Booting the MicroVAX II

Initial Bootup
Top \| Vax	Contact

The first goal of booting the VAX was to get a console up. Getting it to talk to a terminal (an old Tosheba laptop) on the consol serial port was not that difficult, indeed, it worked first time! The DB9 plug on the back isn't the same as a pc serial port, and a conversion cable is needed (instructions about half way down that page).

Once I had a terminal up, I couln't get it to boot at all... I gave up on the disk drives and tried to boot from the network card. Digital designed a protocol, MOP, to allow network booting and other functions. A mopd server is available for various platforms, and this allows you to upload a bootstrap. I followed the somewhat out of date The VAX Network Booting HOWTO by Brain Chase. The howto is quite specific in saying "non lance" ethernet cards are not supported by the bootstrap, and mentions that the DELQA card is not supported. However I have a DELQA-M card, and Hamster mentions that this card is lance based. Since the HOWTO is 3 years old, I tried anyway... my first problem was simple.

We couldn't get the AUI tranciever to do anything, no lights, nothing.

This was a clue. We pulled the ethernet card out, and found it wasn't pushed into the slot. This explained why we couldn't boot from disk either. The QBUS works somewhat like token ring, where the CPU at the top sends a "token" by aserting a signal, cards that do not require access to the bus pass the token on by asserting another signal (wired to the next card through the backplane). Cards that require access to the bus do not pass on the token and use the bus. If a card in the middle is missing, then the token can't be passed on. This is the function of the "Grant Continuity" cards. With the ethernet not fully plugged in, nothing below that card (everything basically) functioned.

After we pushed the network card down fully, we got lights on the AUI connetor and we could also boot from the big disk. This disk has VMS on it, and after about a minute of chugging on it, it failed and dropped back to the chevron (VAX speak for the preboot console). It would appear that the big disk is working, but not fully.

I jumped through the hoops of setting up mopd, rarp and bootparamd on amislave, dodged crazy NFS problems when I unplugged the debian box in the same room as bigvax so I could use it's network connection. Setting the gateway field in bootparams seemed to help, amislave has 3 network cards, and bootparams wasn't passing the correct gateway address to bigvax.

Eventually I got it to pull the kernel down via nfs and boot it:

KA630-A.V1.3

Performing normal system tests.

7..6..5..4..3..

Tests completed.

>>> b/100 xqa0

2..

Bootfile: mopboot

1..0..

>> NetBSD/vax boot [Nov 17 2000 07:23:42] <<

>> Press any key to abort autoboot

Trying BOOTP
bootp: no reply
Trying BOOTPARAMS
boot: client IP address: 192.168.1.3
boot: client name: bigvax
root addr=192.168.1.1 path=/home/vax/root
open netbsd.vax: No such file or directory
> boot netbsd
1160552+56892+194872+84828+99876]=0x185e60
[ preserving 184704 bytes of netbsd a.out symbol table ]

Copyright (c) 1996, 1997, 1998, 1999, 2000
The NetBSD Foundation, Inc. All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.

NetBSD 1.5 (GENERIC) #1: Sun Nov 19 15:44:10 PST 2000

matt@cheetah.local:/u1/kobj/GENERIC

MicroVAX II

total memory = 16372 KB
avail memory = 12448 KB
using 230 buffers containing 920 KB of memory
mainbus0 (root)
ibus0 at mainbus0
uba0 at ibus0: Q22
dz0 at uba0 csr 160100 vec 304 ipl 17

It hangs here.

I didn't know what to do. I decided (after mucking around with different kernel versions) that pulling all the nonessential cards would be a good idea. This resulted in a complete boot and the wonders of configuring with only crazy editors, ed an vi are not my cup of tea! I did most of the configuring using emacs on amislave where the files are local.

Currently I'm compiling bash, and pico, and hopefully I can work on getting the disks going in the morning, when the compilation has finished.... maybe it won't be finished by then, who knows...

Initial Bootup