Hello all,

I'm new to this list and to NetBSD, please bear with me.

I'm trying to netboot a sparc ELC. The problem seems to be
that the secondary loader (boot.net from 1.5) 
fails to connect to a bootparam server resulting in 
'bootparamd failed whoami'. Needless to say, booting doesn't
proceed beyond this.

(someone was having a similar problem a while ago on this list,
was that resolved ? )

Bootparamd is running on a i386 Debian box. Tcpdump shows the
following when the loader tries to call bootparamd, i.e. after
getting boot.net by tftp.


01:17:42.082249 wis06.1023 > 255.255.255.255.sunrpc:  udp 96 (ttl 4, id 0,
len 124)
0x0000   4500 007c 0000 0000 0411 ba4f 851c 7706        E..|.......O..w.
0x0010   ffff ffff 03ff 006f 0068 efdc 0000 0001        .......o.h......
0x0020   0000 0000 0000 0002 0001 86a0 0000 0002        ................
0x0030   0000 0005 0000 0001 0000 0014 0000 0000        ................
0x0040   0000 0000 0000 0000 0000 0000 0000 0000        ................
0x0050   0000                                           ..
...
...

Now, why is the request going to 255.255.255.255 , and 
is 96 an error code ? I'm at loss here.

Rarp seems to be working fine. From the debug log:

/usr/sbin/in.rarpd:[5]  RARP_REQUEST for 8:0:20:3:b4:a5
/usr/sbin/in.rarpd:[5]  trying unit -1 netnum 133.x.x.0 mask ffffff00
/usr/sbin/in.rarpd:[5]  good lookup, maps to 133.x.x.6
/usr/sbin/in.rarpd:[5]  immediate reply sent

Portmapper seems also to be working on the server:

pyxis23:/# rpcinfo -p
program vers proto   port
100000    2   tcp    111  rpcbind
100000    2   udp    111  rpcbind
100003    2   udp   2049  nfs
100003    2   tcp   2049  nfs
100005    1   udp    877  mountd
100005    2   udp    877  mountd
100005    1   tcp    880  mountd
100005    2   tcp    880  mountd
100026    1   udp    672  bootparam

And broadcasting for bootparam servers from another machine
	
othermachine:/usr/local/home/ragga# rpcinfo -b bootparam 1
133.x.x.2 machine1.ac.jp
133.x.x.1 machine2.ac.jp
133.x.x.23 myserver.ac.jp

Is the fact that there are 2 other machines responding harmful in
any way ? AFAIK they know nothing about the ELC.

I've tried the loaders from versions 1.3.3 through 1.5 with exactly
the same result. 
It therefore seems I'm doing something fundamentally wrong but
cannot understand what.
Any ideas ?

TIA,

--
Ragga