From owner-freebsd-questions  Tue May 27 06:54:43 1997
Return-Path: <owner-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id GAA14481
          for questions-outgoing; Tue, 27 May 1997 06:54:43 -0700 (PDT)
Received: from insl1.etec.uni-karlsruhe.de (uucp@insl1.etec.uni-karlsruhe.de [129.13.109.178])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA14476
          for <freebsd-questions@freebsd.org>; Tue, 27 May 1997 06:54:41 -0700 (PDT)
Received: (from uucp@localhost) by insl1.etec.uni-karlsruhe.de (8.8.5/8.8.2) with UUCP id PAA04662 for freebsd-questions@freebsd.org; Tue, 27 May 1997 15:54:07 +0200 (CEST)
Received: (from erb@localhost) by insl2.etec.uni-karlsruhe.de (8.8.5/8.8.2) id PAA13242 for freebsd-questions@freebsd.org; Tue, 27 May 1997 15:53:12 +0200 (CEST)
Message-Id: <199705271353.PAA13242@insl2.etec.uni-karlsruhe.de>
Subject: 2.2.2 diskless problem
To: freebsd-questions@freebsd.org
Date: Tue, 27 May 1997 15:53:12 +0200 (CEST)
Reply-To: erb@inss1.etec.uni-karlsruhe.de
From: erb@inss1.etec.uni-karlsruhe.de (Olaf Erb)
X-Mailer: ELM [version 2.4ME+ PL31 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hi,

I've been setting up a diskless client with netboot, and all works
very will until the kernel wants to execute /sbin/init.

Here's a tcpdump, dummy1 is the server, dummy2 the client:

dummy2.f84f2 > dummy1.nfs: 120 read fh 0,132111/497 914 bytes @ 919984
dummy1.nfs > dummy2.f84f2: reply ok 1016 read

<now the kernel has been loaded>

Console messages appear, then:

NFS SWAP:192.168.0.1:/usr/root/swap
NFS ROOT:192.168.0.1:/usr/root/dummy2

arp who-has dummy2 tell dummy2
arp who-has dummy1 tell dummy2
arp reply dummy1 is-at 8:0:2b:be:14:6d
dummy2.afe10001 > dummy1.nfs: 100 lookup fh 0,132111/495 "sbin"
dummy1.nfs > dummy2.afe10001: reply ok 128 lookup fh 0,132111/68316
dummy2.afe10002 > dummy1.nfs: 100 lookup fh 0,132111/68316 "init"
dummy1.nfs > dummy2.afe10002: reply ok 128 lookup fh 0,132111/621
dummy2.afe10003 > dummy1.nfs: 92 getattr fh 0,132111/621
dummy1.nfs > dummy2.afe10003: reply ok 96 getattr REG 100500 ids 3/7 sz 196608 
dummy2.afe10004 > dummy1.nfs: 104 read fh 0,132111/621 8192 bytes @ 0
dummy1.nfs > dummy2.afe10004: reply ok 1472 read (frag 26890:1480@0+)
dummy1 > dummy2: (frag 26890:1480@1480+)
dummy1 > dummy2: (frag 26890:1480@2960+)
dummy1 > dummy2: (frag 26890:900@7400)
dummy1.1266 > dummy2.globe: udp 114
dummy2 > dummy1: icmp: dummy2 udp port globe unreachable
dummy2.afe10004 > dummy1.nfs: 104 read fh 0,132111/621 8192 bytes @ 0
dummy1.nfs > dummy2.afe10004: reply ok 1472 read (frag 26892:1480@0+)
dummy1 > dummy2: (frag 26892:1480@1480+)
dummy1 > dummy2: (frag 26892:1480@2960+)

this continues forever (read fh, reply..). 
Btw, what is this 'globe' stuff the server sends to the client?

What's going on here?
FYI, server has a de203, client a ne1000 ethernet card.

Thanks,
Olaf
-- 
Argue your limitations, and sure enough, they're yours.
                                                -- Richard Bach, Illusions