5 May
2018
5 May
'18
2:35 a.m.
On May 4, 2018, at 5:41 PM, Robert Armstrong <bob
at jfcl.com> wrote:
Yes... I wasn't referring to the original implementation, but rather to what one could
do when creating a compatible implementation in another DECnet system. I did my Python
version as a single circuit model, but thinking about what Johnny said I'm inclined to
revise it.
It's not surprising the original Multinet didn't do such a thing; it's quite
obvious that the designers had very little clue about DECnet.
paul
It would certainly be plausible to model Multinet
as a multidrop link
I had forgotten about multidrop DDCMP connections. I agree that would
work for what we want to do, but Multinet didn't implement it that way. The
only fix I can see would be to write another DECnet device driver that does.
In the current implementation there's a 1-1 relation between the Multinet
lines, circuits and tunnels -