From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Wed, 9 May 2001 21:26:29 +0000 From: "Heinz J. Mauelshagen" Subject: Re: [linux-lvm] SCSI/LVM problems after power outage Message-ID: <20010509212629.A8512@sistina.com> References: <20010509093510.B4918@sistina.com> <20010509201416.A20867@mercury.mercury.multiweb.nl> Mime-Version: 1.0 In-Reply-To: <20010509201416.A20867@mercury.mercury.multiweb.nl>; from dave@xs4all.nl on Wed, May 09, 2001 at 08:14:16PM +0200 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: List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-lvm@sistina.com On Wed, May 09, 2001 at 08:14:16PM +0200, Dave Wapstra wrote: > On 09 May, 2001 at 09:35:10 +0000, Heinz J. Mauelshagen wrote: > > > > On Tue, May 08, 2001 at 02:05:16PM -0700, Day, Evan wrote: > > > It looks like device 2 is having issues: > > > > > > SCSI disk error : host 2 channel 0 id 2 lun 0 return code = 18000002 > > > > > Ok. Sounds reasonable ;) It seems /dev/sdc is giving troubles. > > > In case the PEs on the flaky device are still readable: > > extending your VG and using pvmove is the way to go. > > > > Steps needed (crossing your fingers that the flaky drive can still stand this): > > > > - install or use an additional drive of at least the size of the flaky one > > - pvcreate that drive > > - vgextend the VG by it > > - pvmove /dev/FlakyDrive > > - wait until pvmove is done > > - vgreduce VG /dev/FlakyDrive > > I'm trying to backup the data that is still available (backup was a > thing that was still in progress... :) ) > > pvmove is not really happy with the current situation. I don't understand you here. Did you try the above steps and something failed? > > Can I just do "vgreduce ftp /dev/sdc" after I finished rescueing most of the data? > Sure. In case you've got a backup, you could remove or reduce LVs so that *no* PEs are allocated on /dev/sdc. You can check with "pvdisplay -v /dev/sdc" which LV(s) have PEs allocated on that PV and run lvremove/lvreduce in order to free that PV. Once pvdisplay tells you that no PEs are allocated (Allocated PE 0) you can run "vgreduce VGName /dev/sdc". > > You should have your LVs offline in case you have an older version than > > LVM 0.9.1 Beta 7, because they had a pvmove related bug which could cause > > oopses when you moved PEs in use. > > > > BTW: you need to configure MD (RAID 1 or 5) or use hardware raid subsystems > > to avoid suffering from such flaky disk problems. > > I haven't looked at RAID yet. The striping option of LVM is not really > an option, since I cannot add any disks to the volume. That's true today. There's a tiny item on the TODO list to enhance this ;-) > > -Dave > > -- > Dave Wapstra > dave@xs4all.nl > _______________________________________________ > linux-lvm mailing list > linux-lvm@sistina.com > http://lists.sistina.com/mailman/listinfo/linux-lvm Regards, Heinz -- The LVM Guy -- *** Software bugs are stupid. Nevertheless it needs not so stupid people to solve them *** =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Heinz Mauelshagen Sistina Software Inc. Senior Consultant/Developer Am Sonnenhang 11 56242 Marienrachdorf Germany Mauelshagen@Sistina.com +49 2626 141200 FAX 924446 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-