--------- Received message begins Here ---------
>
> Jesse Pollard writes:
> > [Albert D. Cahalan]
> >> Jesse Pollard writes:
[snip]
> > synchronize from the sync system call, write_owner from the chown system
> > call - which, by the way, is a resource control problem and has to be/should
> > be disabled on most systems (especially those with quota limits). Several
> > others trace all the way back to Multics.
>
> It is NT that made these into permission bits.
>
> NT lets you chown() a file to yourself (only to yourself) if the
> file's ACL grants you this right. Nobody can chown() a file to
> somebody else, but any file owner can grant the "take ownership"
> right to somebody else. People with WRITE_ACL can grant themselves
> this right too.
>
> WRITE_ACL sure isn't UNIX-like. Neither is DELETE, which lets you
> bypass directory write permission to delete a specified file.
> NT has both of these.
WRITE_ACL is more like the VAX/VMS ACLs - which is where I believe NT got its
model from anyway.
So with DELETE you allow a corrupted directory to be created?
Or is that considered normal to have file names but no file? If I can
delete a file and remove it from the directory entry then I do have write
permission.
> > Most of these are POSIX ACLs.
>
> The withdrawn POSIX draft only has rwx permissions. I have draft 17.
>
I do seem to remember some discussion about more being included (things
like APPEND and EDIT_ACL permissions), but I didn't realize they never made it.
-------------------------------------------------------------------------
Jesse I Pollard, II
Email: pollard@navo.hpc.mil
Any opinions expressed are solely my own.
-
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/
This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:27 EST