[Coco] Back to my COCO's again

[Gene Heskett]

On Tuesday 11 December 2018 20:09:21 Bill Gunshannon wrote:

> On 12/11/18 5:56 PM, Gene Heskett wrote:
> > On Tuesday 11 December 2018 13:17:52 Bill Gunshannon wrote:
> >> On 12/11/18 11:34 AM, Gene Heskett wrote:
> >> That, combined with a lot of fiddling around and I am up to the
> >> next
> >>
> >> stage.  :-)
> >>
> >>
> >> I have a system built for my SDC that includes the TC3 devices.
> >>
> >> However, whenever I try to access a SCSI disk I get UNIT ERROR.
> >>
> >> Other bits of data.  If I have the disk turned off I get NOT READY.
> >>
> >> That makes sense.
> >>
> >> No matter which SCSI device I type (s0, s1, s2 etc) I see a flash
> >>
> >> of the disk activity light before I get the UNIT ERROR message
> >>
> >> Drive is set for Unit 0.
> >
> > Mmm not 100% sure I've encountered that exact behaviour, so do you
> > have a /dd? And is its LSN0 properly prepared? Both for the size of
> > the drive, the offset where the os9 partition ends and the basic
> > vdisks begin? And where on the drive the os9 file you want to use to
> > finish the boot, actually starts, and its length. See the rbf.defs
> > for which bytes in LSN0 are what, its all in there.
>
> I think there is confusion.  I have a bootable SDC image that includes
>
> SCSI devices.  I need this to prep the SCSI disk I plan to install
> NitrOS9
>
> on.  It will be one large partition with only NitrOS9 on it.  I have
> no
>
> need for basic vdisks.  But I have not even gotten that far.  I am
>
> trying to do a "format /s0" and get the UNIT ERROR message.  And
>
> any other command that attempts to access the SCSI gets the same.
>
> But, as I said earlier, I do see the light on the SCSI disk blink
> which
>
> means NitrOS9 is seeing the SCSI Controller and attempting to pass
>
> a command thru to the disk.

Sounds like, but you said earlier it didn't make any diff which disk 
descriptor, and that bothers me because on a scsi systen only the drive 
being addressed should blink. You said it was drive 0 on the drives 
jumpers so if the descriptor is properly configured, then only /s0 
should blink the light.  You can check that with dmode /s0, but have a 
copy of the superdesc.asm in hand because there are more bits in use for 
a hard drive than for a floppy. /dd and /s0 should also be copies with 
only a name change. However where you are booting from an image on the 
SDC  /dd will likely point at the SDC and thats dirt I haven't walked  
on yet, being up to my butt in an out of commission milling machine, and 
likely another few weeks bringing it back to life as I am rebuilding the 
interface from scratch to give me another 2 dozen i/o lines to better 
control it.

If you can copy the output of dmode /s0 to the list, that might have some 
clues/hints.  Maybe, following the trail my ancient wet ram wants to 
follow ATM. Somebody else might well correct/argue with me, and the more 
eyeballs looking at it, the better we'll be. You may want to see if the 
disk surface has data with a "dump /s0@" which should dump whats there 
starting with the drives LSN0. The trained eyeball can look at the first 
256 bytes and determine if its ever been formatted for os9. Seeing that 
might also be helpfull. So would a list of the modules in the os9boot 
file you are booting from. The stock ident would help, but we'll get 
more info from my "vfy os9boot" if you've a copy of it handy.

Take care Bill

[...]

-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>