Excerpts from netbsd: 11-Apr-97 Re: -current on (diskless) .. "Chris G.
Demetriou"@cs. (1807)

> > This didn't work for me with your distributed "netboot'
> > because the TFTP server is not the NFS server.
> 
> Does it require that the TFTP server be the same as the NFS server?  I
> didn't think that our libsa had the requirement, but i'm not
> completely familiar with the libsa code.

The problem is that the bootp/tftp server is used
as NFS server too. There is nothing like a "root server"
tag in bootp. Perhaps one of the existing tags (time server,
name server, domain server, log server, cookie server,
lpr server, impress server, rlp server, swap server, nis server,
ntp server) should be used for this.
(I have the same problem with the i386 netboot too.)

[PCI probe problem]

> Interesting.  How old is your system's firmware?

It's not too old - it's V6.3 (Nov 20 1996).
As I just checked, there is a V6.4 on gatekeeper, but I don't
want to upgrade now.

> 	If so, note the cases in cia_pci.c to handle lossage on the
> 	EB164, and modify them so they apply to the AlphaStation
> 	[56]00 (kn20aa) as well.

That helpes a bit: After a hard reset the probe works fine
until this multi io card (PPB + Tulip + 2xSCSI) is hit. The
3 devices (number 0/1/2 on pci1) are found; while probing
the remaining device numbers on pci1 and then pci0 the
messages appear. After a reboot the messages appear from
the beginning. So it's only a little improvement...

After all, the machine is running quite well.
There are some strange effects from time to time, like
unexpected signals and file truncation.
At the moment, if I log in via rsh, I get parts of /etc/ttys,
/etc/master.passwd and some binary stuff dumped on my
terminal. Via telnet or on the console I get /etc/motd as
expected...

best regards
Matthias Drochner