[ILUG] oom fixes and kernel traffic...

[John P. Looney]

On Tue, Aug 03, 1999 at 01:39:05PM +0100, kevin lyda mentioned:
> jplooney at compapp.dcu.ie spoke thusly:
> > That's the side Microsoft took in the Windows 3.11 days - if a program
> >crashes your machine, that's not our problem - don't run the program.
> >
> > If your machine is unusable, for anyr reason, what so ever, it's Linux's
> >fault.
> 
> really?  try this as root:
> 
> # dd of=/dev/kmem if=/dev/zero
> # dd if=/dev/zero of=`mount | grep ' / '| awk '{print $1}'`
> # rm /bin/init

> not satisfied with shell killing?
> 
> int main() {while(fork());}
 
 There is a difference between killing your machine with a specific
command, designed to kill it, and running netscape or xanim, and it killing
it.

> the oom stuff's annoying, but no os is foolproof.  the universe pumps out
> very ingenius fools.  as already mentioned there's a lot of discussion on
> linux-kernel about oom (again).  there are reasons why your suggested fix
> won't work - i haven't followed the topic so i don't know them any more,
> but pop over and read up on it.

 I just want OOM handling that is 10% as good as Windows, FreeBSD or
Solaris. Is that too much to ask for ?

Kate

-- 
Microsoft - the best reason in the world to drink beer