From patchwork Wed Apr 12 10:41:33 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zorro Lang X-Patchwork-Id: 9677083 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 1E16A60384 for ; Wed, 12 Apr 2017 10:41:51 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 10C62285FE for ; Wed, 12 Apr 2017 10:41:51 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 0574E2860B; Wed, 12 Apr 2017 10:41:51 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 47E7628606 for ; Wed, 12 Apr 2017 10:41:50 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753019AbdDLKlr (ORCPT ); Wed, 12 Apr 2017 06:41:47 -0400 Received: from mx1.redhat.com ([209.132.183.28]:34018 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752336AbdDLKlj (ORCPT ); Wed, 12 Apr 2017 06:41:39 -0400 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 48F05C059740; Wed, 12 Apr 2017 10:41:39 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 48F05C059740 Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com Authentication-Results: ext-mx08.extmail.prod.ext.phx2.redhat.com; spf=pass smtp.mailfrom=zlang@redhat.com DKIM-Filter: OpenDKIM Filter v2.11.0 mx1.redhat.com 48F05C059740 Received: from dhcp-12-168.nay.redhat.com (dhcp-12-168.nay.redhat.com [10.66.12.168]) by smtp.corp.redhat.com (Postfix) with ESMTP id 467DE9ECD1; Wed, 12 Apr 2017 10:41:38 +0000 (UTC) From: Zorro Lang To: fstests@vger.kernel.org Cc: linux-xfs@vger.kernel.org Subject: [PATCH] fstests/xfs: xfs_repair should junk empty attribute leaf blocks Date: Wed, 12 Apr 2017 18:41:33 +0800 Message-Id: <1491993693-9667-1-git-send-email-zlang@redhat.com> X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Wed, 12 Apr 2017 10:41:39 +0000 (UTC) Sender: linux-xfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-xfs@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP There was a bug during log replay, the attr/attr3 leaf verifier reported corruption when encountering a leaf attribute with a count of 0 in the header, as below: Metadata corruption detected at xfs_attr3_leaf block 0x480988/0x1000 commit f714016 from xfsprogs has fixed this bug. This test case will emulate this corruption by xfs_db and use xfs_repair to fix it. Signed-off-by: Zorro Lang --- Hi, xfs_db already support 'write -d' from last year, even though the manpage doesn't record that. But the '-d' option can't be run for everything. I just saw Eric and Darrick send two patches to allow write -d to dqblks and inodes. But they haven't been merged. In this case, I hope to change a field of attr3 on V5 XFS, so I have to find a way to make sure if the current xfs_db can do that. I'm not the one who learn about xfs_db so much, so if someone feel something wrong, please feel free to tell me. I know I can limit this case test on V4 XFS, then there's no 'write -d' problem. But I hope to cover V5 XFS test if I can. Thanks, Zorro tests/xfs/288 | 103 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/xfs/288.out | 2 ++ tests/xfs/group | 1 + 3 files changed, 106 insertions(+) create mode 100755 tests/xfs/288 create mode 100644 tests/xfs/288.out diff --git a/tests/xfs/288 b/tests/xfs/288 new file mode 100755 index 0000000..3dc0a75 --- /dev/null +++ b/tests/xfs/288 @@ -0,0 +1,103 @@ +#! /bin/bash +# FS QA Test 288 +# +# When an attribute leaf block count is 0, xfs_repair should junk +# that leaf directly (as xfsprogs commit f714016). +# +#----------------------------------------------------------------------- +# Copyright (c) 2017 YOUR NAME HERE. All Rights Reserved. +# +# This program is free software; you can redistribute it and/or +# modify it under the terms of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +#----------------------------------------------------------------------- +# + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +echo "QA output created by $seq" + +here=`pwd` +tmp=/tmp/$$ +status=1 # failure is the default! +trap "_cleanup; exit \$status" 0 1 2 3 15 + +_cleanup() +{ + cd / + rm -f $tmp.* +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/attr + +# remove previous $seqres.full before test +rm -f $seqres.full + +# V4 XFS attr type is 'attr' +write_cmd="write" + +# Modify as appropriate. +_supported_fs xfs +_supported_os Linux +_require_scratch +_require_attrs +_require_xfs_db_command write +_require_xfs_db_command addr + +_scratch_mkfs_xfs | _filter_mkfs 2>$tmp.mkfs >/dev/null +. $tmp.mkfs + +# If test on XFS with CRC enabled, after change attr's hdr.count we need +# to recalculate crc. +if [ $_fs_has_crcs -eq 1 ]; then + # This's a trick to check if xfs_db support recalculating + # CRCs of attr3 + _scratch_xfs_db -x -c "daddr 0" \ + -c "type attr3" \ + -c "write -d" 2>/dev/null | \ + grep -q "Cannot recalculate CRCs on this type of object" + if [ $? -eq 0 ];then + _notrun "xfs_db can't recalculate CRCs of attr3" + fi + # V5 XFS attr type is 'attr3' + write_cmd="write -d" +fi + +_scratch_mount + +touch $SCRATCH_MNT/$seq.attrfile +inum=$(stat -c '%i' $SCRATCH_MNT/$seq.attrfile) + +# To get a attr block leaf, we need to extend attr format to extent +# format at least, and the max inode size is half of filesystem +# block size, so write half of block size attr to make sure attr +# out of local format. +fsblocksize=`xfs_info $SCRATCH_MNT|sed 's/=/ /g'|awk '/^data / { print $3 }'` +maxisize=$((fsblocksize/2)) +setfattr -n "user.testattr${seq}" -v `python -c "print ($maxisize * 'a')"` \ + $SCRATCH_MNT/$seq.attrfile +umount $SCRATCH_MNT +_scratch_xfs_db -x -c "inode $inum" \ + -c "addr a.bmx[0].startblock" \ + -c "$write_cmd hdr.count 0" >> $seqres.full + +# This repair should junk above leaf attribute to fix this XFS +_repair_scratch_fs >> $seqres.full + +echo "Silence is golden" + +# success, all done +status=0 +exit diff --git a/tests/xfs/288.out b/tests/xfs/288.out new file mode 100644 index 0000000..2958a5c --- /dev/null +++ b/tests/xfs/288.out @@ -0,0 +1,2 @@ +QA output created by 288 +Silence is golden diff --git a/tests/xfs/group b/tests/xfs/group index 75769f9..a0a03c0 100644 --- a/tests/xfs/group +++ b/tests/xfs/group @@ -285,6 +285,7 @@ 285 dangerous_fuzzers dangerous_scrub 286 dangerous_fuzzers dangerous_scrub dangerous_online_repair 287 auto dump quota quick +288 auto quick repair 290 auto rw prealloc quick ioctl zero 291 auto repair 292 auto mkfs quick