Chris Evans wrote:
>
> On Tue, 26 May 1998, Meelis Roos wrote:
>
> > CE> The problem is probably that you haven't configured resource limits on
> > CE> your box. Try doing just this (RLIMIT_DATA is a good one, RLIMIT_RSS is
> > CE> annoyingly the only RSS_* not yet implemented). Alternatively run crashme
> > CE> so that it uses less memory and/or less concurrency. A command line option
> > CE> I believe.
> >
> > No! The kernel should must crash when the limits are not set.
> > If it crashes, it's a bug.
>
> Are you sure it's "crashed"? Perhaps the system is just "very busy" to the
> extent of appearing crashed, because you have allowed crashme to grab too
> much virtual memory? Does the "crashed" system ping? echo characters to
> the console? switch vt's?
>
Well, how do you define "crash"? :-))
The symptoms I'm talking about are:
For more than 15 minutes:
*) Not responsive to network (no ping, no telnet)
*) Not responsive to console (no VT switching, no
noticeable reaction on keyboard input)
*) No noticeable disk I/O
I have verified this behaviour on two different
machines now (out of two tested)
- andreas
-- Andreas Haumer | email: andreas@xss.co.at | PGP key *x Software + Systeme | phone: +43.1.6001508 | available Buchengasse 67/8 | +43.664.3004449 | on request. A-1100 Vienna, Austria | fax: +43.1.6001507 |- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu