WRITE(06). CDB: error in /var/log/messages

[Dan Nelson]

In the last episode (Jan 24), Ryan Petty said:
> I am not a new user to FreeBSD, but I am seeing some errors in
> /var/log/messages that I am not familiar with and have had difficulty
> finding a definition and ultimately a resolution.
> 
> I have researched the FreeBSD documentation, but my searches lead to
> developer information which does not answer my question (at least in
> a way I can understand it).
> 
> A sample of the errors is found below:
> 
> Jan 24 11:03:44 homer /kernel: (da0:ahc0:0:0:0): WRITE(06). CDB: a b a2 73 8 0
> Jan 24 11:03:44 homer /kernel: (da0:ahc0:0:0:0): Deferred Error: MEDIUM ERROR info:17f asc:14,1
> Jan 24 11:03:44 homer /kernel: (da0:ahc0:0:0:0): Record not found field replaceable unit: 1
> Jan 24 11:04:12 homer /kernel: (da0:ahc0:0:0:0): WRITE(06). CDB: a 11 65 df 20 0
> Jan 24 11:04:12 homer /kernel: (da0:ahc0:0:0:0): Deferred Error: MEDIUM ERROR info:17f asc:14,1
> Jan 24 11:04:12 homer /kernel: (da0:ahc0:0:0:0): Record not found field replaceable unit: 1
> Jan 24 11:05:29 homer /kernel: (da0:ahc0:0:0:0): WRITE(06). CDB: a 0 2 1f 20 0
> Jan 24 11:05:29 homer /kernel: (da0:ahc0:0:0:0): Deferred Error: MEDIUM ERROR info:17f asc:14,1
> Jan 24 11:05:29 homer /kernel: (da0:ahc0:0:0:0): Record not found field replaceable unit: 1
> 
> What do these errors mean and how do I fix them--or is this a sign of
> a dead or dying drive or SCSI controller?

Looks like a drive about to fail.  Depending on the vendor, you may
have to run a diagnostic program that displays the SMART info before
they'll replace it.  You can also dump the same data with the
sysutils/smartmontools port.

-- 
	Dan Nelson
	dnelson at allantgroup.com