[rescue] Now what, SGI Challenge L Department

Sheldon T. Hall shel at cmhcsys.com
Fri Jan 16 16:10:57 CST 2004


Lionel Peterson writes ...
> --- "Sheldon T. Hall" <shel at cmhcsys.com> wrote:
> >
> > I'm getting memory errors on my Challenge L.  So far, there's not
> > much of a pattern, and none have been uncorrectible.
>
> OK, time for a dumb question - why do you feel you need to correct
> these? By virtue of the design of your machines memory subsystem, these
> appear to *me* as non-problems.

Well, that's why I'm asking!  I don't know if they are a problem or not.
Seemingly, they are all getting caught and corrected.

Of course, that makes one wonder ... in a non-ECC machine, would it be
common to have a single-bit memory error every few days?  Since they
wouldn't be automatically caught or corrected, such memory errors would
probably be harmful.

> Having said that - are you sure there was ever a time when these errors
> *didn't* occur?

No, I'm not.  The logging arrangements on the Challenge, as I got it, were
pretty lame.  I suppose they were the IRIX 6.5 defaults, since the OS n the
machine seemed to be a fresh install, but in any case, very little of use
was getting logged.  I improved this, and got more information, and at some
point these errors showed up.

Were the errors there before, but not logged?  I dunno.  Were they logged,
but somewhere I didn't see?  Beats me.

> I suspect (and I defer to the group for a final answer)
> that these are common errors, and you only noticed them recently - they
> have always been happening... ABsent a pattern, I don't know what you
> could do (aside from reseating all the SIMMs ;^)...

I guess that's sorta what I was asking ... "Is this normal?"

-Shel



More information about the rescue mailing list