[Discuss] On Btrfs raid and odd-count disks

[Dan Ritter]

On Fri, Apr 12, 2013 at 10:51:21AM -0400, Derek Atkins wrote:
> Richard Pieri <richard.pieri at gmail.com> writes:
> > Both ZFS and Btrfs provide facilities for automatically "erasing" write
> > holes. The process is called "scrubbing". The scrubbing process walks
> > through the entire file system tree, recalculates all file and metadata
> > checksums, and compares them to the stored checksums. Errors are
> > repaired using replica data. Oracle's documentation recommends a weekly
> > scrubbing schedule for consumer-grade disks and a monthly scrubbing
> > schedule for server-grade disks.
> 
> Fair enough...  I don't know if standard (e.g. DM-level) RAID5 or RAID6
> provide for said "scrubbing"?  Or detecting/handling disk read or (or
> worse, disk write) failures.

Disk read and write failures are events which ought to be
reported by the disk interface. I've certainly seen enough of
them. There are occasions where they won't be, though... the one
time that happened to me, it was an earlyish 3Ware RAID card
that turned out to have problems with non-passive PCI risers.

mdadm has a sort of scrub facility available, in which it reads
all the bits -- see /sys/block/$array/md/sync_action 

Most HW RAID controllers have something similar available.

-dsr-