Cache oddness

[dsmith44]

Hello,

 

I had an issue with an SSD that is part of my three drive cache, I think due to cabling, lots of CRC errors

 

Reseating all the cables and booting back into rescue mode I ran Btrfs scrub on the device and it corrected some 1,000,000 errors correctly, no failed repairs.

 

Rebooted into unraid full and it for some reason then ran a resize for two hours then settled down again.

 

Finally I have a situation where the actual Btrfs seems fine although in need of a balance, but unraid thinks only two of the three disks are actually in the cache.

 

Smarctl -t on affected SSD is showing nothing.

 

Not quite sure how to progress, any ideas?

 

Thanks

 

root@unraid:~# btrfs fi show /mnt/cache
Label: none  uuid: 80776be4-065e-404e-a103-acb4b5026956
	Total devices 3 FS bytes used 322.01GiB
	devid    1 size 256.17GiB used 256.17GiB path /dev/sdi1
	devid    2 size 447.14GiB used 70.00GiB path /dev/sdg1
	devid    3 size 447.14GiB used 322.17GiB path /dev/sdh1

root@unraid:~# btrfs fi usage /mnt/cache
Overall:
    Device size:		   1.12TiB
    Device allocated:		 648.34GiB
    Device unallocated:		 502.11GiB
    Device missing:		     0.00B
    Used:			 644.03GiB
    Free (estimated):		 251.72GiB	(min: 251.72GiB)
    Data ratio:			      2.00
    Metadata ratio:		      2.00
    Global reserve:		 240.80MiB	(used: 0.00B)

Data,RAID1: Size:322.14GiB, Used:321.47GiB
   /dev/sdg1	  69.00GiB
   /dev/sdh1	 320.14GiB
   /dev/sdi1	 255.14GiB

Metadata,RAID1: Size:2.00GiB, Used:553.47MiB
   /dev/sdg1	   1.00GiB
   /dev/sdh1	   2.00GiB
   /dev/sdi1	   1.00GiB

System,RAID1: Size:32.00MiB, Used:96.00KiB
   /dev/sdh1	  32.00MiB
   /dev/sdi1	  32.00MiB

Unallocated:
   /dev/sdg1	 377.14GiB
   /dev/sdh1	 124.97GiB
   /dev/sdi1	   1.01MiB
root@unraid:~#

 

unraid-diagnostics-20190930-2242.zip

Edited October 1, 2019 by dsmith44

[JorgeB]

Stop the array, if Docker/VM services are using the cache pool disable them, unassign all cache devices, start array to make Unraid "forget" current cache config, stop array, reassign the 3 cache devices, re-enable Docker/VMs if needed, start array.