Making sense of error codes

Alexander Motin mav at
Thu May 5 20:23:07 UTC 2011

Mike Neimoyer wrote:
> First post to the group, and I'm a bit of a newbie in general, so please
> be gentle. :)  I'll provide as much information as I know how to, and if
> there's more you need, please let me know and I'll do my best!  Thanks.
> Recently, I've had a tape library's autoloader stop being recognized. It
> was working just fine, and then the backup software (amanda) began
> reporting errors that it couldn't communicate.  No changes were made to
> the system and it had run a backup earlier that morning successfully.
> Multiple attempts to re-establish communications failed, checked all
> cables, verified that the system saw the device by "camcontrol devlist"
> (it did, but the unit was listed simply as "pass3" instead of the usual
> "ch0,pass3").  Was eventually told to shutdown and powercycle both
> units, but without success.
> Basics:
> Dell Powervault 124t via SCSI
> Checking the /var/log/messages file, after the  I see:
> May 2 16:44:24 jazz kernel: sa0 at ahc1 bus 0 target 12 lun 0
> May 2 16:44:24 jazz kernel: sa0: <CERTANCE ULTRIUM 2 1914> Removable
> Sequential Access SCSI-3 device
> May 2 16:44:24 jazz kernel: sa0: 160.000MB/s transfers (80.000MHz,
> offset 127, 16bit)
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): MODE SENSE(06). CDB: 1a
> 20 1d 0 20 0
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): HARDWARE FAILURE asc:40,84
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): (null)
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): fatal error, failed to
> attach to device
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): lost device
> May 2 16:44:24 jazz kernel: (ch0:ahc1:0:12:1): removing device entry
> So, there is a hardware failure somewhere, but I am not sure how to
> determine if it's the internal card or the external tape library unit
> itself (in a perfect world, I'd have another tape library unit I could
> just swap out, but alas.... :)  ).
> I've tried looking up the ASC (40, 84), but can't seem to find a
> agreed-upon meaning for them.  I could, quite probably in fact, simply
> not be looking in the right place.
> Can any kind soul point me in the right direction for figuring this
> issue out?  Thanks!

According to SPC-3 specification "asc:40,84" means: "DIAGNOSTIC FAILURE
ON COMPONENT NN (80H-FFH)", where NN is 84H.

Alexander Motin

More information about the freebsd-scsi mailing list