On Tue, 2002-01-15 at 12:05, John J. Trammell wrote:
> On Tue, Jan 15, 2002 at 10:12:00AM -0600, Phil Mendelsohn wrote:
> > On Tue, Jan 15, 2002 at 09:18:43AM -0600, John J. Trammell wrote:
> > > On Mon, Jan 14, 2002 at 10:56:02AM -0600, Joel Rosenberg wrote:
> > > > On Mon, 2002-01-14 at 09:59, John J. Trammell wrote:
> > > > > [snip]
> > > > > Another good reason I haven't seen yet is that if say /home/
> > > > > has its own partition, a user app going nuts and filling up
> > > > > the partition won't trash the machine.  Nice segmentation there.
> > > > > 
> > > > Don't disk quotas do a better job of that?  
> > > >  
> > > 
> > > Perhaps in /home, but another poster mentioned /var...
> > 
> > OK -- but from a security standpoint, no one that isn't capable of
> > fixing the system should be allowed to write in such a way that they
> > can crash the system.  Actually, no user should be allowed to do
> > *anything* they can't fix, but we have to be careful starting down
> > that path.  
> > 
> > Either way, you are correct from a pragmatic point of view (I'd
> > say), but from an idealistic point of view, one could argue that
> > *every* user (even daemons and root) should have quotas, disk space
> > being finite.
> > 
> 
> Good, because I'm a pragmatist.  :-)
> 
> The thought of sitting down to work out new disk quotas every time I
> add a user gives me the willies.  It is far less work for me in my
> current position to just partition.
> 
> -- 
I'm fairly sure that that process can and should be automated, if you're going to be doing it at all often. 
-- 
-------------------------------------
There's a widow in sleepy Chester
  Who weeps for her only son;
There's a grave on the Pabeng River,
  A grave that the Burmans shun,
And there's Subadar Prag Tewarri
  Who tells how the work was done.
-------------------------------------