Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rochester!pt.cs.cmu.edu!sei!sei.cmu.edu!pdb
From: pdb@sei.cmu.edu (Patrick Barron)
Newsgroups: comp.os.vms
Subject: Re: (TCP/IP + DECnet) / (1 * DEQNA) = works!
Message-ID: <1457@aw.sei.cmu.edu>
Date: Fri, 29-May-87 15:48:00 EDT
Article-I.D.: aw.1457
Posted: Fri May 29 15:48:00 1987
Date-Received: Sat, 6-Jun-87 15:47:56 EDT
References: <8705290339.AA09048@seismo.CSS.GOV>
Sender: netnews@sei.cmu.edu
Reply-To: pdb@ci.sei.cmu.edu.UUCP (PUT YOUR NAME HERE)
Distribution: world
Organization: Carnegie-Mellon University, SEI, Pgh, Pa
Lines: 15

In article <8705290339.AA09048@seismo.CSS.GOV> SYNFUL@DRYCAS.CLUB.CC.CMU.EDU (Marc Shannon) writes:
>The only trick we had was that DECnet must be started before TCP/IP...It
>only wants to speak to certain network devices, I guess.  (Sounds like the
>making of a computer soap opera!)

There's a real good reason for that:  When DECnet starts up, it goes out and
changes the DEQNA's Ethernet address.  If anything starts to talk to TCP
between the time you start TCP and the time you start DECnet (after TCP
has already responded to an ARP or two), then everyone who was talking to
you now has the wrong Ethernet address, and you lose.

For similar reasons, DECnet-Ultrix needs to be started before TCP under
Ultrix.

--Pat.