Re: securety related question...

On Sun, 6 Oct 2002, Erika Pacholleck wrote:
> [06.10.2002] Bill maltby - LFS Related <-- :
> > On Sun, 6 Oct 2002, Richard Lightman wrote:
> > 
> > > Could you check that you can bind a rw device ro? I read somewhere
> > > that there was a patch to keep separate mount flags for each mount,
> > > but it does not appear to be there yet:
> > 
> > I get the same results with the same version of mount that you have.
> > Further, if I do a "remount,ro", the remount affects the "base" mount as
> > well as the bound mount. So, until separate flags for the bound mount
> > point become available, there is reduced security available for the "host"
> > partition if it must be mounted rw.
> 
> Isn't it logical? The --bind is the only option (no additionals) and its
> purpose is to make the *already mounted* partition available at another
> place, that *same* partition (and this means with all its flags).

This is logical for me only if I presume that the designer/coder does not
provide functionality that I would provide. Consider:
  - a mount is only a redirection operation upon an i-node
  - this re-direct-associated control structure, could have a set of
    flags totally independent of the "base" mounting
  - given that, it seems rational that different "permissions" could hold,
    depending upon which path was used to access the position
  - so I could have write permissions through one path and not have them
    in another path.

There are some "complications" that would have to be handled, such as
those you mention.

> How are we going to end up, if --bind would allow to change the flags,

It would changing only the flags associated with the re-direction, not the
base mounting, in my theoretical implementation.

> suddenly a by intention mounted ro partition will be rw, and what you
> believe protected by ro is suddenly deleted via the bound mounting point!

So? I see no problem with that. Further, it is no different than current
(flawed) *IX implementations for directories. Have some fun, try this.

Cd to some directory. From another session, rm -rf the directory in whcich
the other is positioned. Now go to the first session again. Do pwd. Then
do "cd ..". Interesting?

-- 
Bill Maltby
[email protected]

-- 
Unsubscribe: send email to [email protected]
and put 'unsubscribe lfs-security' in the subject header of the message