Solved WD 8TB CAM errors

[rubenerd]

Hey everyone,

Has anyone had problems with the new Western Digital Red 8TB NAS (5400, not 7200) drives under generic 11.1-RELEASE? I'm running two with GELI in a ZFS mirror on an HP Gen8 Microserver, but after three months have started seeing strings of CAM errors.

(ada0:ahcich3:0:0:0): WRITE_FPDMA_QUEUED. ACB: 61 08 f8 f0 00 40 68 01 01 00 00 00 00
(ada0:ahcich3:0:0:0): CAM status: Command timeout
(ada0:ahcich3:0:0:0): Retrying command
ahcich3: Timeout on slot 16 port 0
(ada0:ahcich3:0:0:0): FLUSHCACHE48. ACB: ea 00 00 00 00 40 00 00 00 00 00 00
(ada0:ahcich3:0:0:0): CAM status: Command timeout
(ada0:ahcich3:0:0:0): Retrying command

IIRC, I saw these errors in the 8.0-RELEASE days with a Samsung spinning, but not loading ahci(4) mitigated it. That was a long time ago though, so I'm almost certain this is just a dying drive.

Curiously, smartd output is consistent with the other working drive. I'm also running two of these drives with btrfs on Ubuntu in a lab at work (*nervous twitch*) and haven't seen issues.

Before sending it back to WD, anyone seen something similar with these specific drives?

 

[ralphbsz]

Timeout could easily be caused by communication problems (SATA cable, port) rather than by the drive itself. Try reseating cables, switching the two drives, and similar debugging techniques.

A drive internally failing without any trace of a problem visible in SMART data is possible, but very unlikely.

 

[rubenerd]

Hey Ralph, cheers. Your suspicions were correct. I moved the drives to another microserver, booted and they worked. Either a problem with the connectors on the suspect machine, or maybe the RAID card. Either way, I know where to troubleshoot now.