From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Hoffrichter Subject: Re: [linux-lvm] Best practice for lvm on raid-5 Message-ID: <20040202182846.GO5191@deepthought.hausboot.org> References: <20040202153506.GN5191@deepthought.hausboot.org> <20040202162025.GA27150@muon.de> Mime-Version: 1.0 Content-Disposition: inline In-Reply-To: <20040202162025.GA27150@muon.de> Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: Date: Mon Feb 2 13:30:02 2004 List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-lvm@sistina.com On Mon, Feb 02, 2004 at 05:20:25PM +0100, Nils Juergens wrote: > > Currently I have /dev/md0 as my only physical volume, but this will lead > > to problems if I am going to expand the raid 5 by adding another 80 GB > > harddrive to (as pvresize is not implemented again yet, as I understand). > > Unless I am mistaken, it is not even possible to expand /dev/md0 by adding > another drive in the first place. And because linux software raid has > no way of knowing which blocks are in use and which are not, i do not see > how this could ever be possible. Yes, it should be possible using raidreconf. I have not testet it yet, but at least the manpage states it. Converting a raid 0 to a raid 5 of same size worked without problem. > If you add another array (raid1 or raid5) as /dev/md1, you can vgextend your > VG. Yes, I'm aware of this, but for this I have to add another 3 disk to the system, and I want to avoid this. CU all, Jens -- GPG: 1024D/CF884D50 F2E8 F7FC F823 6464 4E9D EFAB 6EE9 8B9C CF88 4D50 ------------------------------------------------------------------------ Jens Hoffrichter / joho@hausboot.org / Joho@IRC / Fon: +43-650-5376989 A dream is a wish your heart makes. (Seen on a parade in Walt Disney World)