Has anyone seen/heard of ATMega128s spuriously
erasing EEPROM contents?

We're seeing a small but increasing number of
instances of erases (or at least, rewrites to $FF).
The number is too small for us to be certain of a
pattern in locations, but so far we have at a minimum
two distinct locations which have been erased,
possibly three (it's hard to know for certain because
two of the examples have had an erase occur in
a memory location which triggers a rewrite of a
large portion of the EEPROM, including those
four bytes, which hides the evidence).

Nothing in the errata; I'm on my way to the
AVRFreaks forum to check out what they have
to say.

The chip is protected by a supervisory chip which
keeps it in reset below 2.9V- it should be good
to operate down to 2.7V.

Mike H.
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist