[Info-vax] What would you miss if DECnet got the chop? Was: "bad select 38" (OpenSSL on VMS)

[Dirk Munk]

Michael Moroney wrote:
> Dirk Munk <munk at home.nl> writes:
>
>> It just occurred to me that you will have to look into these matters anyway.
>
>> TCP/IP services 10.5 will be based on Multinet, and Multinet has a PWIP
>> driver.
>
>> DECnet-Plus is build on top of the PWIP driver, so you will have to
>> check if the Multinet PWIP driver is DECnet-Plus compatible, or perhaps
>> use the HPE PWIP driver?
>
>> Multinet also has a full blown IPv6 implementation, does that also apply
>> to their PWIP driver?
>
>> Interesting issues to sort out for VSI.
>
> Yes, you're right. The issue of whether DECnet V over IP works will come
> up when the new stack work gets going. Nobody here is going to let the
> new stack break things, I'm sure.

I didn't expect anything else.

>
> Of interest to many, I had to get into the DECnet V files yesterday,
> unrelated to this discussion, and I noticed there was an "IPV6" class in
> the CMS libraries.  But it is based on an old release and it appears little
> was done. It was probably cancelled or postponed indefinitely pretty early
> on.

Well, the IPv6 implementation of HP TCP/IP services is quite 
rudimentary, only a Unix style UI, no VMS style UI etc.

RFC2126 was written by DEC engineers, so I'm not surprised that there 
are IPv6 references in the DECnet Phase V sources.

I assume the further development of IPv6 itself was halted at some 
stage, and subsequently the implementation of RFC2126 was also halted.

> So if the time ever comes I have a little something.  But we won't have
> anything that wasn't "checked in" that would have been in the private
> directories of the people working on it.
>
> Right now I know little of OSI or the three RFCs for OSI over IP, the
> PWIP driver or Multinet.  I will be trying Multinet on test systems
> shortly.
>