Strange mounting issue with ext3

pgf at laptop.org pgf at laptop.org
Tue Apr 7 21:03:38 PDT 2009


 > Roderich Schupp wrote:
 > > On Tue, Apr 7, 2009 at 6:28 PM, Mikkel L. Ellertson
 > > <mikkel at infinity-ltd.com> wrote:
 > >> The last thing I want is having a user-permission-squashed ext3
 > >> partition. For that matter, I am not sure I like the idea of an
 > >> auto-mounted ext[23] file system at all. The last thing I want is
 > >> someone to be able to copy a SUID program from a removable file
 > >> system without having root access in the first place.
 > > 
 > > That whole paragraph makes no sense to me.
 > > 
 > > Chers, Roderich
 > > 
 > It should have been two paragraphs.
 > 
 > Because I sometimes have drives with files from more then one user,
 > I do not want all the files accessible by the console user just by
 > plugging in the drive. Mounting the drive as a
 > "user-permission-squashed" file system would allow this.

i guess i don't understand this.  user-permission-squashed only
makes explicit what is already implicitly true -- i.e., if you
don't have control over who's plugging in the drive, then how do
you have control over whether they're root or not?  if you do
have control, then then don't set up the system to do squashed
mounts.  if you don't, then, well, shouldn't you give up on the
sanctity of uid?

 > Most of the time, I do not want ext[23} formatted drives to be
 > auto-mounted in the first place, even if they get mounted normally,
 > and not "user-permission-squashed".
 > 
 > But I may be too paranoid - I tend to use an encrypted file system
 > and/or a locked hard drive for things I want to keep private. I

now that, i understand.

paul
=---------------------
 paul fox, pgf at laptop.org


More information about the hal mailing list