I bought a new HDD and had to reconfigure some things in my case last night to add it and tidy things up, and after rebooting, my two-SSD BTRFS cache started throwing errors (although both were existing and working fine before).

 

I unplugged and replugged them back in to make sure everything was seated correctly and they were. Any thoughts on what might be happening here?

 

emrys-diagnostics-20200828-0725.zip

One of the cache devices (cache1) dropped offline at some point in the past:

 

Aug 28 06:35:33 Emrys kernel: BTRFS info (device sdf1): bdev /dev/sdf1 errs: wr 16291, rd 8662, flush 349, corrupt 0, gen 0

 

See here for what to do and how to better monitor the pool.

3 minutes ago, johnnie.black said:

One of the cache devices (cache1) dropped offline at some point in the past:

 

Aug 28 06:35:33 Emrys kernel: BTRFS info (device sdf1): bdev /dev/sdf1 errs: wr 16291, rd 8662, flush 349, corrupt 0, gen 0

 

See here for what to do and how to better monitor the pool.

 

Thanks for the response. I was already using that script, which is what notified me there were errors.

 

I may have accidentally unplugged one of the cache SSDs while the system was still on (long story), so I reset the stats per the post you linked to and will wait to see if any further errors occur (unless there's something I manually need to run to fix it after having run into an issue like that).

You should also run a scrub and make sure all errors were corrected, mentioned on the linked post.

2 hours ago, johnnie.black said:

You should also run a scrub and make sure all errors were corrected, mentioned on the linked post.

Can you advise what running a scrub is or how to do it? I'm not sure what that means and the linked post doesn't go into details into what running a scrub entails other than correcting issues.

Scrub checks data integrity, and it will bring the dropped SSD up to date with the other one, you can run it by clicking on cache on the main page, then scrub.