Amendment to my Cry for Help

Corbett J. Klempay cklempay@acm.jhu.edu
Sat Dec 12 16:20:16 1998


One thing I just realized a few minutes ago after talking to our lab
admin...not only did putting 2 Lite-Ons in our master die, but they have
also tried putting 2 actual Tulips (both Netgear 310's).  They report that
the behavior was no different between 2 Lite-Ons and 2 Tulips; operation
is erratic and intermittent (he said "it's seemingly random; sometimes it
works, sometimes it doesn't find the card...and when it works, it dies
under load").

Some ideas we had to try: (comments welcome)

	- See if a current 2.1.x kernel behaves any differently
	- Try alternate combinations of which PCI slots the cards are
	  seated in
	- Put the 2 cards in a compute node, where the machine would be
	  running a single UDMA Fireball 6.4; we are worried there's some
	  sketchy interplay involving the Promise (NCR 875-based) SCSI
	  card/BIOS  

------------------------------------------------------------------------------
Corbett J. Klempay			         Quote of the Week:
http://www2.acm.jhu.edu/~cklempay  "Perl is worse than Python because
				    people wanted it worse." Larry Wall

PGP Fingerprint: 7DA2 DB6E 7F5E 8973 A8E7  347B 2429 7728 76C2 BEA1
------------------------------------------------------------------------------