#!/bin/bash
# set -x
set -e

# lfs_migrate: a simple tool to copy and check files.
#
# To avoid allocating objects on one or more OSTs, they should be
# deactivated on the MDS via "lctl --device {device_number} deactivate",
# where {device_number} is from the output of "lctl dl" on the MDS.
#
# To guard against corruption, the file is compared after migration
# to verify the copy is correct and the file has not been modified.
# This is not a protection against the file being open by another
# process, but it would catch the worst cases of in-use files, but
# to be 100% safe the administrator needs to ensure this is safe.

RSYNC=${RSYNC:-rsync}
LFS_MIGRATE_RSYNC_MODE=${LFS_MIGRATE_RSYNC_MODE:-false}
ECHO=echo
LFS=${LFS:-lfs}
LFS_SIZE_OPT="-s"

usage() {
    cat -- <<USAGE 1>&2
usage: lfs_migrate [-c <stripe_count>] [-h] [-l] [-n] [-q] [-R] [-s] [-y] [-0]
                   [file|dir ...]
    -c <stripe_count>
       restripe file using the specified stripe count
    -h show this usage message
    -l migrate files with hard links (skip by default for rsync)
    -n only print the names of files to be migrated
    -q run quietly (don't print filenames or status)
    -R restripe file using default directory striping
    -s skip file data comparison after migrate
    -y answer 'y' to usage question
    -0 input file names on stdin are separated by a null character

The -c <stripe_count> option may not be specified at the same time as
the -R option.

If a directory is an argument, all files in the directory are migrated.
If no file/directory is given, the file list is read from standard input.

e.g.: lfs_migrate /mnt/lustre/dir
      lfs find /test -O test-OST0004 -size +4G | lfs_migrate -y
USAGE
    exit 1
}

OPT_CHECK=y
OPT_STRIPE_COUNT=""

while getopts "c:hlnqRsy0" opt $*; do
    case $opt in
	c) OPT_STRIPE_COUNT=$OPTARG;;
	l) OPT_NLINK=y;;
	n) OPT_DRYRUN=n; OPT_YES=y;;
	q) ECHO=:;;
	R) OPT_RESTRIPE=y;;
	s) OPT_CHECK="";;
	y) OPT_YES=y;;
	0) OPT_NULL=y;;
	h|\?) usage;;
    esac
done
shift $((OPTIND - 1))

if [ "$OPT_STRIPE_COUNT" -a "$OPT_RESTRIPE" ]; then
	echo ""
	echo "$(basename $0) error: The -c <stripe_count> option may not" 1>&2
	echo "be specified at the same time as the -R option." 1>&2
	exit 1
fi

if [ -z "$OPT_YES" ]; then
	echo ""
	echo "lfs_migrate is currently NOT SAFE for moving in-use files." 1>&2
	echo "Use it only when you are sure migrated files are unused." 1>&2
	echo "" 1>&2
	echo "If emptying an OST that is active on the MDS, new files may" 1>&2
	echo "use it.  To stop allocating any new objects on OSTNNNN run:" 1>&2
	echo "  lctl set_param osp.<fsname>-OSTNNNN*.max_create_count=0'" 1>&2
	echo "on each MDS using the OST(s) being emptied." 1>&2
	echo -n "Continue? (y/n) "
	read CHECK
	[ "$CHECK" != "y" -a "$CHECK" != "yes" ] && exit 1
fi

# if rsync has --xattr support, then try to copy the xattrs.
$RSYNC --help 2>&1 | grep -q xattr && RSYNC_OPTS="$RSYNC_OPTS -X"
$RSYNC --help 2>&1 | grep -q acls && RSYNC_OPTS="$RSYNC_OPTS -A"
# If rsync copies lustre xattrs in the future, then we can skip lfs (bug 22189)
strings $(which $RSYNC) 2>&1 | grep -q lustre && LFS=:

# rsync creates its temporary files with lenient permissions, even if
# permissions on the original files are more strict. Tighten umask here
# to avoid the brief window where unprivileged users might be able to
# access the temporary file.
umask 0077

