All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lukas Czerner <lczerner@redhat.com>
To: fstests@vger.kernel.org
Subject: [PATCH] ext4/309: Test read-only external journal device
Date: Fri, 17 Jul 2020 12:55:44 +0200	[thread overview]
Message-ID: <20200717105544.3201-1-lczerner@redhat.com> (raw)

We should never be able to mount ext4 file system read-write with
read-only external journal device. Test it.

The test was based on generic/050.

Signed-off-by: Lukas Czerner <lczerner@redhat.com>
---
 tests/ext4/309     | 135 +++++++++++++++++++++++++++++++++++++++++++++
 tests/ext4/309.out |  32 +++++++++++
 tests/ext4/group   |   1 +
 3 files changed, 168 insertions(+)
 create mode 100755 tests/ext4/309
 create mode 100644 tests/ext4/309.out

diff --git a/tests/ext4/309 b/tests/ext4/309
new file mode 100755
index 00000000..356044c4
--- /dev/null
+++ b/tests/ext4/309
@@ -0,0 +1,135 @@
+#! /bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2009 Christoph Hellwig.
+# Copyright (c) 2020 Lukas Czerner.
+#
+# FS QA Test No. 309
+#
+# Copied from tests generic/050 and adjusted to support testing
+# read-only external journal device on ext4.
+#
+# Check out various mount/remount/unmount scenarious on a read-only
+# logdev blockdev.
+#
+seqfull=$0
+seq=`basename $0`
+seqres=$RESULT_DIR/$seq
+echo "QA output created by $seq"
+
+here=`pwd`
+tmp=/tmp/$$
+status=1	# failure is the default!
+
+_cleanup()
+{
+	cd /
+	blockdev --setrw $SCRATCH_LOGDEV
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_supported_fs generic
+_supported_os Linux
+
+_require_scratch_nocheck
+_require_scratch_shutdown
+_require_logdev
+_require_local_device $SCRATCH_DEV
+_require_local_device $SCRATCH_LOGDEV
+_require_norecovery
+
+_scratch_mkfs >/dev/null 2>&1
+_require_metadata_journaling $SCRATCH_DEV
+
+#
+# Mark the log device read-only
+#
+echo "setting log device read-only"
+blockdev --setro $SCRATCH_LOGDEV
+
+#
+# Mount it, and make sure we can't write to it, and we can unmount it again
+#
+echo "mounting with read-only log device:"
+_try_scratch_mount 2>&1 | _filter_ro_mount | _filter_scratch
+if [ "${PIPESTATUS[0]}" -eq 0 ]; then
+	echo "touching file on read-only filesystem (should fail)"
+	touch $SCRATCH_MNT/foo 2>&1 | _filter_scratch
+fi
+
+echo "unmounting read-only filesystem"
+_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot
+
+echo "setting log device read-write"
+blockdev --setrw $SCRATCH_LOGDEV
+
+echo "mounting with read-write log device:"
+_try_scratch_mount 2>&1 | _filter_scratch
+
+echo "touch files"
+touch $SCRATCH_MNT/{0,1,2,3,4,5,6,7,8,9}{0,1,2,3,4,5,6,7,8,9}
+
+echo "going down:"
+_scratch_shutdown -f
+
+echo "unmounting shutdown filesystem:"
+_scratch_unmount 2>&1 | _filter_scratch
+
+echo "setting log device read-only"
+blockdev --setro $SCRATCH_LOGDEV
+
+#
+# Mounting a filesystem that requires log-recovery fails even with
+# -o norecovery unless the fs device is read only or it's mounted
+# read only
+#
+echo "mounting filesystem that needs recovery with a read-only log device:"
+_try_scratch_mount 2>&1 | _filter_ro_mount | _filter_scratch
+
+echo "unmounting read-only filesystem"
+_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot
+
+
+echo "mounting filesystem with -o norecovery with a read-only log device:"
+_try_scratch_mount -o norecovery 2>&1 | _filter_ro_mount | _filter_scratch
+echo "unmounting read-only filesystem"
+_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot
+
+#
+# This is the way out if the log device really is read-only.
