue is an awesome tool. I didn't have a device to clone to, so I ended up writing it out to an image on my main array. used losetup to assign it to a loop device, mounted it and boom all my files were there!