[Coco] nitros9.org

Wed May 17 14:53:05 EDT 2006

Mark,

I think you've misunderstood my problem. The script worked
and I ended up with a floppy that had a kernel, sysgo and
a handfull of commands. Unfortunatly when I tried to boot from
that floppy the boot failed.

I'm installing the 6309 version of nitros9 on a new CF card. This
way there won't be any interference between versions. As luck would
have it I was pondering the advice from Robert G. and your Email
arrived as I was building a bootdisk that actually worked!!! It turns
out that there are some module between level 2 and level 3 of Nitros.
In particular the changes were in with the naming of the graphics modules!!
When I saw that windint.io was no longer distributed with L3 I simply
commented out. It took a very close look at "standard.bl" to catch the
change. Somehow I managed to catch cc3io/vtio change. I also managed
to corrupt the modules directory on the CF card so I reformatted the card
and reinstalled all the software again.

Unfortunatly the CF card that I got yesterday is showing the same issue as
the first card I got from you. When I write to one of the hdb "disks" it's
all lost when I reset or power cycle the machine. The card works fine with
nitros though. I ordered another card online yesterday so I'll have to wait
until it comes in before I can make the new system bootable without a floppy. :-(
On the positive side I can back the card up via dd on my linux box and when
the new card comes in use dd to set up the new card!!

Thanks everyone for your help. If I upset anyone with my fairly negative attitude
please understand that I was up until 2am working on the problem and spent better
then 9hrs trying to build a bootable floppy and was more then a little frustrated.

