[SunRescue] NetBSD 1.4 miniroot problem fixed

[jwbirdsa at carfallin.picarefy.com]

>> ..... (I have an old SUN0207 which I'm using
>> as my magic install disk.)
>
>Great idea.... I keep a spare drive set up for my VAXen, but had not
>thought about that for the suntoyz.  I should probably do that.  I
>have a spare applecrate that I could throw one of those 200mb things
>in and use that as a portable booter.  How have you got yours ID'd
>and set up?  Is it a live bootable system, or just a miniroot on swap?
 
   Actually, I have several different miniroots on it, and a filesystem
which contains all the install sets for sun3 and sparc.

>This was what I was wondering about the clock speed for the CPU.
>The sun3's run at 12/16/25(?) mhz, and the 4/260's are at around
>25/30 somethihng?

   3/50		15.7MHz or thereabouts
   3/75,3/1xx	16.67MHz
   3/60,3/80	20MHz
   3/2xx	25MHz
   3/4xx	33MHz

   4/2xx	16.67MHz (!)

>What if you half clock the cpu?   Will that
>allow enough time for the scsi to settle?  If that is the case,
>does the code need a wait loop for a couple of instructions to
>allow the scsi to catch up with the cpu?  Nothing fancy, but just
>a millisecond or two. 

   Don't know. To be honest, I've never had very good luck with Sun-3
SCSI controllers, even under SunOS. I'd replaced them all with Sun-2
units, which might not have been quite as fast but actually worked
reliably. I wish somebody would write a driver for them.

>That scsi thing is indeed a black art.....(:+{{.... and those
>sun vme crates must require Merlin's calibre of black art.

   Yeah. I just spent an afternoon with some unpleasant termination
issues. The thing seems to work OK (knock on wood) with just one drive
on it and the drive using its internal terminator. When I tried chaining
a ZIP drive on after it, using the ZIP's internal terminator, I got a
number of SCSI timeouts. The machine almost always managed to recover,
but sometimes it would sit and twiddle its thumbs for literally five
minutes in the process! Unfortunately, it munched the OS one of the
times that it didn't recover and won't boot multiuser anymore (csh
coredumps!), but in single-user mode I've been able to do enough testing
that I think the SCSI problem is over. (knock on wood some more)
   (Incidentally, NetBSD doesn't seem to handle removable media drives
very well. It recognizes that the drive is removable, and seems to sync
the drive OK when umounting the filesystem, but once you remove the
disk, you'll be plagued by occasional "cache synchronization failed"
messages until the next time you reboot.)
   My current problem with the 3/4xx is "ie0: DMA underrun" messages,
which occur very occasionally when transferring enormous files via FTP.
So far, these seem to simply be annoyances; the data appears to be
intact.

>I can never seem to get two of the DB50 cables on-line in series on
>the scsi bus.  One is fine, but the second one blows up.

   I once got three in a row working -- three shoeboxes hanging off a
3/60. That was the system that wouldn't work if it was terminated, and
it insisted that the first two cables had to be short and the last one
long. Any other arrangement, including making all three cables short
ones, was error city.

   Random rant: why is it that mass storage busses all have so many
problems? MFM, RLL, and ESDI worked pretty darn well but the two-disk
limit was a bummer. SCSI allows for plenty of expansion, but it's pretty
much plug-and-pray. IDE is still limited to two drives and has problems
with devices that don't get along, although these days with twin
controllers the norm you can usually separate devices, and when you can
get a 40-gigabyte disk for less than $300 it doesn't matter so much if
you can't have very many.
   But why hasn't anybody come up with anything better?

   --James B.