From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mx3.redhat.com (mx3.redhat.com [172.16.48.32]) by int-mx1.corp.redhat.com (8.11.6/8.11.6) with ESMTP id iAK8cKr13940 for ; Sat, 20 Nov 2004 03:38:20 -0500 Received: from polop.usc.edu (polop.usc.edu [128.125.10.9]) by mx3.redhat.com (8.12.11/8.12.11) with ESMTP id iAK8c9GZ021279 for ; Sat, 20 Nov 2004 03:38:09 -0500 Received: from polop.usc.edu (localhost.localdomain [127.0.0.1]) by polop.usc.edu (8.12.11/8.12.11) with ESMTP id iAK8c8sA003537 for ; Sat, 20 Nov 2004 00:38:08 -0800 Received: (from garrick@localhost) by polop.usc.edu (8.12.11/8.12.11/Submit) id iAK8c8X0003535 for linux-lvm@redhat.com; Sat, 20 Nov 2004 00:38:08 -0800 Date: Sat, 20 Nov 2004 00:38:08 -0800 From: Garrick Staples Subject: Re: Re: [linux-lvm] Software RAID 5 Resizing and LVM Message-ID: <20041120083808.GS3139@polop.usc.edu> References: <1100732495.2496.6.camel@Slacktop.local> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="4uMdeSyPFwfKNOAv" Content-Disposition: inline In-Reply-To: <1100732495.2496.6.camel@Slacktop.local> Reply-To: LVM general discussion and development List-Id: LVM general discussion and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , List-Id: To: LVM general discussion and development --4uMdeSyPFwfKNOAv Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable > Date: Wed, 17 Nov 2004 10:53:00 -0600 > From: Jetkins@austinlogistics.com > Subject: Re: [linux-lvm] Software RAID 5 Resizing and LVM > To: adam@slackguy.net >=20 >=20 > I asked pretty much that exact question on Monday! :) The only difference > was that mine is a hardware RAID. Someone replied with a possible > workaround if you're using LVM2, but if you're using LVM1 like me, and you > pvcreate'd your PV across the entire disk like me, then you're screwed li= ke > me. :( >=20 > If you partitioned the logical disk and pvcreate'd a partition, then you > should be able to create a new partition on the newly-available space, th= en > pvcreate a second PV and add it into the VG. The wonders of 20/20 > hindsight. I'm using the entire, unpartitioned, block device on my boxes on advice fro= m my DataDirect Networks vendor. Apparently putting a partition table at the fr= ont of the block device offsets each write request 512 bytes, which is disaster= ous for performance. By using the whole device, the writes are correctly byte aligned to the arrays. --=20 Garrick Staples, Linux/HPCC Administrator University of Southern California --4uMdeSyPFwfKNOAv Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.3 (GNU/Linux) iD8DBQFBnwJw0SBUxJbm9HMRAjnaAJ4xxkIjk56+lX+yib2x/lH7o4WdGgCgkkxQ fqis/gJ3YDUYg7V+xcgw/Kk= =WuNq -----END PGP SIGNATURE----- --4uMdeSyPFwfKNOAv--