Re: Ext3 filesystem info?

Jeff Haumont (haumont@acm.org)
Tue, 21 Sep 1999 10:21:00 -0500


"Theodore Y. Ts'o" wrote:

> Date: Mon, 20 Sep 1999 11:18:42 -0500 (CDT)
> From: Jesse Pollard <pollard@tomcat.admin.navo.hpc.mil>
>
> >Anyone know if the coda project intends to make their ACLs work the same way
> >as these do for ext2? It'd be great to have one set of common commands to
> >learn for working with ACLs ...
>
> I'd be much happier if the ACL interface moved into the VFS layer, and up
> to individual file systems to support/not support ACL's. This way the
> implementation would become isolated from the kernel support, allowing a
> common set of user utilities, and user/OS interface.
>
> The problem is that different, already established filesystems: AFS,
> Coda, NTFS, etc., all have different ACL semantics. For example, AFS
> only has an ACL on a per-directory basis. I'm not sure about Coda, but
> it may be the same as AFS. NTFS uses 128 bit UUID's in its ACL's to
> name users and groups. The POSIX acl interface uses uid_t and gid_t for
> user and group id's.
>
> So it would be *nice* to do this, but there's quite a lot of design work
> to make the interfaces similar enough that a single interface could be
> used at both the UI and system call level. I won't say that it's
> impossible, but it's definitely non-trivial.
>

The common interface wouldn't have to try to use every feature of the ACLs in the
underlying filesystem. Even a least-common-demoninator interface would be much
better than nothing. Maybe restrict to directory ACLs only, etc.

Another option would be to write some massively intelligent userspace library that
could query the filesystem and divine the correct way to do things. Not something
I'd want to try to debug, verify to be correct and secure, or maintain though ...

Jeff

--
Jeff Haumont <haumont@acm.org>

- 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/