ocfs2: fix rename vs unlink race

If another node unlinks the destination while ocfs2_rename() is waiting on a
cluster lock, ocfs2_rename() simply logs an error and continues. This causes
a crash because the renaming node is now trying to delete a non-existent
inode. The correct solution is to return -ENOENT.

Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>

1 files changed, 10 insertions, 3 deletions

diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c
index 729259016c18..989ac2718587 100644
--- a/fs/ocfs2/namei.c
+++ b/fs/ocfs2/namei.c
@@ -1105,9 +1105,16 @@ static int ocfs2_rename(struct inode *old_dir,
                 goto bail;
         }
 
-        if (!new_de && new_inode)
-                mlog(ML_ERROR, "inode %lu does not exist in it's parent "
-                     "directory!", new_inode->i_ino);
+        if (!new_de && new_inode) {
+                /*
+                 * Target was unlinked by another node while we were
+                 * waiting to get to ocfs2_rename(). There isn't
+                 * anything we can do here to help the situation, so
+                 * bubble up the appropriate error.
+                 */
+                status = -ENOENT;
+                goto bail;
+        }
 
         /* In case we need to overwrite an existing file, we blow it
          * away first */