Hello all,


I am working on getting a NetBSD cluster with 64 nodes up and running.
Currently they are setup to boot from floppys with GRUB on them. I am
putting new NICs in so that they can PXE boot . They nodes are
diskless and I have a head which assigns fixed addresses using dhcp
and a NFS running on a separate machine (separate from the dhcp server
- aka head).


I have numerous questions as I am just learning how all of this works.
As usual the netbsd.org documentation is very helpful so I am figuring
out a lot as I go along.

My question is a fairly pedestrian one. in my /var/log/messages I
notice that the switch is requesting an ip address from the dhcp
server. The message (repeated many times) looks like this:

Jun 30 13:00:00 frodo dhcpd: DHCPDISCOVER from 00:04:CC:CC:CC:CC via
ex1: network 192.168.1.2/24: no address pool

ex0 is connected to the intranet and the outside world
ex1 is connected to the switch for the cluster

My initial guess is that i dont have something in /etc/rc.conf. Or
/etc/dhcpd.conf. Should I have an entry which identifies the switch
MAC using the gateway address? Any help is greatly appreciated.

Thank you,
-- 
Michael Bubb
optPart, LLC - Hoboken, NJ
cel 201.736.0870
fax 201.377.1717