# This is needed for 1.8 Interoperability and can be removed in the future
$LFS getstripe --help 2>&1 | grep -q stripe-size && LFS_SIZE_OPT="-S"

lfs_migrate() {
	while IFS='' read -d '' OLDNAME; do
		$ECHO -n "$OLDNAME: "

		# avoid duplicate stat if possible
		TYPE_LINK=($(LANG=C stat -c "%h %F" "$OLDNAME" || true))

		# skip non-regular files, since they don't have any objects
		# and there is no point in trying to migrate them.
		if [ "${TYPE_LINK[1]}" != "regular" ]; then
			echo -e "not a regular file, skipped"
			continue
		fi

		# working out write perms is hard, let the shell do it
		if [ ! -w "$OLDNAME" ]; then
			echo -e "no write permission, skipped"
			continue
		fi

		if [ "$OPT_DRYRUN" ]; then
			echo -e "dry run, skipped"
			continue
		fi

		if [ "$OPT_RESTRIPE" ]; then
			UNLINK=""
		else
		# if rsync copies Lustre xattrs properly in the future
		# (i.e. before the file data, so that it preserves striping)
		# then we don't need to do this getstripe/mktemp stuff.
			UNLINK="-u"

			[ "$OPT_STRIPE_COUNT" ] && COUNT=$OPT_STRIPE_COUNT ||
				COUNT=$($LFS getstripe -c "$OLDNAME" \
					2> /dev/null)
			SIZE=$($LFS getstripe $LFS_SIZE_OPT "$OLDNAME" \
			       2> /dev/null)

			[ -z "$COUNT" -o -z "$SIZE" ] && UNLINK=""
			SIZE=${LFS_SIZE_OPT}${SIZE}
		fi

		# first try to migrate inside lustre
		# if failed go back to old rsync mode
		if [[ $LFS_MIGRATE_RSYNC_MODE == false ]]; then
			if $LFS migrate -c${COUNT} ${SIZE} "$OLDNAME"; then
				$ECHO "done"
				continue
			else
				echo "falling back to rsync-based migration"
				LFS_MIGRATE_RSYNC_MODE=true
			fi
		fi

		if [ -z "$OPT_NLINK" -a ${TYPE_LINK[0]} -gt 1 ]; then
			echo -e "multiple hard links, skipped"
			continue
		fi

		NEWNAME=$(mktemp $UNLINK "$OLDNAME.tmp.XXXXXX")
		if [ $? -ne 0 -o -z "$NEWNAME" ]; then
			echo -e "\r$OLDNAME: can't make temp file, skipped" 1>&2
			continue
		fi

		[ "$UNLINK" ] && $LFS setstripe -c${COUNT} ${SIZE} "$NEWNAME"

		# we use --inplace, since we created our own temp file already
		if ! $RSYNC -a --inplace $RSYNC_OPTS "$OLDNAME" "$NEWNAME";then
			echo -e "\r$OLDNAME: copy error, exiting" 1>&2
			rm -f "$NEWNAME"
			exit 4
		fi

		if [ "$OPT_CHECK" ] && ! cmp -s "$OLDNAME" "$NEWNAME"; then
			echo -e "\r$NEWNAME: compare failed, exiting" 1>&2
			exit 8
		fi

		if ! mv "$NEWNAME" "$OLDNAME"; then
			echo -e "\r$OLDNAME: rename error, exiting" 1>&2
			exit 12
		fi
		$ECHO "done"
	done
}

if [ "$#" -eq 0 ]; then
	if [ "$OPT_NULL" ]; then
		lfs_migrate
	else
		tr '\n' '\0' | lfs_migrate
	fi
else
	while [ "$1" ]; do
		if [ -d "$1" ]; then
			lfs find "$1" -type f -print0 | lfs_migrate
		else
			echo -en "$1\0" | lfs_migrate
		fi
		shift
	done
fi