Date: Tue, 2 Jan 2018 15:34:05 -0500 From: "Karl Vogel" <vogelke@pobox.com> To: freebsd-questions@freebsd.org Subject: Re: zpool fragmentation Message-ID: <20180102203405.GA17447@bsd118.wpafb.af.mil> In-Reply-To: <CA%2BtpaK0r=DAOi5pA%2BYHC4u7C12EjRc6sUE1tdQJU-EW5YWFVqw@mail.gmail.com> References: <CAGBmCT6tjinhjQrzs08zVUoK6OzUB1HzovN=cOmAWOfNzrRPsA@mail.gmail.com> <CA%2BtpaK0r=DAOi5pA%2BYHC4u7C12EjRc6sUE1tdQJU-EW5YWFVqw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Jan 02, 2018 at 01:52:54PM -0600, Adam Vande More wrote:
> On Tue, Jan 2, 2018 at 12:40 PM, pete wright <nomadlogic@gmail.com> wrote:
>
>> are there any coping mechanisms people use to reduce fragmentation?
>
> Keep lots of free space.
If you create an unmounted empty filesystem with a reservation to
prevent the zpool from filling up, it gives you behavior similar to
ufs's reserved blocks. I reserve 5-10% to make sure the system doesn't
become completely unusable:
root# zfs create -o reservation=50G -o mountpoint=none rpool/reservation
--
Karl Vogel I don't speak for the USAF or my company
Sheep learn to recognise celebrity faces from different angles
--"New Scientist", 8 Nov 2017
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20180102203405.GA17447>