Date: Mon, 18 Jan 2010 22:05:00 +0000 From: Mike Clarke <jmc-freebsd2@milibyte.co.uk> To: freebsd-questions@freebsd.org Cc: Pieter de Goeje <pieter@degoeje.nl> Subject: Re: Newbie gmirror questions Message-ID: <201001182205.00543.jmc-freebsd2@milibyte.co.uk> In-Reply-To: <4B538459.7090601@infracaninophile.co.uk> References: <201001152334.52978.jmc-freebsd2@milibyte.co.uk> <201001172122.15128.jmc-freebsd2@milibyte.co.uk> <4B538459.7090601@infracaninophile.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sunday 17 January 2010, Matthew Seaman wrote: > However, one of the really amazingly brilliant things about geom is > that just about any disk / storage related thing can be a geom > provider, and geom constructs will nest very happily. =A0Here's a howto > for setting up gmirror across a pair of slices: > > http://people.freebsd.org/~rse/mirror/ That's a very interesting article. Since I'll be able to configure the=20 mirror on the new drives before installing any software my approach can=20 be a bit simpler. In the example he's using a single partition for the whole disk but=20 reduces the size if the partition by one block so that the mirror's=20 meta data doesn't get misinterpreted as whole disk meta data. Since I=20 anticipate using only the first 2 partitions for a couple of mirrors=20 and using the rest of the disk as plain partitions then I don't think I=20 need to do this but might it still be a good idea to reduce the last=20 partition by one block anyway in case my usage changes in the future? =2D-=20 Mike Clarke
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201001182205.00543.jmc-freebsd2>