>>> Hmm, if a parity error happens, does it mean it's going to happen
>>> again and again ?
>> Most likely.  And if ECC can't cover it up you'll get all kinds of
>> nastiness from it.
> Do you have a reference for this?  My understanding is that most
> memory errors are transient, not permanent.

I didn't write it, but it matches my experience. While I have
occasionally seen memory errors that never recur, in my experience, a
memory error usually indicates that further errors at the same location
will follow.  (Occasionally errors at different locations within the
same FRU, such as can be caused by flaky seating, but more often the
very same bit in error.)

Why my experience has been that makes an interesting topic for
speculation; my own guess would be that I stick to older machines with
lower RAM densities, for which background radioactivity is
significantly less likely to introduce errors.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B