Josh Hope wrote:
> 
> I am writing this letter regretably from the Mac OS, because...well, it's 
> a long story.
> 
> I was going out of dt, so I could login with another username besides 
> root. I hadn't logged out of the system at all ever since I re-installed 
> everything after formatting the drives.
> 
> Anyways, I logout of all dt consoles, then I type exit from my main 
> console...and it just sat there...
> 
> I waited about 10 minutes, nothing happened. I tried ^C, ^Z, 
> etc...(banged the keyboard out of frustration ;))...to no ado!
> 
> So, I hit the restart button...
> 
> I then booted into single user, and ran fsck -p. It showed a few errors, 
> which I was expecting. It fixed them all up and marked the file systems 
> clean. I then attempted a reboot -n...result: Bad system call.

Are you binaries and kernel in sync?  You shouldn't get this message if
they are...

> I sorta 
> paniced there, because I got those a lot when in single user on 
> OpenBSD...without thinking, I tried exiting single user. Then, the 
> console just stayed there. Just like before, nothing happened after 
> exiting out of single user...
> 
> So, I think...maybe my tty's were corrupted or something. I boot into Mac 
> OS and re-build the tty's using Installer (1.0e, might I add)...
> 
> Then I boot again. I pop into single user, do an fsck -p (no file systems 
> were corrupt), then I exit'ed from single user. Immediately after 
> exiting, I got a "Bad system call". I'm not sure what program caused it, 
> since its name was not displayed, but it was immediately after exiting 
> single user...
> 
> The rest of the boot went just as expected. Until it changed security 
> level from 1 to 0 (or is it 0 to 1? I forget)...
> 
> It then just sat there, and started displaying these messages:
> 
> init: getty repeating too quickly on port /dev/ttye0, sleeping
> 
> I then hit the trusty interrupt switch and did a trace just for the heck 
> of it, and these are the results (in case they help someone help me):
> 
> _Debugger(d51ba,b2806,7cfeac,32be,3911) + 6
> _nmihand(3911,915004,6000,400c000,6b5c200) + 26
> _lev7intr(?)
> _vm_page_activate(182e80) + 12
> _vm_fault(6b5c200,400c000,1,0) + bcc
> _trap(8,402a249,400cb36) + 4b4
> _addrerr() + e8
> 
> I've been using the same kernel for months and months and months, without 
> one single hitch at all. Why does it screw up now?
> 
> Anyways, to summarize what I did the past week...
> 
> I was in contact via e-mail with the port maintainer of OpenBSD/mac68k. 
> He informed me that it was realtively safe to install OpenBSD over my 
> existing NetBSD system. I went ahead and downloaded the binaries and 
> kernel. I then installed them...

In and of itself, this shouldn't be a problem....

> Upon booting, after detecting my 3 SCSI devices (on 0 internal HD, on 1 
> SyQuest EZ, on 3 internal CD with nothing in it), it started filling my 
> screen with parity errors...

Don't know about this one.

> Anyways, I then stuck in my NetBSD GENERICSBC #26 kernel and tried to 
> boot OpenBSD with that (stupid? maybe...)

Not quite stupid, but more below...

> That got me to single user, but with every command I made I got "bad 
> system call" errors...

Here's your problem.  Mixing a NetBSD kernel with OpenBSD binaries that
are not of the same vintage causes the above errors.  Basically, I believe
that you shared libc is out of sync (although the static one is, too, so
any statically compiled programs would have the same trouble).
Essentially, some program is making a system call (the stubs are in libc)
which is referenced by a number.  Your kernel doesn't have that number
implemented in it at the moment, so you get that error.  It's quite
possible that this is the reason getty is causing you trouble as well.

> I just ran mkfs on the Mac side on both my partitions (a 140MB /usr 
> partition on my internal drive, and a 100MB / partition on my EZ drive), 
> then re-installed all the NetBSD binaries and my trusty GENERICSBC #26 
> kernel...

I assume that this ended the "bad system call" troubles?

> Now, earlier today, I did install the base.tar.gz package from Steve's 
> binaries. After installing this, I got the thing where telnetting into my 
> system caused the connection to be immediately closed...

Hmmmm....did you install a new kernel as well?  It might be that this is
caused by a similar problem...

> So, now I am left not knowing what I should do. Any suggestions or help 
> would be greatly appreciated...
> 
> And, I'm really hoping I don't have to re-format. I just got finished 
> compiling/installing the many programs I use! Just today! :/
> 
> Thank you all, and an apology for the long letter...

What kind of Mac is this again?  If it's supported by 1.2.1, you might
consider going back to that for the few months before 1.3 is available...

I hope this helps some.

Later.

-- 
Colin Wood                                 cwood@ichips.intel.com
Component Design Engineer - MD6                 Intel Corporation
-----------------------------------------------------------------
I speak only on my own behalf, not for my employer.