[zfs-discuss] Issue with data

Richard Laager rlaager at wiktel.com
Thu Sep 26 06:08:37 EDT 2013


On Thu, 2013-09-26 at 02:19 -0700, Joachim Grunwald wrote:
> I did set up the pool  "zpool create -o ashift=12 tank raidz2 /dev/sda 
> /dev/sdb /dev/sdc /dev/sdd /dev/sde /dev/sdf" and a single drive 2T pool 
> "zpool create -o ashift=12 single /dev/sdg"

Short answer: Just stop worrying about this. You set ashift=12 for
future-proofing, you're using filesystems not zvols and have not reduced
recordsize, and you're not bothered by the space usage.

If you want to test this to see the difference, put a zvol of the same
size on each with blocksize=4k. Then fill the zvol (zeroes should be
fine if compression is disabled). The zvol should take up more space on
the raidz pool because of the round-up.

People see this happen and are surprised, because they expect the space
usage to be the same as it would be with RAID-5 or RAID-6.

-- 
Richard
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://list.zfsonlinux.org/pipermail/zfs-discuss/attachments/20130926/d3507af4/attachment.sig>


More information about the zfs-discuss mailing list