Nathan Moore said...

|[nmoore@honker ~]$ host honker
|honker.workstations.winona.edu has address 199.17.152.193
|
|[root@diepholz ~]# host diepholz
|Host diepholz not found: 3(NXDOMAIN)

Host talks to an NFS server.  You have
not changed your DNS to know about diepholz .

|for diepholz, the ifconfig output follows,
|
|[root@diepholz ~]# ifconfig
|eth0      Link encap:Ethernet  HWaddr 00:50:8D:C1:E3:05
|           inet addr:199.17.152.97  Bcast:199.17.152.255  Mask: 255.255.255.0
...
|I've also posted the /etc/hosts file below,
|
|[root@diepholz ~]# cat /etc/hosts
|# Do not remove the following line, or various programs
|# that require network functionality will fail.
|127.0.0.1       diepholz        localhost.localdomain   localhost

This should really be:

127.0.0.1	localhost.localdomain   localhost
199.17.152.97	diepholz

|So, I understand that the inter-cluster ssh works because of the hand- 
|curated /etc/hosts file quoted above, but I don't understand why the  
|two non-working machines are not broadcasting their IP/hostname pair.

I'm not awar eof how they would broadcast it unless
you actually have each system set up as a DNS server
for its own name as a domain, which would be a bit
strange,  You need to update your DNS server, unless
you're using dynamic DNS, which I haven't really
messed with.