Hi,

 

I've started to get some errors lately, and I think it's one of my cache drives that's creating problems.

 

I've got no clue what to look for, so maybe some one here can help me?

 

Attached is the latest diagnostics from the server.

 

Best regards,

berta-diagnostics-20180322-1954.zip

Read errors on one of your cache devices (cache4):

 

Mar 22 19:46:52 Berta kernel: print_req_error: I/O error, dev sdz, sector 20476128
Mar 22 19:46:52 Berta kernel: BTRFS error (device sdu1): bdev /dev/sdz1 errs: wr 610, rd 41, flush 0, corrupt 0, gen 0

 

Likely a bad cable:

 

199 CRC_Error_Count         -OSRCK   098   098   000    -    1253

 

Replace SATA cable.

 

 

Also:

 

Mar 19 10:48:31 Berta kernel: md: disk2 read error, sector=5269787856
Mar 19 10:48:31 Berta kernel: md: disk2 read error, sector=5269787864
Mar 19 10:48:31 Berta kernel: md: disk2 read error, sector=5269787872

 

Check cables and run an extended SMART test on that disk.

Thanks,

 

I'll have a look at i later today! 

I finally got some time to look at the problem. I did the sugested things and now the log is floating with "BTRFS csum failed".

 

Anything I can do?

berta-diagnostics-20180429-1943.zip

Run a correcting scrub on the pool and make sure are errors are corrected, if not backup, format and restore.