[CUWiN-Dev] Setting Up UML Image

[Quantum Scientific]

On Friday 18 March 2005 18:55, David Young wrote:
> I've spent way more than one minute, and we are no nearer a solution to
> your problem than you were a few days ago, because you hasten to take
> offense at my questions, instead of answering

The answer doesn't matter to me at this point, but obviously I was trying to 
build the system inside a container file.  There are probably three commands 
to create the file and format it with UFS-44bsd, so it can be operated on 
virtualized in any OS.  There are a few more commands to set up a chroot 
environment in that file and build NetBSD, depending on the host OS.  This 
would be the platform that others can work in to contribute to the project, 
if one chooses to -not- make NetBSD their life, and this is what I was 
working toward, as you know.

I suggest it would be helpful to others if instructions for this were 
published (don't ask me), so volunteers wouldn't waste days struggling with 
the idiosyncrasies of one-time NetBSD preparation, and can concentrate 
instead on getting used to a -running- NetBSD and contributing.  The BSD 
barrier is high enough as it is!  I've heard from others in the past day or 
so who have struggled against this very inaccessibility problem, like me.  It 
remains a mystery how -you- build the development image.  There can be none 
of this high-flautin' uber-guru affectation in a cooperative project, or 
people will just quietly go away.  Unfortunately, I'm not so quiet.  

Just one command was all I needed (mounting loopback) , and it cost me a full 
day with no success.  I had tried all permutations of vn0, vnd0, etc (which 
names don't make sense anyway for a lo device... what do we actually call it?  
A loopback device, or a virtual device?), and they all failed.  Maybe it had 
something to do with the Live CD, I don't know...  but I wasn't about to 
install NetBSD on one of my systems just to prepare an image I was going to 
UML in Debian.

In my opinion now, BSD is a non-starter.  It is just too primitive, and there 
is absolutely no advantage.  It is a mistake.  I'm just sayin' it, being a 
Texan.


> My question was, what is this about UML?  Maybe I can help you with that.

I never got to the point of UML, and that is not a problem anyway.  I needed 
the NetBSD image, to run in a UML environment.

Carl Cook