networks@dartvax.UUCP (Special Projects Group) (10/31/85)
Many thanks to all those that replied to my problems with UDP and
with talk (which uses UDP.) I started to reply in person to everyone
but was being flooded, so ...
For anyone else interested, there were two problems. The first is
that I trusted 'talk' to be correct and it's not. It does not
take into account the machine dependencies. The VAXen stores
bytes in one order and the SUN in the other so while it is a valid
address on one end, it's reversed on the other. A htons(port)
fixes that nicely. (The politest comment I heard about talk was
that it was brain-dead. The comments got a lot worse...)
The second problem is that the UDP that came with earlier releases
of 4.2BSD aparently did not checksum UDP correctly while it seems
that SUN fixed this. The later releases of 4.2 are rumored to have
this fixed and fixes were posted to net.unix-wizards sometime last
year, I believe.
So, again, thanks for all the help.
--
David C. Kovar -- Special Projects Group
USNET: {linus|decvax|cornell|astrovax}!dartvax!networks
ARPA: networks%dartmouth@csnet-relay
CSNET: networks@dartmouth
US Mail: Kiewit Computation Center
Dartmouth College
Hanover NH
03755
Phone: (603) 646-3144