diff mbox series

e2fsck: subtract acl blocks when setting i_file_acl to zero

Message ID 8e8f277d-6222-5f63-0dcb-a17771a0deff@huawei.com
State Rejected
Headers show
Series e2fsck: subtract acl blocks when setting i_file_acl to zero | expand

Commit Message

Li Jinlin March 17, 2022, 4:54 p.m. UTC 
We got issue as follows:
    [root@localhost ~]# e2fsck -a img
    img: recovering journal
    img: Truncating orphaned inode 188 (uid=0, gid=0, mode=0100666, size=0)
    img: Truncating orphaned inode 174 (uid=0, gid=0, mode=0100666, size=0)
    img: clean, 484/128016 files, 118274/512000 blocks
    [root@localhost ~]# e2fsck -fn img
    e2fsck 1.46.5 (30-Dec-2021)
    Pass 1: Checking inodes, blocks, and sizes
    Inode 174, i_blocks is 2, should be 0.  Fix? no
    
    Inode 188, i_blocks is 2, should be 0.  Fix? no
    
    Pass 2: Checking directory structure
    Pass 3: Checking directory connectivity
    Pass 4: Checking reference counts
    Pass 5: Checking group summary information
    
    img: ********** WARNING: Filesystem still has errors **********
    
    img: 484/128016 files (24.6% non-contiguous), 118274/512000 blocks


File ACL would be set to zero in release_inode_blocks(), but the
blocks count will not be subtract acl blocks, which causes this issue.

To slove this issue, subtract acl blocks when setting i_file_acl to
zero. 

Signed-off-by: LiJinlin <lijinlin3@huawei.com>
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
---
 e2fsck/super.c        |  7 +++++--
 lib/ext2fs/ext2fs.h   |  5 +++++
 lib/ext2fs/ext_attr.c | 15 +++++++++++++--
 3 files changed, 23 insertions(+), 4 deletions(-)

Comments

Zhiqiang Liu March 21, 2022, 1:59 a.m. UTC | #1 
For truncating orphaned inode, we should not remove its ACL in release_inode_blocks(),
otherwise it may cause safety problems due to loss of acl.

diff --git a/e2fsck/super.c b/e2fsck/super.c
index 31e2ffb..6249e12 100644
--- a/e2fsck/super.c
+++ b/e2fsck/super.c
@@ -236,6 +236,10 @@ static int release_inode_blocks(e2fsck_t ctx, ext2_ino_t ino,
         ext2fs_iblk_sub_blocks(fs, EXT2_INODE(inode),
                 pb.truncated_blocks);

+    /* do not clean up file acl if the inode is truncating type */
+    if (inode->i_links_count)
+        return 0;
+
     blk = ext2fs_file_acl_block(fs, EXT2_INODE(inode));
     if (blk) {
         retval = ext2fs_adjust_ea_refcount3(fs, blk, block_buf, -1,
Theodore Ts'o Jan. 18, 2023, 5:53 a.m. UTC | #2 
This patch is not correct.  We don't need to create a new
ext2fs_adjust_ea_refcount4 to return the h_blocks field in the
extended attribute block, since that only thing that we support is a
single xattr block.

The real issue, as Zhiqiang Liu pointed out in [1], is that we should
not be clearing the i_file_acl block is the inode is only being
truncated, and not being unlinked.

[1] https://lore.kernel.org/all/ed518b11-3c38-1c1f-a75d-3293c91f17d4@huawei.com/

						- Ted
diff mbox series

Patch

diff --git a/e2fsck/super.c b/e2fsck/super.c
index 9495e029..715a8dc9 100644
--- a/e2fsck/super.c
+++ b/e2fsck/super.c
@@ -194,6 +194,7 @@  static int release_inode_blocks(e2fsck_t ctx, ext2_ino_t ino,
 	blk64_t				blk;
 	errcode_t			retval;
 	__u32				count;
+	__u32				blocks;
 
