Date: Wed, 13 Jun 2012 01:21:08 +0200 From: Damien Fleuriot <ml@my.gd> To: Nenhum_de_Nos <matheus@eternamente.info> Cc: "freebsd-stable@freebsd.org" <freebsd-stable@freebsd.org> Subject: Re: lost ZFS pool Message-ID: <D0275186-A74B-4473-8F7A-13585C2ABA40@my.gd> In-Reply-To: <e2619f0d9d8c11085ebde96e6841851a.squirrel@eternamente.info> References: <e2619f0d9d8c11085ebde96e6841851a.squirrel@eternamente.info>
next in thread | previous in thread | raw e-mail | index | archive | help
On 12 Jun 2012, at 20:17, "Nenhum_de_Nos" <matheus@eternamente.info> wrote: > hail, >=20 > I write just to make sure its dead. I've lost the first disk on a ZFS pool= (jbod). Now I can't > mount it with only the second disk. The first disk clicks to death :( >=20 > [root@optimus ~]# zpool status > pool: pool > state: UNAVAIL > status: One or more devices could not be opened. There are insufficient > replicas for the pool to continue functioning. > action: Attach the missing device and online it using 'zpool online'. > see: http://www.sun.com/msg/ZFS-8000-3C > scrub: none requested > config: >=20 > NAME STATE READ WRITE CKSUM > pool UNAVAIL 0 0 0 insufficient replicas > label/zfs1 UNAVAIL 0 0 0 cannot open > label/zfs2 ONLINE 0 0 >=20 > I have a spare disk (blank), but even though I can't make it online again .= .. >=20 > is there any hope I can read the files in that disk ? >=20 > thanks, >=20 > matheus >=20 Johan has the right of it, you're splitting data between your 2 devices a la= raid0, this means both disks are required for the pool to function. Your data is gone and you're not recovering any of it. You will want to follow Johan's recommendation of using mirror (raid1), rai= dz (raid5) or even raidz2 (raid6) to ensure data integrity and availability.= Note that using raid does *not* exempt you from running backups. As has been pointed out already (Jeremy Chadwick was that you ?), with large= disks ( >1TB ) you encounter the risk of another disk in your raid array fa= iling while you're rebuilding on the replacement drive, because of the high I= O. Also, while I am a proponent of using the full disk for ZFS (as opposed to a= slice), it is recommended to use slices and shave off roughly 100mb from th= e disk space available when creating your pool. This way, if an old disk fails and you replace it with one the same size but= from another manufacturer (the size may vary a bit, imagine if it's 2mb sma= ller, you can't use it in your pool !) you won't face the problem.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?D0275186-A74B-4473-8F7A-13585C2ABA40>