Return-Path: Received: from mail1.g1.pair.com ([66.39.3.162]:34691 "EHLO mail1.g1.pair.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727620AbfBKQZM (ORCPT ); Mon, 11 Feb 2019 11:25:12 -0500 Received: from mail1.g1.pair.com (localhost [127.0.0.1]) by mail1.g1.pair.com (Postfix) with ESMTP id 46B485476C4 for ; Mon, 11 Feb 2019 11:25:11 -0500 (EST) Received: from harpe.intellique.com (labo.djinux.com [82.225.196.72]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail1.g1.pair.com (Postfix) with ESMTPSA id E267760B0C3 for ; Mon, 11 Feb 2019 11:25:10 -0500 (EST) Date: Mon, 11 Feb 2019 17:23:59 +0100 From: Emmanuel Florac Subject: RAID hiccup, LVM information wiped, XFS FS gone, how to proceed? Message-ID: <20190211172359.71aff9d6@harpe.intellique.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; boundary="Sig_/HqlO8AI6QMX=wjTnbhd_ZeZ"; protocol="application/pgp-signature" Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Linux-XFS list --Sig_/HqlO8AI6QMX=wjTnbhd_ZeZ Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Hi everyone, I have a lost filesystem at hand: the RAID controller had a hiccup (or someone tripped on the cord, or did some other mistake, I have no way to know because the system is in some other country anyway). The configuration was quite simple: RAID array -> pv -> vg -> lv -> XFS. I've hexedit'ed the /dev/sdb and it looks like the first few kB have been wiped out: no LVM signature, no VG UUID... However after the first few blocks, it looks very similar to another disk>vg>lv configuration which I hexedit side by side (there are series of 00 with regularly placed "FIB" values interspersed with other values, then after a while what is obviously file content) so my guess is that the data is probably still there (better be, there are 80 TB missing). xfs_repair can't seem to be able to do anything on the raw disk, alas. Any ideas? I don't even know why I can't seem to find the XFSB signature on disk: when hexediting (for instance) /dev/vg0/somelv, I see it starting with XFSB; but editing /dev/sdc (which is hosting the above mentioned lv), it's not there. (I know, I could simply restore the vgcfgbackup if it existed... please let's suppose this command doesn't exist). --=20 ------------------------------------------------------------------------ Emmanuel Florac | Direction technique | Intellique | | +33 1 78 94 84 02 ------------------------------------------------------------------------ --Sig_/HqlO8AI6QMX=wjTnbhd_ZeZ Content-Type: application/pgp-signature Content-Description: Signature digitale OpenPGP -----BEGIN PGP SIGNATURE----- iEYEARECAAYFAlxhoZ8ACgkQX3jQXNUicVYPLACfYCzIlGOF33E0gvyzOkMNSY/m 4MEAn2HYTVWO1aNpQpWcOhrlMFEt2gpz =Pgm8 -----END PGP SIGNATURE----- --Sig_/HqlO8AI6QMX=wjTnbhd_ZeZ--