Okay Bill and everyone, I've got answers a lot of the way there:

[early stuff snipped out from boot message]

>>
>> nubus0 at mainbus0
>> ae0 at nubus0: address 02:60:8c:06:d9:50, type EtherNet card, 32KB memory
>> macvid0 at nubus0: Futura II LX
>> macvid0: Unknown video card ID 0x417 -- Not installing interrupt routine.
>
>Looks ok, I think.
>
>> >What does netstat -r say?
>>
>> Well, here's netstat -nr, from Macii:
>>
>> macii# netstat -nr
>> Routing tables
>>
>> Internet:
>> Destination      Gateway            Flags     Refs     Use    Mtu  Interface
>> 10               link#1             UC          0        0      -  ae0
>> 10.0.0.1         127.0.0.1          UGHS        1       24      -  lo0
>> 127.0.0.1        127.0.0.1          UH          1        0      -  lo0
>>
>>
>> >From "centris:
>>
>> centris# netstat -nr
>> Routing tables
>>
>> Internet:
>> Destination      Gateway            Flags     Refs     Use    Mtu  Interface
>> 10               link#2             UC          0        0      -  ae0
>> 10.0.0.10        127.0.0.1          UGHS        1       24      -  lo0
>> 127.0.0.1        127.0.0.1          UH          1        0      -  lo0
>
>Looks ok, but I'm confused by the link#2 comment. (Q: to the list) is that
>reflecting the fact that ae0 was the second ethernet thing found, or is it
>reflecting a link media choice on ae0? If it's the latter, could it be
>choosing a media which isn't hooked up?

Bill, here's what netstat -ain showed me--maybe since there is onboard
ethernet (sn0) recognized then the ae0 shows up as link#2:

centris# netstat -ain
Name  Mtu   Network     Address              Ipkts Ierrs    Opkts Oerrs  Coll
sn0*  1500  <Link>      08:00:07:6c:79:60        0     0        0     0     0
ae0   1500  <Link>      02:60:8c:06:d9:50        0     0        0     0     0
ae0   1500  10          10.0.0.10
                        224.0.0.1              	 0     0        0     0     0
ppp0* 1500  <Link>                               0     0        0     0     0
ppp1* 1500  <Link>                               0     0        0     0     0
lo0   32768 <Link>                              24     0       24     0     0
lo0   32768 127         127.0.0.1
                        224.0.0.1             	24     0       24     0     0
sl0*  296   <Link>

>
[snip]

>>
>> When I do ifconfig ae0 as you suggest I get an error message:
>>
>> macii# ifconfig ae0 inet 10.0.0.1 netmask 255.255.0.0
>> macii# Apr 25 10:58:05 macii /netbsd: ae0: device timeout
>>
>> The device time message worries me, seems like the card is not responding
>> to the software
>>
>> Any other ideas I should try, or info to supply?
>
>I don't know what to ask. I think the fact that macii's ethernet card won't
>ifconfig is a BAD sign. I bet the ping failure is a symptom of this
>problem.

To this point Alan Briggs suggests:

  >I think this is the crux of the problem.  I think it's related to the
  >fact that the RasterOps Colorboard 364 is not a known beastie as far
  >as interrupt handling goes.  I would have expected a hang on the macII
  >instead of a timeout, though.  Hmm...  Although, if the interrupt was
  >already triggered and the IRQ line was basically wedged in the
  >"interrupt active" state, it might not trigger a processor interrupt...
  >
  >In any event, it would probably be worthwhile to get a video card into
  >the MacII on which we can clear video interrupts, or follow the
  >information on http://www.macbsd.com/macbsd/howto/video.html to help
  >make that card a supported video card.  I should warn you, though, that
  >the Colorboard 264 code looks a bit hairy and I doubt the 364 is any
  >simpler...

  >Actually, I just noticed that the 650 is probably having the same
  >problem with the Futura II LX.  It would be great if we could get the
  >video interrupt clearing information for both of those cards.
  >
  >Thanks!
  >-allen

I'm going to try to supply video interupt information on the Futura LX and
also a Futura SX card that I have to swap with the RasterOps.

Lastly, if anyone's still with me, this idea thanks to Mark Andres:

  >Second, a couple of hints on debugging your network problems.  1) Ethernet
  >works under MacOS but not under NetBSD.  Have you tried testing half and
  >half.  Leave the C650 in MacOS and put the II in NetBSD.  Does the network
  >run OK?  Then, try it vice versa.  2) When pinging, try IP addresses
  >instead of host names.  Does it make a difference?

>
>Maybe try the half-boot idea? Let macii come up in MacOS and centris in
>NetBSD.
>
The half boot macii in Mac OS and centris in NetBSD worked fine with either
the onboard ethernet (sn0) or the 3Com nubus ethernet--after removing the
nubus video card from centris and using the builtin video (bye bye dt for
this experiment, sigh).

Now if the interrupt stuff can be solved, I get my cake and eat it too.

BTW the pinging worked equally badly with IP address or host names--then
worked equally well in the half boot mode.

Thanks to all who helped on this.

-----------------------------
Michel Chalufour
michelc@tiac.net