Frank

On Wed, May 17, 2006 at 11:49:04AM -0500, Mark Marlette wrote:
> Frank,
> 
> Use the Source Luke....... :)
> 
> A few basic things. A script is nothing more than an automated process 
> of commands you could enter in manually.
> 
> With that said, look at the script. OS9gen will complain about a 
> missing module. It might scroll right on by you. Use the command that 
> is contained in the script and find where it is broken. Not sure if you 
> copied the script of if you are modifing the one that was shipped with 
> you system, that was used to create your system.
> 
> Hopefully you are using a copy in a totally different version 
> directory. On my system I have every version of Nitros ever released 
> with a script to build the boot disk. A pain when you changes releases 
> but when you do a remake, a pleasure.
> 
> Always, always, always, make sure the floppy boot works first before 
> you try to do a hard or HDB boot creation.
> 
> Hopefully you have floppy boot disk made so that if you do mess the 
> CF's auto boot from HDB, then you can still get to the CF via the 
> floppy boot.
> 
> Don't give up. It is a simple process, study the script, understand it, 
> try the commands manually if you want to REALLY understand them.
> 
> Mark
> 
> Quoting Frank Pittel <fwp at deepthought.com>:
> 
> >On Wed, May 17, 2006 at 07:31:39AM -0400, Robert Gault wrote:
> >>*j Means non-existent module. t Means the system has tried to boot. While 
> >>t
> >>normally shows up at the beginning of the sequence, it can appear again. 
> >>So
> >>t*j means that a module can't be found so Boot failed.
> >
> >Do you know of any method other then trial and error to find out 
> >which module is
> >missing?
> >
> >>Given what you are attempting, my guess is that you either do not have
> >>/term in your bootfile or are missing sysgo or maybe grfdrv. Whatever it
> >>is, I doubt you are using the script supplied by Cloud-9 for booting from
> >>their systems.
> >
> >Forgive my ignorance as I'm new to nitros9 where would I find /term 
> >and is the
> >bootfile the same as a bootlist file? When I got the Superide Mark 
> >installed
> >the 6809 version of nitros for me. Since my coco had a 6809 it seemed 
> >like the
> >thing to do. I have since replaced the 6809 with a 6309 and would 
> >like to take
> >advantage of the extra performance and capabilities of it. I got a 
> >new CF card
> >and after formatting it I copied the files from the 6309 nitros 
> >distro disks I
> >got at the fest a few weeks ago. I also installed the drivers for 
> >superdrive and
> >drivewire. I tried using the mb script along with the supplied 
> >bootlist file. That
> >gave me a bootable floppy but I couldn't access the ide controller or 
> >CF card.
> >No amount of uncommenting modules in the bootlist file would result 
> >in a kernel
> >that would allow me to access the CF card. When I got the CF card 
> >preconfigured
> >from cloud9 it included a couple of scripts to build a kernel and the 
> >bootlists
> >that those scripts used. That and the comments in the scripts and 
> >bootlists files
> >led me to believe that they were provided by cloud9. If that's not the 
> >case I
> >apologize to the author of the files.
> >
> >Frank
> >
> >
> >
> >
> >
> >>
> >>Frank Pittel wrote:
> >>>Chris,
> >>>
> >>>I've been using the "old" os9gen that I have. I've copied the build 
> >>scripts
> >>>and bootlists that cloud9 put on the CF card for me to the new card. I've
> >>>fixed
> >>>up the scripts and bootlists with the correct paths for the drivers and
> >>>whether
> >>>or not I boot with a "boot" floppy or from hdb the result is the same. 
> >>all
> >>>seems
> >>>well at first. I get a line and a half of dots. (.) and then it starts to
> >>>load modules
> >>>for about two and a half lines. At this point all seems well and normal.
> >>>At the
> >>>end of the module list it list i2x0. At this point it should continue 
> >>with
> >>>a capital
> >>>c (C) and then finish booting. In my case it displays t*j (i2xot*j) and
> >>>then displays
> >>>"boot failed". I've tried booting with both sysgo files included with
> >>>nitros and the
> >>>result is the same. I've run ident on the two sysgo files and they show 
> >>up
> >>>as being
> >>>good.
> >>>
> >>>Frank
> >>>
> >>>
> >>>
> >>>On Tue, May 16, 2006 at 09:56:44PM -0500, Christopher Hawks wrote:
> >>>
> >>>>Frank Pittel wrote:
> >>>>
> >>>>>I noticed that the page is now back up. Unfortunatly I couldn't find
> >>>>>what I
> >>>>>was hoping for. As I understand the latest and greatest version of
> >>>>>os9gen has
> >>>>>a bug in it which presents itself when building a kernel for use on a 
> >>CF
> >>>>>card
> >>>>>that is partitioned for use by HDB and Nitros9
> >>>>>
> >>>>>I am having another problem that I'm hoping I can get some help with. I
> >>>>>have a
> >>>>>CF card that I bought today with the idea of installing both HDB and
> >>>>>Nitros9 onto.
> >>>>>(Interestingly enough I think I found a bug with HDB and it's dealing
> >>>>>with the
> >>>>>particular brand of card I have. Any data written to the card with
> >>>>>"save" or "copy"
> >>>>>is lost at reboot. However if written to the virtual file with "backup"
> >>>>>it's fine.
> >>>>>The card works fine with nitros though. <Boisy if you have any interest
> >>>>>in this let
> >>>>>me know and I'll send you the card when I get a replacement>) What I
> >>>>>need is a "bootlist"
> >>>>>and script that will allow me to gen a bootable floppy. Amoung the
> >>>>>modules I need
> >>>>>installed is the hardware clock on my superide card, Superdriver and
> >>>>>drivewire. I have
> >>>>>the modules I need but I just can't seem to build a os9boot file that
> >>>>>will boot. :-(
> >>>>>It would help if someone could email me what you have working for you
> >>>>>for the 6309.
> >>>>
> >>>>Frank:
> >>>>
> >>>>	The os9gen bug shows up first when to you run "link.bas" after
> >>>>	backing up the bootdisk to an HDB-DOS disk. It will report the
> >>>>	bootfile size and location, but, will say the bootfile size is zero
> >>>>bytes! The bug causes _all_ bootfiles to be the fragmented type, and
> >>>>"link.bas" can't handle this.
> >>>>
> >>>>	If "link.bas" gives a real file size, you don't have the bug. The
> >>>>change has been committed to CVS.
> >>>>
> >>>>
> >>>>--
> >>>>Christopher R. Hawks
> >>>>HAWKSoft
> >>>>---------------------------------------------------------
> >>>>Any research done on how to efficiently use computers has been long lost
> >>>>in the mad rush to upgrade systems to do things that aren't needed by
> >>>>people who don't understand what they are really supposed to do with 
> >>them.
> >>>>	-- Graham Reed, in a.s.r.
> >>>>
> >>>>--
> >>>>Coco mailing list
> >>>>Coco at maltedmedia.com
> >>>>http://five.pairlist.net/mailman/listinfo/coco
> >>>
> >>>
> >>
> >>--
> >>Coco mailing list
> >>Coco at maltedmedia.com
> >>http://five.pairlist.net/mailman/listinfo/coco
> >
> >--
> >Coco mailing list
> >Coco at maltedmedia.com
> >http://five.pairlist.net/mailman/listinfo/coco
> >
> 
> 
> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco