From b51b816e769842f9ef422896f89b0cbbcf230b01 Mon Sep 17 00:00:00 2001
From: Joseph Qi <joseph.qi@linux.alibaba.com>
Date: Thu, 10 Nov 2022 17:14:42 +0800
Subject: [PATCH] fsck.xfs: try to replay log if dirty during xfs_repair

It is possible that when the machine crashes, the fs is in inconsistent
state with the journal log not yet replayed. This can drop the machine
into the rescue shell because currently fsck.xfs does not address the
journal replay, which brings maintenance trouble.

So try to replay log if it is dirty using mount/umount way which is
suggested in xfs_repair manpage. Note that we only do this for root
device so that we don't want to touch data device.

Administrators have to enable this feature by systemd command options
fsck.mode=force and fsck.repair=yes. For other options, we leave it
behave like before, e.g. drop to the rescue shell if repair detects
any corruptions and need administators repair manually.

Signed-off-by: Joseph Qi <joseph.qi@linux.alibaba.com>
---
 fsck/xfs_fsck.sh | 31 +++++++++++++++++++++++++++++--
 1 file changed, 29 insertions(+), 2 deletions(-)

diff --git a/fsck/xfs_fsck.sh b/fsck/xfs_fsck.sh
index 6af0f224..768611c7 100755
--- a/fsck/xfs_fsck.sh
+++ b/fsck/xfs_fsck.sh
@@ -31,10 +31,12 @@ repair2fsck_code() {
 
 AUTO=false
 FORCE=false
+REPLAY=false
 while getopts ":aApyf" c
 do
 	case $c in
-	a|A|p|y)	AUTO=true;;
+	a|A|p)		AUTO=true;;
+	y)		REPLAY=true;;
 	f)      	FORCE=true;;
 	esac
 done
@@ -64,7 +66,32 @@ fi
 
 if $FORCE; then
 	xfs_repair -e $DEV
-	repair2fsck_code $?
+	error=$?
+	if [ $error -eq 2 ] && [ $REPLAY = true ]; then
+		for x in $(cat /proc/cmdline); do
+			case $x in
+				root=*)
+					ROOT="${x#root=}"
+				;;
+				rootflags=*)
+					ROOTFLAGS="-o ${x#rootflags=}"
+				;;
+			esac
+		done
+
+		test -b "$ROOT" || ROOT=$(blkid -t "$ROOT" -o device)
+		if [ $DEV = $ROOT ]; then
+			echo "Replaying log for $DEV"
+			mkdir -p /tmp/replay_mnt || exit 1
+			mount $DEV /tmp/replay_mnt $ROOTFLAGS || exit 1
+			umount /tmp/replay_mnt
+			rm -d /tmp/replay_mnt
+
+			xfs_repair -e $DEV
+			error=$?
+		fi
+	fi
+	repair2fsck_code $error
 	exit $?
 fi
 
-- 
2.24.4