+# Doesn't mean it's a good idea in practice, more a last resort
+# data recovery hack. Either the underlying fs device needs
+# to be read only as well, or we mount the file system read only
+#
+echo "setting fs device read-only"
+blockdev --setro $SCRATCH_DEV
+echo "mounting filesystem with -o norecovery with a read-only fs and log device:"
+_try_scratch_mount -o norecovery 2>&1 | _filter_ro_mount | _filter_scratch
+echo "unmounting read-only filesystem"
+_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot
+echo "setting fs device read-write"
+blockdev --setrw $SCRATCH_DEV
+
+echo "mounting filesystem with -o norecovery,ro with a read-only log device:"
+_try_scratch_mount -o norecovery,ro 2>&1 | _filter_ro_mount | _filter_scratch
+echo "unmounting read-only filesystem"
+_scratch_unmount 2>&1 | _filter_scratch | _filter_ending_dot
+
+echo "setting log device read-write"
+blockdev --setrw $SCRATCH_LOGDEV
+
+#
+# But log recovery is performed when mount with -o ro as long as
+# the underlying device is not write protected.
+#
+echo "mounting filesystem that needs recovery with -o ro:"
+_try_scratch_mount -o ro 2>&1 | _filter_scratch
+
+# success, all done
+echo "*** done"
+rm -f $seqres.full
+status=0
diff --git a/tests/ext4/309.out b/tests/ext4/309.out
new file mode 100644
index 00000000..4b7e136a
--- /dev/null
+++ b/tests/ext4/309.out
@@ -0,0 +1,32 @@
+QA output created by 309
+setting log device read-only
+mounting with read-only log device:
+mount: device write-protected, mounting read-only
+touching file on read-only filesystem (should fail)
+touch: cannot touch 'SCRATCH_MNT/foo': Read-only file system
+unmounting read-only filesystem
+setting log device read-write
+mounting with read-write log device:
+touch files
+going down:
+unmounting shutdown filesystem:
+setting log device read-only
+mounting filesystem that needs recovery with a read-only log device:
+mount: device write-protected, mounting read-only
+mount: cannot mount device read-only
+unmounting read-only filesystem
+umount: SCRATCH_DEV: not mounted
+mounting filesystem with -o norecovery with a read-only log device:
+mount: SCRATCH_MNT: wrong fs type, bad option, bad superblock on SCRATCH_DEV, missing codepage or helper program, or other error
+unmounting read-only filesystem
+umount: SCRATCH_DEV: not mounted
+setting fs device read-only
+mounting filesystem with -o norecovery with a read-only fs and log device:
+mount: device write-protected, mounting read-only
+unmounting read-only filesystem
+setting fs device read-write
+mounting filesystem with -o norecovery,ro with a read-only log device:
+unmounting read-only filesystem
+setting log device read-write
+mounting filesystem that needs recovery with -o ro:
+*** done
diff --git a/tests/ext4/group b/tests/ext4/group
index a1adc553..5e80e480 100644
--- a/tests/ext4/group
+++ b/tests/ext4/group
@@ -56,3 +56,4 @@
 306 auto rw resize quick
 307 auto ioctl rw defrag
 308 auto ioctl rw prealloc quick defrag
+309 shutdown mount auto quick
-- 
2.21.3


             reply	other threads:[~2020-07-17 10:55 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-17 10:55 Lukas Czerner [this message]
2020-07-26 15:02 ` [PATCH] ext4/309: Test read-only external journal device Eryu Guan
2020-07-27  9:40   ` Lukas Czerner
2020-07-27 10:26 ` [PATCH v2] ext4/002: " Lukas Czerner
2020-08-07 14:41   ` tytso

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200717105544.3201-1-lczerner@redhat.com \
    --to=lczerner@redhat.com \
    --cc=fstests@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.