Having trouble with my cache. 

 

Cache is two 1TB SSDs. Original symptom was docker failing and unable to start. Find Common Problems stated the cache was read-only. I assumed (probably incorrectly) that one of the drives was failing. 

 

After disabling docker and vms, I changed the appropriate shares to Cache: Yes and ran the Mover. Not all data was able to move with errors in the mover logs. Once the mover finished, I removed the (I thought) offending drive from the cache pool and allowed BTRFS to being rebuild, but that failed and again forced it into readonly. 

Jan 11 21:25:16 galaxy kernel: BTRFS info (device sdk1): balance: ended with status: -5
Jan 11 21:25:16 galaxy kernel: BTRFS info (device sdk1: state EA): forced readonly

 

Full diagnostics attached. Starting to think my cache is corrupted. I'm happy to jettison the cache and start fresh, but i'm not exactly clear where to begin. Yes, I ran this server with overclocked memory and I may be paying the price for that now. I have removed the memory overclock since the diagnostics were taken. 

 

Thanks

 

galaxy-diagnostics-20230111-2137.zip

Jan 11 21:21:41 galaxy kernel: BTRFS info (device sdk1): bdev (efault) errs: wr 0, rd 0, flush 0, corrupt 15, gen 0
Jan 11 21:21:41 galaxy kernel: BTRFS info (device sdk1): bdev /dev/sdk1 errs: wr 0, rd 0, flush 0, corrupt 10, gen 0

 

Btrfs is detecting data corruption on both devices, this usually points to a RAM problem, recommend running memtest, then probably best to backup and re-format the pool.