linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Nikolay Borisov <nborisov@suse.com>
To: linux-btrfs@vger.kernel.org
Cc: Nikolay Borisov <nborisov@suse.com>
Subject: [PATCH 6/7] btrfs-progs: Add test for super block recovery
Date: Tue,  5 Dec 2017 10:39:48 +0200	[thread overview]
Message-ID: <1512463189-24724-7-git-send-email-nborisov@suse.com> (raw)
In-Reply-To: <1512463189-24724-1-git-send-email-nborisov@suse.com>

This functionality regressed some time ago and it was never caught. Seems no
one complained of that, but to be sure add a regression test to prevent future 
regressions.

Signed-off-by: Nikolay Borisov <nborisov@suse.com>
---
 tests/fsck-tests/029-superblock-recovery/test.sh | 64 ++++++++++++++++++++++++
 1 file changed, 64 insertions(+)
 create mode 100755 tests/fsck-tests/029-superblock-recovery/test.sh

diff --git a/tests/fsck-tests/029-superblock-recovery/test.sh b/tests/fsck-tests/029-superblock-recovery/test.sh
new file mode 100755
index 000000000000..beb78d6ccc22
--- /dev/null
+++ b/tests/fsck-tests/029-superblock-recovery/test.sh
@@ -0,0 +1,64 @@
+#!/bin/bash
+# Test that any superblock is correctly detected
+# and fixed by btrfs rescue
+
+source "$TOP/tests/common"
+
+check_prereq btrfs
+check_prereq mkfs.btrfs
+check_prereq btrfs-select-super
+
+setup_root_helper
+
+rm -f dev1
+run_check truncate -s 260G dev1
+loop=$(run_check_stdout $SUDO_HELPER losetup --find --show dev1)
+
+# Create the test file system.
+run_check $SUDO_HELPER "$TOP"/mkfs.btrfs -f "$loop"
+
+function check_corruption {
+	local sb_offset=$1
+	local source_sb=$2
+
+
+	# First we ensure we can mount it successfully
+	run_check $SUDO_HELPER mount $loop "$TEST_MNT"
+	run_check $SUDO_HELPER umount "$TEST_MNT"
+
+	# Now corrupt 1k of the superblock at sb_offset
+	run_check $SUDO_HELPER dd bs=1K count=1 seek=$(($sb_offset + 1)) if=/dev/zero of="$loop"
+
+	#if corrupting one of the sb copies, copy it over the initial superblock
+	if [ ! -z $source_sb ]; then
+		local shift_val=$((16 << $source_sb * 12 ))
+		run_check $SUDO_HELPER dd bs=1K count=4 seek=64 skip=$shift_val if="$loop" of="$loop"
+	fi
+
+	run_mustfail "Mounted fs with corrupted superblock" \
+		$SUDO_HELPER mount $loop "$TEST_MNT"
+
+	# Now run btrfs rescue which should fix the superblock. It uses 2
+	# to signal success of recovery use mayfail to ignore that retval
+	# but still log the output of the command
+	run_mayfail $SUDO_HELPER "$TOP"/btrfs rescue super-recover -yv "$loop"
+	if [ $? != 2 ]; then
+		_fail "couldn't rescue super"
+	fi
+
+	run_check $SUDO_HELPER mount $loop "$TEST_MNT"
+	run_check $SUDO_HELPER umount "$TEST_MNT"
+}
+
+_log "Corrupting first superblock"
+check_corruption 64
+
+_log "Corrupting second superblock"
+check_corruption 65536 1
+
+_log "Corrupting third superblock"
+check_corruption 268435456 2
+
+# Cleanup
+run_check $SUDO_HELPER losetup -d "$loop"
+rm -f dev1
-- 
2.7.4


  parent reply	other threads:[~2017-12-05  8:39 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-05  8:39 [PATCH 0/7] Misc btrfs-progs cleanups/fixes Nikolay Borisov
2017-12-05  8:39 ` [PATCH 1/7] btrfs-progs: Explictly state test.sh must be executable Nikolay Borisov
2017-12-05  8:57   ` Qu Wenruo
2017-12-05  8:39 ` [PATCH 2/7] btrfs-progs: Factor out common print_device_info Nikolay Borisov
2017-12-05  9:02   ` Qu Wenruo
2017-12-05  8:39 ` [PATCH 3/7] btrfs-progs: Remove recover_get_good_super Nikolay Borisov
2017-12-05  9:10   ` Qu Wenruo
2017-12-05  8:39 ` [PATCH 4/7] btrfs-progs: Use list_for_each_entry in write_dev_all_supers Nikolay Borisov
2017-12-05  9:14   ` Qu Wenruo
2017-12-05  9:16     ` Nikolay Borisov
2017-12-07  9:10   ` [PATCH v2] btrfs-progs: Replace usage of list_for_each with list_for_each_entry Nikolay Borisov
2017-12-07  9:59     ` Qu Wenruo
2017-12-05  8:39 ` [PATCH 5/7] btrfs-progs: Document logic of btrfs_read_dev_super Nikolay Borisov
2017-12-05  9:21   ` Qu Wenruo
2017-12-05  8:39 ` Nikolay Borisov [this message]
2017-12-05  9:33   ` [PATCH 6/7] btrfs-progs: Add test for super block recovery Qu Wenruo
2017-12-05 10:04     ` Nikolay Borisov
2017-12-05 11:12       ` Qu Wenruo
2017-12-05 11:26         ` Nikolay Borisov
2017-12-05 12:13           ` Qu Wenruo
2018-01-23 15:07       ` David Sterba
2018-01-23 15:29         ` Nikolay Borisov
2018-01-23 15:39           ` David Sterba
2017-12-05  8:39 ` [PATCH 7/7] btrfs-progs: Fix super-recovery Nikolay Borisov
2017-12-05  9:35   ` Qu Wenruo
2018-01-15  9:17 ` [PATCH 0/7] Misc btrfs-progs cleanups/fixes Nikolay Borisov
2018-01-23 15:40   ` David Sterba

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=1512463189-24724-7-git-send-email-nborisov@suse.com \
    --to=nborisov@suse.com \
    --cc=linux-btrfs@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).