🚀 go-pugleaf

I think I need to clarify my earlier post.  Many people have sent me
suggestions consisting of 'use ps -aux' to find the problem.  The
problem is that I tried that weeks ago (I've been programming in Unix
for 8 years) and the only oddity that I can find is init using 12% of
the CPU time, which since I have only one NeXTdimension, it's hard to
find out what the noraml is for a ND machine (it's .1% or so on my
NeXTcube).  I know this has something to do with the problem (I meant to
mention it int the last post), but I can find absolutely NO reason for
this machine to be doing that kind of loading.  

For what it's worth, the machine uses NFS to mount 2 other machines (4
partitions total), and is a NetInfo client, and uses NTP for time
service.  None of those should have anything to do with the problem.

On a slightly unrelated note, I upgraded the ND board to 20Mb and it
slowed down even more.  I was told by NeXT that it didn't matter what
order the SIMMS went in (since we have no documentation, and never got
any,even after harassing them for it), but apparantly something doesn't
like 4x4+4x1.

I am NOT a happy NeXT customer right now, I've gotten nothing but the
runaround even getting what I consider to be "normal" information that
should be in the manuals, (some of which we have been unable to get out
of NeXT even though we should have gotten them with our machines).  As
someone doing research and development work for a major university (The
University of Texas at Austin) it would be hard for me to suggest anyone
buy a ND machine at this point.  The concepts are great, someone needs
to teach NeXT how to support a machine.  I can't even get them to answer
a simple question (like SIMM order) wihtout going through my "official
support person".  This is NOT a company I want to deal with.

Chris
-- 
| Chris Petrilli                                             ____
| petrilli@gnu.ai.mit.edu             ``Quayle is a bozoe.'' \  /
| I don't even speak for myself.                              \/