if you want to reproduce a machine check which is a result of hardware
problem try with "dd".
First Create a very huge file like 1 GB.
I did a program in C to create a very huge file but of course a shell
script works as well.
Then use "dd" with a very huge block size as huge as you can and use dd
to copy the file into another file so that you use
all your RAM.
This will reproduce a machine check if you have hardware problems, at
least this was my experience with my Alphastation 500

good luck

Rick


On Sat, 13 Sep 2003 kpneal@pobox.com wrote:

> On Sun, Sep 14, 2003 at 01:40:23AM +0200, Riccardo.Veraldi@fi.infn.it wrote:
> >
> > it could be your ecache
> > I had very similar errors like yours on alphastation 500/333
> > and te problem was a bad ecache module.
> > they coudl nto be changed (motherboard embedded) so I just had to throw my
> > alpha away but my error code was different from 660, so maybe it;s not the
> > case of your problem, hope so for you.
>
> Geeze, I hope it isn't that. This cache is soldered onto the board as
> well. I checked, because I was wondering if I could bump it up to
> 4MB of cache. No such luck.
>
> I think I did notice jumpers to disable the Bcache on the motherboard.
> Disabling the cache would suck, but it would suck less than discarding
> the board.
>
> I did notice that the documented maximum amount of memory for this
> board is a half gig. I've got a whole gig. I yanked out the top two
> DIMMs and I'm going to see how that goes. The trivial test shown
> below didn't cause a crash. If all goes well I'll swap the two sticks
> that are in with the two that are out. I need to know if I have a
> bad stick of memory or not.
>
> > On Sat, 13 Sep 2003 kpneal@pobox.com wrote:
> >
> > > On Sat, Sep 13, 2003 at 01:30:35AM -0400, kpneal@pobox.com wrote:
> > > > I'll see if I can get the initial panic message next time it
> > > > happens.
> > >
> > > Well, that was easy. I have an mfs /tmp.
> > >
> > > % dd if=/dev/zero of=/tmp/blarg
> > >
> > > Warning: received processor correctable error.
> > > Warning: received processor correctable error.
> > > Warning: received processor correctable error.
> > > Warning: received processor correctable error.
> > >
> > > unexpected machine check:
> > >
> > >     mces    = 0x1
> > >     vector  = 0x670
> > >     param   = 0xfffffc0000006068
> > >     pc      = 0xfffffc000051ca74
> > >     ra      = 0xfffffc0000300ac8
> > >     code    = 0x98
> > >     curproc = 0xfffffc00051245c8
> > >         pid = 119, comm = mount_mfs
> > >
> > > I'll try rearranging memory tomorrow. Oh, and the 'reboot' command
> > > in ddb gave me another "correctable" error followed by a hang.
> > >
> > > *sigh*
> > >
> > > This is what happens when a household member decides to show who is in
> > > charge by opening the back door to let the cats get fresh air. Having
> > > outside air come in in the middle of the summer thereby bringing the
> > > temperature in the computer room into the mid-90's (humid) just can't be
> > > good for machines. I can imagine the temperature in the box being 10+
> > > degrees higher than the room air temp, placing it dangerously close to
> > > the outside of the operational bounds.
> > >
> > > Said household member is now removed and isn't coming back. Now I
> > > just have to clean up the damage.
> > >
> > > I'm going to bed.
> > >
> > > Thanks for the help. Let's all cross our fingers for bad memory
> > > and not a bad board.
> > > --
> > > Kevin P. Neal                                http://www.pobox.com/~kpn/
> > >
> > > "What is mathematics? The age-old answer is, of course, that mathematics
> > >  is what mathematicians do." - Donald Knuth
> > >
> > >
> --
> Kevin P. Neal                                http://www.pobox.com/~kpn/
> "Oh, I've heard that paradox a couple of times, but there's something
> about a cat dying and I hate to think of such things."
>   - Dr. Donald Knuth speaking of Schrodinger's cat, December 8, 1999, MIT
>
>