fdisk(8) vs gpart(8), and gnop

Steven Hartland killing at multiplay.co.uk
Mon Jun 2 09:09:04 UTC 2014


----- Original Message ----- 
From: "Matthew Ahrens" <mahrens at delphix.com>
> On Sun, Jun 1, 2014 at 8:55 AM, Nathan Whitehorn <nwhitehorn at freebsd.org>
> wrote:
> 
>> On 06/01/14 08:52, Steven Hartland wrote:
>>
>>> ----- Original Message ----- From: "Mark Felder" <feld at freebsd.org>
>>>
>>>  On May 31, 2014, at 20:57, Freddie Cash <fjwcash at gmail.com> wrote:
>>>>
>>>>  There's a sysctl where you can set the minimum ashift for zfs. Then you
>>>>> never need to use gnop.
>>>>>
>>>>> I believe it's part of 10.0?
>>>>>
>>>>
>>>> I've not seen this yet. What we need is to port the ability to set
>>>> ashift at pool creation time:
>>>>
>>>> $ zpool create -o ashift=12 tank mirror disk1 disk2 mirror disk3 disk4
>>>>
>>>> I believe the Linux zfs port has this functionality now, but we still do
>>>> not.
>>>>
>>>
> I am strongly against implementing "-o ashift=12"[*].  If we need to
> explicitly tell ZFS what sector size to use, we should make that intention
> clear with an appropriately named property, for example "-o
> device_sector_size=4K".  Even better, this property should be per-disk.
> 
> --matt
> 
> [*]  Once we implement an appropriately-named property, I would be OK with
> also allowing the "-o ashift=12" as a Linux compatibility feature.

Being able to override the detected sector size for a disk would get my vote
as that would result in everything else just working and would be able to
have different ashift per top level vdev.

That said I'm not sure if that should be a ZFS option or just a OS option?

IIRC solaris / illumos has that ability already?

    Regards
    Steve


More information about the freebsd-fs mailing list