From: Sam Vilain (sam_at_vilain.net)
Date: Mon 10 Feb 2003 - 16:56:51 GMT
On Fri, 07 Feb 2003 11:48, you wrote:
> I'm not saying that. I'm just saying that it's unlikely that the RAM
> has suddenly failed on this new machine. I might suspect it a bit
> more on a 4-year-old one.
New hardware is just as likely as old hardware to have a fault which only
exhibits itself occasionally. Of course, every time a piece of hardware
is touched by grubby ungrounded fingers it `rolls again' on its chances of
developing a fault :-).
A better point would be that a Dell server is almost certain to have ECC
RAM, making the chance of a fault very slim indeed.
If only PC hardware would dump its memory to swap in the event of a system
crash like Sparc hardware does ... would make debugging this sort of thing
easier.
Oh, well look at that there is a patch to Linux to do that (assuming your
VM/kernel is not so smashed that it can't even write to swap). Take a
look at the `crash' debian package. Worth installing on every machine,
I'd say.
-- Sam Vilain, sam_at_vilain.net"Automatic" simply means that you cannot repair it yourself. - anon.