 	if (!ext2fs_inode_has_valid_blocks2(fs, EXT2_INODE(inode)))
 		return 0;
@@ -238,8 +239,10 @@  static int release_inode_blocks(e2fsck_t ctx, ext2_ino_t ino,
 
 	blk = ext2fs_file_acl_block(fs, EXT2_INODE(inode));
 	if (blk) {
-		retval = ext2fs_adjust_ea_refcount3(fs, blk, block_buf, -1,
-				&count, ino);
+		retval = ext2fs_adjust_ea_refcount4(fs, blk, block_buf, -1,
+				&count, ino, &blocks);
+		if (retval == 0)
+			ext2fs_iblk_sub_blocks(fs, EXT2_INODE(inode), blocks);
 		if (retval == EXT2_ET_BAD_EA_BLOCK_NUM) {
 			retval = 0;
 			count = 1;
diff --git a/lib/ext2fs/ext2fs.h b/lib/ext2fs/ext2fs.h
index 68f9c1fe..8ebebf31 100644
--- a/lib/ext2fs/ext2fs.h
+++ b/lib/ext2fs/ext2fs.h
@@ -1289,6 +1289,11 @@  extern errcode_t ext2fs_adjust_ea_refcount3(ext2_filsys fs, blk64_t blk,
 					   char *block_buf,
 					   int adjust, __u32 *newcount,
 					   ext2_ino_t inum);
+extern errcode_t ext2fs_adjust_ea_refcount4(ext2_filsys fs, blk64_t blk,
+					   char *block_buf,
+					   int adjust, __u32 *newcount,
+					   ext2_ino_t inum,
+					   __u32 *blocks);
 errcode_t ext2fs_xattrs_write(struct ext2_xattr_handle *handle);
 errcode_t ext2fs_xattrs_read(struct ext2_xattr_handle *handle);
 errcode_t ext2fs_xattrs_read_inode(struct ext2_xattr_handle *handle,
diff --git a/lib/ext2fs/ext_attr.c b/lib/ext2fs/ext_attr.c
index d36fe68d..1538a53a 100644
--- a/lib/ext2fs/ext_attr.c
+++ b/lib/ext2fs/ext_attr.c
@@ -237,9 +237,10 @@  errcode_t ext2fs_write_ext_attr(ext2_filsys fs, blk_t block, void *inbuf)
 /*
  * This function adjusts the reference count of the EA block.
  */
-errcode_t ext2fs_adjust_ea_refcount3(ext2_filsys fs, blk64_t blk,
+errcode_t ext2fs_adjust_ea_refcount4(ext2_filsys fs, blk64_t blk,
 				    char *block_buf, int adjust,
-				    __u32 *newcount, ext2_ino_t inum)
+				    __u32 *newcount, ext2_ino_t inum,
+				    __u32 *blocks)
 {
 	errcode_t	retval;
 	struct ext2_ext_attr_header *header;
@@ -264,6 +265,8 @@  errcode_t ext2fs_adjust_ea_refcount3(ext2_filsys fs, blk64_t blk,
 	header->h_refcount += adjust;
 	if (newcount)
 		*newcount = header->h_refcount;
+	if (blocks)
+		*blocks = header->h_blocks;
 
 	retval = ext2fs_write_ext_attr3(fs, blk, block_buf, inum);
 	if (retval)
@@ -275,6 +278,14 @@  errout:
 	return retval;
 }
 
+errcode_t ext2fs_adjust_ea_refcount3(ext2_filsys fs, blk64_t blk,
+				    char *block_buf, int adjust,
+				    __u32 *newcount, ext2_ino_t inum)
+{
+	return ext2fs_adjust_ea_refcount4(fs, blk, block_buf, adjust,
+					  newcount, 0, NULL);
+}
+
 errcode_t ext2fs_adjust_ea_refcount2(ext2_filsys fs, blk64_t blk,
 				    char *block_buf, int adjust,
 				    __u32 *newcount)