On Tue, 29 Jul 1997, Rick Copeland wrote:
> Sounds exciting!  Why don't you send the exact proceedure to the list?

I've had semi-simillar luck with the 1.2G kernel image provided by Ragge.
I built the system up with the 1.1A tapes you sent me (thanks again), from
there I got all my hardware quirks worked out to the point where I felt my
system was stable.

I recopied the 1.1A miniroot, booted up the miniroot, and mounted my
existing 1.1A filesystems.  From there I copied over the kernel, boot, and
bootblock images from ftp.luth.se:/pub/NetBSD/vaxport.  I placed the
netbsd and boot files over the older 1.1A ones on my existing filesystems,
and then I disklabeled my boot drive with the new bootblocks.

Next I copied over a set of tar.gz 1.2G binaries I had laying around.  I
removed a lot of the 1.1A directories and their contents as they'd be
replaced with 1.2G files.  (The whole procedure is a mildly challenging
exercise in space managment).  From there I just untarred the 1.2G
binaries into their appropriate locations.

The system works pretty well, but I've found an interesting difficulty in
that if I use nearly any of the standard Unix commands but supply a
non-existant filename as an argument.  Command like `ls -l nofile' causes
the system to panic.  That's where the `semi' in semi-simillar came from.
Ragge assured me that he's fixed that one -- and apparently Robert hasn't
experienced the same problem.  I guess now it's just a matter of me
getting a more recent 1.2G kernel :-)  I'll work on that tonight.

I also have to clean up my /etc directory.  I kept my 1.1A /etc so I've
got sort of a hybrid system.  Probably not a good thing.

BTW, I don't recommend the above procedures if ye are not brave of heart.

-brian.
---------------------------------------------------------------------------
Brian D. Chase         Systems Coordinator        brian.chase@carpediem.com
-- Compression, Inc. - 13765 Alton Pkwy, Suite B - Irvine, CA 92618, USA --