> On Thu, Dec 09, 1999 at 04:32:31PM -0600, Bret Indrelee wrote:
> > If processes can get a highly accurate time value from some sort of global
> > clock, it allows a pair of processes to create a covert channel for passing
> > information. The less secure program monitors the time variences of the
> > high-security program in order to get information about or from them.
> Linux simply does not support real compartmentation and probably never will.
While the first part is true, I really hope that the latter isn't. Once
you leave the small departmental file-server and start to get to larger,
more capable servers (e.g. OLTP systems, etc.), the need for fault tolerance
and fault isolation becomes much greater.
> It only makes sense on mainframes anyways, in the PC age you simply buy another
> box and separate them with a 10 cm gap of air (patent pending) security border.
Excuse me? Who says you can't build a mainframe with commodity processors?
(SGI Origin 2000 - mips microprocessors, IBM SP2 - PowerPC processors,
UNISYS OPUS - pentium Pro processors, UNISYS HMP - Xeon processors).
(Besides, have you checked IBM's revenue figures for 1998 yet? Their
software revenue alone was more than twice Microsofts in 1997, and I
expect it to be roughly the same in 1998, and 1999). IBM's total
revenue may eclipse $100B this year - that is not from selling
PC's, by the way.
What market segment do you consider Linux suitable to fill?
> Complicating the OS is simply not worth it.
Complicating the applications because you don't want to make the
operating system more robust is not a viable fault-tolerance
policy in the long run.
scott
-
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/