Date: Mon, 31 Dec 2012 02:52:13 +0200 From: Kimmo Paasiala <kpaasial@gmail.com> To: Nikolai Lifanov <lifanov@mail.lifanov.com> Cc: freebsd-fs@freebsd.org Subject: Re: Zpool made of ZVOLs? Message-ID: <CA%2B7WWScB2G0WbSfu7UJPwaCEXiPP%2BZdDO9PPgsVpB8QuzYYKfA@mail.gmail.com> In-Reply-To: <973d506368c26dda4f618556b071e454@mail.lifanov.com> References: <973d506368c26dda4f618556b071e454@mail.lifanov.com>
index | next in thread | previous in thread | raw e-mail
On Mon, Dec 31, 2012 at 1:25 AM, Nikolai Lifanov <lifanov@mail.lifanov.com> wrote: >> Is it possible and even recommended to create a zpool out of block >> devices that are actually ZVOLs? >> >> I'd like to have a separate zfs pool for poudriere because building >> ports with it disrupts the snapshotting on my main pool if the >> poudirere jails are created on it. At the same time I don't want to >> resort to using external disks to create a pool that is only used for >> a jail or two. >> >> Regards, >> >> -Kimmo >> > > I'm sure that this is not what you want. > I think you are trying to snapshot your root pool and volatility + size of > poudriere data makes this impractical. The first pool would still be the > provider of zvols for the second pool. > If you can't have pools backed by separated devices (i.e. rpool on separate > disks from tank), then you can always arrange the data that poudriere is > interfering with to be on a different dataset. > So, from tank/poudriere/... arrangement you can move to tank/ROOT/... and > tank/DATA/poudriere. > > - Nikolai Lifanov I solved this by a vnode backed md(4) disk. I know that I could also use little more fine grained snapshoting and backups. Mow I'm just creating recursive back ups for the whole pool and backing up everything using one zfs send, zfs receive operation.help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CA%2B7WWScB2G0WbSfu7UJPwaCEXiPP%2BZdDO9PPgsVpB8QuzYYKfA>