Messages in this thread | | | Date | Wed, 3 Mar 1999 19:03:15 -0500 (EST) | From | Alexander Viro <> | Subject | Re: EXT2_UNRM_FL |
| |
On Wed, 3 Mar 1999, Theodore Y. Ts'o wrote:
> I guess I'm little confused about how NO_UNLINK would be useful. Does > it protect against file renames as well as unlinks? What about > truncates? If it doesn't protect the data, what are the applications > where it's useful?
NO_UNLINK is the third member of the same group as APPEND_ONLY and IMMUTABLE. Semantics: you can't create or remove a link to such object (i.e. no rename, no link, no rmdir, no unlink). Wrt to setting/changing such flag it is equivalent to other two. Examples: mkdir ~luser/.rhosts and set no-unlink on it (immutable is an overkill, append-only has unpleasant side-effects); anything you want to modify but don't want to remove; damn it, core in pwd of process that dies in inconvenient times ;-) Besides, it's pretty logical element of the same group - all of them give namespace-level protection, but other two also give contents protection. Which may be not what you want. It's BSD thing (just as other two are). Support of that thing on VFS level costs us nothing. Filesystems may map whatever they want on this bit.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |