Alright so I noticed that for a while some folders will randomly start being set to read-only. Copying a large VM onto a share almost always causes the folder to lock up and abort the file copy.

Looking at my syslog I noticed that it keeps throwing BTRFS errors like every second. It starts here: 

 

Apr 10 19:31:53 Boniweb kernel: ata7.00: failed command: WRITE FPDMA QUEUED
Apr 10 19:31:53 Boniweb kernel: ata7.00: cmd 61/00:f8:58:fe:76/0a:00:04:00:00/40 tag 31 ncq dma 1310720 ou
Apr 10 19:31:53 Boniweb kernel:         res 40/00:ff:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)
Apr 10 19:31:53 Boniweb kernel: ata7.00: status: { DRDY }
Apr 10 19:31:53 Boniweb kernel: ata7: hard resetting link
Apr 10 19:31:53 Boniweb kernel: ata7: SATA link up 6.0 Gbps (SStatus 133 SControl 300)
Apr 10 19:31:59 Boniweb kernel: ata7.00: qc timeout (cmd 0xec)
Apr 10 19:31:59 Boniweb kernel: ata7.00: failed to IDENTIFY (I/O error, err_mask=0x4)
Apr 10 19:31:59 Boniweb kernel: ata7.00: revalidation failed (errno=-5)
Apr 10 19:31:59 Boniweb kernel: ata7: hard resetting link
Apr 10 19:32:04 Boniweb kernel: ata7: link is slow to respond, please be patient (ready=0)
Apr 10 19:32:09 Boniweb kernel: ata7: COMRESET failed (errno=-16)
Apr 10 19:32:09 Boniweb kernel: ata7: hard resetting link
Apr 10 19:32:14 Boniweb kernel: ata7: link is slow to respond, please be patient (ready=0)
Apr 10 19:32:19 Boniweb kernel: ata7: COMRESET failed (errno=-16)
Apr 10 19:32:19 Boniweb kernel: ata7: hard resetting link
Apr 10 19:32:24 Boniweb kernel: ata7: link is slow to respond, please be patient (ready=0)
Apr 10 19:32:54 Boniweb kernel: ata7: COMRESET failed (errno=-16)
Apr 10 19:32:54 Boniweb kernel: ata7: limiting SATA link speed to 3.0 Gbps
Apr 10 19:32:54 Boniweb kernel: ata7: hard resetting link
Apr 10 19:32:59 Boniweb kernel: ata7: COMRESET failed (errno=-16)
Apr 10 19:32:59 Boniweb kernel: ata7: reset failed, giving up
Apr 10 19:32:59 Boniweb kernel: ata7.00: disabled
Apr 10 19:32:59 Boniweb kernel: ata7: EH complete
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#7 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#7 CDB: opcode=0x2a 2a 00 04 76 fe 58 00 0a 00 00
Apr 10 19:32:59 Boniweb kernel: print_req_error: I/O error, dev sdh, sector 74907224
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#8 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#8 CDB: opcode=0x2a 2a 00 04 76 f8 58 00 06 00 00
Apr 10 19:32:59 Boniweb kernel: print_req_error: I/O error, dev sdh, sector 74905688
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#9 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#9 CDB: opcode=0x2a 2a 00 04 76 ee 58 00 0a 00 00
Apr 10 19:32:59 Boniweb kernel: print_req_error: I/O error, dev sdh, sector 74903128
Apr 10 19:32:59 Boniweb kernel: BTRFS error (device sdh1): bdev /dev/sdh1 errs: wr 1, rd 0, flush 0, corrupt 0, gen 0
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#10 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#10 CDB: opcode=0x2a 2a 00 04 76 e8 58 00 06 00 00
Apr 10 19:32:59 Boniweb kernel: print_req_error: I/O error, dev sdh, sector 74901592
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#11 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00
Apr 10 19:32:59 Boniweb kernel: sd 7:0:0:0: [sdh] tag#11 CDB: opcode=0x2a 2a 00 04 76 de 58 00 0a 00 00
Apr 10 19:32:59 Boniweb kernel: print_req_error: I/O error, dev sdh, sector 74899032

A reboot always fixes the disk for a while, even though there is no sign in the unraid gui that anything was ever wrong with it... 

 

Now I'm about to reformat that cache disk (it's a 2 months old SSD) as soon as the parity check that's currently running is done. However I'm not sure if that will fix it. Maybe someone can at least tell if it's a hardware or software problem?

boniweb-diagnostics-20190410-1946.zip

Edited April 10, 20197 yr by Boro

This looks more like a connection problem, btrfs errors are just the result of the hardware errors, replace both cables to see if helps.