btrfs: restrict snapshotting to own subvolumes
authorDavid Sterba <dsterba@suse.cz>
Wed, 15 Jan 2014 17:15:52 +0000 (18:15 +0100)
committerBen Hutchings <ben@decadent.org.uk>
Tue, 1 Apr 2014 23:58:45 +0000 (00:58 +0100)
commit d024206133ce21936b3d5780359afc00247655b7 upstream.

Currently, any user can snapshot any subvolume if the path is accessible and
thus indirectly create and keep files he does not own under his direcotries.
This is not possible with traditional directories.

In security context, a user can snapshot root filesystem and pin any
potentially buggy binaries, even if the updates are applied.

All the snapshots are visible to the administrator, so it's possible to
verify if there are suspicious snapshots.

Another more practical problem is that any user can pin the space used
by eg. root and cause ENOSPC.

Original report:
https://bugs.launchpad.net/ubuntu/+source/apparmor/+bug/484786

Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
[bwh: Backported to 3.2:
 - Adjust context
 - Use the same cleanup code for success and error cases, as done
   upstream in commit ecd188159efa
   ('switch btrfs_ioctl_snap_create_transid() to fget_light()')]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
fs/btrfs/ioctl.c

index 618ae6f..7cbe2f8 100644 (file)
@@ -1327,12 +1327,17 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file,
                        printk(KERN_INFO "btrfs: Snapshot src from "
                               "another FS\n");
                        ret = -EINVAL;
-                       fput(src_file);
-                       goto out;
+               } else if (!inode_owner_or_capable(src_inode)) {
+                       /*
+                        * Subvolume creation is not restricted, but snapshots
+                        * are limited to own subvolumes only
+                        */
+                       ret = -EPERM;
+               } else {
+                       ret = btrfs_mksubvol(&file->f_path, name, namelen,
+                                            BTRFS_I(src_inode)->root,
+                                            transid, readonly);
                }
-               ret = btrfs_mksubvol(&file->f_path, name, namelen,
-                                    BTRFS_I(src_inode)->root,
-                                    transid, readonly);
                fput(src_file);
        }
 out: