If you have another Unix box with X installed, you can run X software on
your headless Mac quite easily.  All you need to do, is telnet to your
headless Mac and set the display to your Unix box running X.

Example:
In csh, type:

setenv DISPLAY ipaddress:0.0

In sh or bash, type:

DISPLAY=3Dipaddress:0.0
export DISPLAY

You can also use an app like MI/X on a Mac OS machine.  Same thing applies.

Of course, ipaddress should be replaced with the ip address of the machine
that you want your headless box to display on.  From there, just type
something like:  xterm

And an xterm window will appear on your remote machine.

This will make configuring your headless Mac much easier and much more fun.
Plus, you can run your machine from a much less powerful terminal, and stil=
l
take full advantage of your headless machine.


> From: D=F6m=F6t=F6r Guly=E1s <dog@dog4.dyndns.org>
> Date: Sun, 20 Nov 2005 21:43:22 +0100
> To: Michael <macallan18@earthlink.net>
> Cc: port-macppc@NetBSD.org
> Subject: Re: Unofficial macppc 2.1 ISO for old world machines
>=20
> Yay, I got the installer running, at last, and was able to install
> netbsd and boot from HD.
>=20
> So far so good. Though I can neither use the graphics card nor the
> keyboard (graphics card ignored by OF, keyboard hangs netbsd), it
> seems like netbsd is running alright. And this is how it went:
> - unplugged mobo battery and mains, let it sit it out
> - tried booting from unofficial iso, no go
> - tried booting from official iso, no go, of course, but it might
> have done something to OF
> - tried booting from unofficial iso again, just for kicks, and
> voi'la, it magically works.
>=20
> I can live with the fact that the computer runs headless once its
> configured, though having to use vi for the first steps makes things
> a lot harder. Also, having to use zTerm under Classic over a USB
> serial adapter is making things a lot harder. Flaky as hell.
>=20
> Alright, back to work...thanks for the help so far, you'll hear from
> me :)
>=20
> Cheers,
> - DoG
>=20
>=20
> On 20.11.2005, at 21:03, Michael wrote:
>=20
>> Hello,
>>=20
>>> Alright, I got a little further this time, then went back :/
>>>=20
>>> Using the commands suggested, namely
>>> boot scsi-int/sd@3:,\OFWBOOT.XCF NETBSD.MACPPC
>>> got the whole thing running. Not at the first try, but I tried
>>> mocking around, and hell if I know what I did, but after several
>>> repetions and mucking with the real-base and load-base settings, it
>>> worked.
>>=20
>> That happens to everyone. Welcome to weird Apple OF.
>>=20
>>> I got into the installer several times, but it gave me timeouts on
>>> the
>>> MESH controller, and locked up while trying to format the HD. So I
>>> went and changed the SCSI chain, plugged things into the external
>>> SCSI
>>> port, and I was stuck again. Unfortunately, I didn't write down the
>>> exact output when things worked, and also failed to put down the OF
>>> variables.
>>=20
>> Hmm, what kind of harddisk do you use? Some IBM by any chance?
>> If so - does it have a jumber labeled 'target initiated sync
>> negotiation' or something like that? If it does - disable this feature
>> since it confuses the mesh controller. Do you have a ZIP drive? It may
>> mess up things too, if you have one disconnect it and see if things
>> change.
>>=20
>>> In the meantime I managed to mess up the real-base var and had to
>>> take out the mobo battery several times. Now, it seems like more ore
>>> less independent of the real-base value, I get to one of the
>>> following messages:
>>>=20
>>>> 0 > boot scsi-int/sd@3:,\OFWBOOT.XCF NETBSD.MACPPC loading XCOFF
>>>> tsize=3DC860 dsize=3D28C bsize=3D25C0 entry=3DE00000
>>>> SECTIONS:
>>>> .text    00E00000 00E00000 0000C860 000000E0
>>>> .data    00E0D000 00E0D000 0000028C 0000C940
>>>> .bss     00E0D290 00E0D290 000025C0 00000000
>>>> loading .text, done..
>>>> loading .data, done..
>>>> clearing .bss, done..
>>>>=20
>>>>>> NetBSD/macppc OpenFirmware Boot, Revision 1.10
>>>>>> (builds@works.netbsd.org, Mon Oct 24 22:31:34 UTC 2005)
>>>> no active package4125672-
>>=20
>> Check your SCSI chain, especially termination. Try different cables,
>> different connectors, remove all unnecessary devices and so on.
>> Mesh is
>> PICKY. OF is PICKY. Our driver isn't as fault tolerant as it probably
>> should be either, but this looks a lot like imperfect cabling.
>> You may also want to check your memory modules - MacOS is a lot more
>> tolerant here ( probably because of their weird memory manager )
>>=20
>> have fun
>> Michael
>=20