[4068] in linux-net channel archive
Re: repeat of previous querry
daemon@ATHENA.MIT.EDU (Paul Levin)
Fri Aug 16 23:13:43 1996
To: Dave Platt <dplatt@iq.tvsoft.com>
cc: linux-net@vger.rutgers.edu
In-reply-to: Your message of "Fri, 16 Aug 1996 14:00:45 PDT."
<199608162100.OAA01711@iq.TVSoft.com>
Date: Fri, 16 Aug 1996 18:20:13 -0400
From: Paul Levin <Paul_Levin@postbox.ius.cs.cmu.edu>
Dave...
Thanks for your suggestion. I don't see any redirect
messages in my syslog. However, a local network guy
suggested that the reason that the machines can't
talk to each other when isolated is that there is
no local hardware address resolution being done
(that's normally handled by the network). I'll have
to create an /etc/ethers file and then do an
/sbin/arp -f to load it to get this to work.
I've gotten a binary of tcpdump working on my lb1
machine and it appears there are lots of dropped
frames when it communicates w/ lb2 (if I interpret
the output correctly). This again is consistent
with the UDP bad checksum and socket destroy delayed errors
I get. Have you heard of a case where a machine
does not calculate the checksums correctly?
Thanks again for your help.
Best Regards
PDLevin The Robotics Institute, CMU
