Skip to contentSkip to conversation

btrfs-progs: cmds/subvolume: try to delete subvolume by id when its path can't be reoslved

PATCH In review Qu Wenruo 3 participants 3 messages Jun 28, 2021 – Aug 17, 2021 Btrfs file system

lore
Share & tools
lore

Qu Wenruo

author
#0 lore raw

PATCH 1/1 btrfs-progs: cmds/subvolume: try to delete subvolume by id when its path can't be reoslved

There is a recent report of ghost subvolumes where such subvolumes has no ROOT_REF/BACKREF, and 0 root ref.
But without an orphan item, thus kernel won't queue them for cleanup.

Such ghost subvolumes are just here to take up space, and no way to delete them except by btrfs check, which will try to fix the problem by adding orphan item.

There is a kernel patch submitted to allow btrfs to detect such ghost subvolumes and queue them for cleanup.

But btrfs-progs will not continue to call the ioctl if it can't find the full subvolume path.

Thus this patch will loose the restriction by allowing btrfs-progs to continue to call the ioctl even if it can't grab the subvolume path.

---
 cmds/subvolume.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
cmds/subvolume.c +16 -1
index 9bd178082495..4881ba5be4b1 100644
@@ -258,6 +258,7 @@ static int cmd_subvol_delete(const struct cmd_struct *cmd,
 	char	*path = NULL;
 	DIR	*dirstream = NULL;
 	int commit_mode = 0;
+	bool subvol_path_not_found = false;
 	u8 fsid[BTRFS_FSID_SIZE];
 	u64 subvolid = 0;
 	char uuidbuf[BTRFS_UUID_UNPARSED_SIZE];
@@ -319,6 +320,18 @@ static int cmd_subvol_delete(const struct cmd_struct *cmd,
 
 		path = argv[cnt];
 		err = btrfs_util_subvolume_path(path, subvolid, &subvol);
+		/*
+		 * If the subvolume is really not referred by anyone, and
+		 * refs is 0, newer kernel can handle it by just adding an
+		 * orphan item and queue it for cleanup.
+		 *
+		 * In this case, just let kernel to handle it, we do no
+		 * extra handling.
+		 */
+		if (err == BTRFS_UTIL_ERROR_SUBVOLUME_NOT_FOUND) {
+			subvol_path_not_found = true;
+			goto again;
+		}
 		if (err) {
 			error_btrfs_util(err);
 			ret = 1;
@@ -395,8 +408,10 @@ again:
 
 	if (subvolid == 0)
 		pr_verbose(MUST_LOG, "'%s/%s'\n", dname, vname);
-	else
+	else if (!subvol_path_not_found)
 		pr_verbose(MUST_LOG, "'%s'\n", full_subvolpath);
+	else
+		pr_verbose(MUST_LOG, "subvolid=%llu\n", subvolid);
 
 	if (subvolid == 0)
 		err = btrfs_util_delete_subvolume_fd(fd, vname, 0);
Signature · 1 line
2.32.0
#1 lore raw
Quoted 11 lines · Qu Wenruo

There is a recent report of ghost subvolumes where such subvolumes has no ROOT_REF/BACKREF, and 0 root ref.
But without an orphan item, thus kernel won't queue them for cleanup.

Such ghost subvolumes are just here to take up space, and no way to delete them except by btrfs check, which will try to fix the problem by adding orphan item.

There is a kernel patch submitted to allow btrfs to detect such ghost subvolumes and queue them for cleanup.

Quoted 7 lines

But btrfs-progs will not continue to call the ioctl if it can't find the full subvolume path.

Thus this patch will loose the restriction by allowing btrfs-progs to continue to call the ioctl even if it can't grab the subvolume path.

Added to devel, thanks. Please send a test case.

#2 lore raw
Quoted 15 lines · David Sterba
Quoted 14 lines · Qu Wenruo

There is a recent report of ghost subvolumes where such subvolumes has no ROOT_REF/BACKREF, and 0 root ref.
But without an orphan item, thus kernel won't queue them for cleanup.

Such ghost subvolumes are just here to take up space, and no way to delete them except by btrfs check, which will try to fix the problem by adding orphan item.

There is a kernel patch submitted to allow btrfs to detect such ghost subvolumes and queue them for cleanup.

But btrfs-progs will not continue to call the ioctl if it can't find the full subvolume path.

Quoted 7 lines
Quoted 4 lines

Thus this patch will loose the restriction by allowing btrfs-progs to continue to call the ioctl even if it can't grab the subvolume path.

Added to devel, thanks. Please send a test case.

That needs kernel support, and I didn't see kernel patch get merged yet.

Thanks,
Qu