diff options
| author | Ned Bass <[email protected]> | 2012-12-12 15:55:51 -0800 |
|---|---|---|
| committer | Brian Behlendorf <[email protected]> | 2012-12-13 08:57:11 -0800 |
| commit | 7afcf5b1da83549bfba70a61fae7a00eaa63c2b0 (patch) | |
| tree | 50234a20990c1ffc0106bda9d6e72de32fd08704 | |
| parent | 2ae10319624869eef0954cdfcdb6f5e33cba2e09 (diff) | |
Avoid ELOOP on auto-mounted snapshots
Ensure that the path member pointers are associated with the
newly-mounted snapshot when zpl_snapdir_automount() returns. Otherwise
the follow_automount() function may be called repeatedly, leading to an
incorrect ELOOP error return. This problem was observed as a 'Too many
levels of symbolic links' error from user-space commands accessing an
unmounted snapshot in the .zfs/snapshot directory.
Signed-off-by: Brian Behlendorf <[email protected]>
Closes #816
| -rw-r--r-- | module/zfs/zpl_ctldir.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/module/zfs/zpl_ctldir.c b/module/zfs/zpl_ctldir.c index 7dfaf6ebc..a2d8fa9fe 100644 --- a/module/zfs/zpl_ctldir.c +++ b/module/zfs/zpl_ctldir.c @@ -357,6 +357,13 @@ zpl_snapdir_automount(struct path *path) return ERR_PTR(error); /* + * Ensure path->dentry points to the dentry for the root of the + * newly-mounted snapshot, otherwise this function may be called + * repeatedly which can lead to an incorrect ELOOP error return. + */ + follow_up(path); + + /* * Rather than returning the new vfsmount for the snapshot we must * return NULL to indicate a mount collision. This is done because * the user space mount calls do_add_mount() which adds the vfsmount |