TCP chokes between Ultrix 4 and 3.1c systems

[Len Lattanzi]

In article <1990Oct9.040440.23066 at watcgl.waterloo.edu> idallen at watcgl.waterloo.edu (Ian! D. Allen [CGL]) writes:
:>Yet another amusing feature.  I think this may have come up quite some time
:>back.  When I use ftp, telnet, or even mail to a DS3100 running Ultrix 3.1c
:>from my DS5000 running Ultrix 4.0, the connection hangs whenever 512 or more
:>bytes are sent from a single operation (e.g., mail messages, ftp files,
:>or commands in telnet).  The same thing happens when the folks on the other
:>end try to talk to me.  Neither of us have any problems talking to non-Ultrix
:>systems, and I can talk to other Ultrix 4.0 systems just fine.
:
:Aha!  We see this behaviour between real MIPS machines and our Ultrix
:3.1C DECsystem 5400's.  BSD Vaxen, Suns, other machines, have no
:problem talking to the 5400's, but connecitons from those three MIPS
:machines hang on large packets.  And it's asymmetric -- we can rcp *out*
:of the DS5400, but not in.
:-- 
:-IAN! (Ian! D. Allen) idallen at watcgl.uwaterloo.ca idallen at watcgl.waterloo.edu
: [129.97.128.64]  Computer Graphics Lab/University of Waterloo/Ontario/Canada


The first M/120's did not support trailers on enet packets and silently
hung connections to other hosts that sent them. (just a guess but try
turning off trailers on all interfaces -- I have no idea if Ultrix
3.1/4.0 enet hangs have similar causes) 
-- 
\
 Len Lattanzi	({ames,pyramid,decwrl}!mips!len) <len at mips.com>
 I would have put a disclaimer here but I already posted the article.