Harddrive beginning to expire?

[Dan Nelson]

In the last episode (Aug 17), doug at polands.org said:
> I'm seeing this entry in my /var/log/messages approx. every three hours:
> 
> Aug 17 06:09:07 judea /kernel: (da5:ahc0:0:5:0): WRITE(06). CDB: a 0 1 f9 a 0
> Aug 17 06:09:07 judea /kernel: (da5:ahc0:0:5:0): RECOVERED ERROR asc:5d,0
> Aug 17 06:09:07 judea /kernel: (da5:ahc0:0:5:0): Failure prediction threshold exceeded field replaceable unit: 1
> Aug 17 09:28:52 judea /kernel: (da5:ahc0:0:5:0): WRITE(06). CDB: a 0 1 f9 a 0
> Aug 17 09:28:52 judea /kernel: (da5:ahc0:0:5:0): RECOVERED ERROR asc:5d,0
> Aug 17 09:28:52 judea /kernel: (da5:ahc0:0:5:0): Failure prediction threshold exceeded field replaceable unit: 1
> 
> This 4.9-STABLE box is running 7 SCSI drives in a vinum stripped array.
> Question, is this the beginning of the end for drive da5?

It probably still has some life left in it.  It looks like you don't
have automatic write reallocation enabled, since the block number is
the same on both requests.  You can enable it by running "camcontrol
mode da5 -e -m 1 -P 3", and setting AWRE to 1.  That will let the drive
remap that disk block to a spare one.  You can monitor how many blocks
have been reallocated by viewing the grown defect list: "camcontrol
defects da5 -f phys -G".  If you use -P instead of -G, you can see the
primary defect list, which is a list of all the bad blocks found when
the disk was shipped.

-- 
	Dan Nelson
	dnelson at allantgroup.com