I was successfully running netbsd 1.3.2 for almost a year until my
hard drive decided to crap out on me.

I now am trying to bring the DECstation 5000/133 back to live by
netbooting the install kernal so that I can then do an install on a local
disk (I have no idea what PROM is running, how do I determine this?). 

Bootp is provided by the university and I have it set to send

sa=<ip address of the tftp server>:bf=install

I have put the install.ecoff kernel (renamed install) in /tftp and
/tftpboot (just to cover all bases!), and made the permissions readable by
all.

when I issue:
>>boot 3/tftp

I get back in about 10 seconds:
?IO: 3/tftp, msg err (0)

I get the following from tcpdump which seems to suggest that the bootp
stuff is working since the DECstation (narvi.stanford.edu) is contacting
the tftp server (loki.stanford.edu), but I can't determine what is
happening, or not happening, next.  My tftpd line in inetd.conf is
uncommented, and inetd has been restarted.

loki> /usr/pubsw/sbin/tcpdump -vv -e host loki and port 69
tcpdump: listening on hme0
18:57:47.860749 8:0:2b:2b:f0:c5 8:0:20:9b:25:18 ip 60:
narvi.Stanford.EDU.49692 > fastloki.Stanford.EDU.tftp: 16 RRQ "install"
(DF) (ttl 255, id 18)

loki> /usr/pubsw/sbin/tcpdump -vv -e host narvi
tcpdump: listening on hme0
18:57:47.786150 8:0:2b:2b:f0:c5 Broadcast ip 342:
narvi.Stanford.EDU.bootpc > 255.255.255.255.bootps: xid:0x1bc20000
C:narvi.Stanford.EDU [|bootp] (DF) (ttl 255, id 17)
18:57:47.860743 8:0:2b:2b:f0:c5 8:0:20:9b:25:18 ip 60:
narvi.Stanford.EDU.49692 > fastloki.Stanford.EDU.tftp: 16 RRQ "install"
(DF) (ttl 255, id 18)


Does anyone know what "msg err (0)" means.  Any suggestions on what to do
next?

thanks

Nik


Nik Ingle
Dept. Applied Physics
Stanford University 
Stanford, CA 94305-4090
(650) 725-2296