Well, my experience doesn't quite cover the older alpha's, so I'll
have to ask for some assistance from someone else with one.  The
problems that I was having with the multia was seen in receiving
packets, not in sending them.  There are basically two possibilities,
that I made a mistake somewhere (pretty high probability) or that
there is another problem.

So, I'll ask a couple of questions of the list in general,
	o Has anyone with an older alpha got netbooting
	  to work with the boot blocks that I provide?
	o Have there been any time periods recently (last
	  couple of months) that netbooting has been broken,
	  since the boot blocks that you get from my system
	  are relatively recent (no idents, though...)?
	o Are there any known issues with netbooting the
	  3000/300?

And could you provide me with a tcpdump from the bootp server,
since that would help narrow down the problem.  Also, the version
of NetBSD that you are trying to boot might be helpful, and what
the console says before failing.  (Using a serial console makes
this a _lot_ easier.)

On 917438605 seconds since the Beginning of the UNIX epoch
Carey Stevens wrote:
>
>"R. C. Dowdeswell" wrote:
>
>> Sure, the problem is probably that the netboot from that page is
>> specifically for multia/newer alpha SRM consoles.  Try my slightly
>> updated and therefore cooler page:
>> http://www.imrryr.org/NetBSD/alpha/netboot/
>> which asks an additional question "Broken Firmware [] Yes [] No"
>> Try Yes for that, and unless I have the wrong boot image lying around
>> that should work a lot better.
>
>Thanks, but I have tried both without any success. The netboot still
>does not seem to send another bootp request.
>
>The following is output from bootpd showing that no request is made from
>the second stage bootstrap program (called netboot.new.no for now) after
>the first console request
>
>esp23 is my bootp server (linux box)
>[root@esp23 bootp-2.4.3]# bootpd -d 4
>bootpd: info(6):   bootptab mtime: Thu Jan 28 00:29:04 1999
>bootpd: info(6):   reading "/etc/bootptab"
>bootpd: info(6):   read 1 entries (1 hosts) from "/etc/bootptab"
>bootpd: info(6):   recvd pkt from IP addr 0.0.0.0
>bootpd: info(6):   bootptab mtime: Thu Jan 28 00:29:04 1999
>bootpd: info(6):   request from Ethernet address 08:00:2B:3D:80:CB
>bootpd: info(6):   found 130.216.5.29 (esv29)
>bootpd: info(6):
>bootfile="/usr/localpeople/netbsd/boot/netboot.new.no"
>bootpd: info(6):   vendor magic field is 0.0.0.0
>bootpd: info(6):   sending reply (with no options)
>bootpd: info(6):   setarp 130.216.5.29 - 08:00:2B:3D:80:CB
>
>esu34 is my tftp/nfs server (SGI 02) system log
>Jan 28 00:38:32 6D:esu34 tftpd[4651]: esv29.auckland.ac.nz: read request
>for /usr/localpeople/netbsd/boot/netboot.new.no: success
>
>>
>> (If you are interested in the problem, see PR#6446).
>
>What does see PR#6446 mean? I am interested if it will help me solve my
>problem. Ideally I just want it work though. :-)

PR#6446 is a Problem Report in the GNATS bug tracking database.
With the information that you are giving me however, I think that
this bug is not affecting you.

>
>
>
>--
>---------------------------------------------------------
>Carey Stevens
>Department of Engineering Science, University of Auckland
>Private Bag 92019, Auckland, NEW ZEALAND.
>Tel: +64 9 3737599 x7490
>Fax: +64 9 3737468
>---------------------------------------------------------
>
>

>

 == Roland Dowdeswell
 == http://www.imrryr.org/~elric/