Last night, I decided to reinstall the system software in my Softkicked A3000 (running WB 2.04 with 2meg chip RAM and 4 meg fast RAM). As far as I could tell, everything went fine with the installation. However, when I rebooted, the machine guru-ed (8000 0004) before the startup sequence completed.
After much investigations, it appears that said crash can be invoked by any access to the ram: disk. This also occurs when booting in 1.3 mode.
After removing all references to ram: from my startup sequence, I can now boot my machine (in fact, I'm writing this using terminus on it right now).
Can anyone venture a guess as to why accessign RAM: would crash my machine when everything else is apparently OK?
On 15-Feb-96 09:08:23, Dave Schaumann wrote this about HELP! Accessing RAM: crashes my A3000!;
> Last night, I decided to reinstall the system software in my > Softkicked A3000 (running WB 2.04 with 2meg chip RAM and 4 meg > fast RAM). As far as I could tell, everything went fine with > the installation. However, when I rebooted, the machine guru-ed > (8000 0004) before the startup sequence completed. > After much investigations, it appears that said crash can be > invoked by any access to the ram: disk. This also occurs when > booting in 1.3 mode. > After removing all references to ram: from my startup sequence, > I can now boot my machine (in fact, I'm writing this using terminus > on it right now). > Can anyone venture a guess as to why accessign RAM: would crash > my machine when everything else is apparently OK? > Thanks in advance... > -Dave
An older version of Ramdisk.device maybe ?
Ken...................
--
EMail: K...@kenwyn.demon.co.uk * Motorola Motivated A1200T 060/50 18Meg Amiga (the TRUE) Multitasking - Multiplatform - Multimedia Machine Amiga OS3, MS-DOS 6.2 , Macintosh System 7.5.1
Someone also suggested this to me in e-mail. Does wb2.04 use the ramdisk.device? I see no mention of it in _Using the System Software_.
Also, I have an update on the behavior. I originally wrote:
>> After much investigations, it appears that said crash can be >> invoked by any access to the ram: disk.
It turns out this is not entirely true. I can, for example, use makedir to create a directory on ram:. I also discovered that I can use terminus to download a file to ram:.
Some things that *do* cause my machine to crash: in the shell, changing directory to ram: and attempting to use copy to copy a file to ram:. If I use makedir to create a directory on ram:, the ram: icon appears on the workbench. If I then double-click the icon to open it, the window opens, but the machine immediately locks up followed shortly by a visit to the guru.
I'm beginning to think I have a bad kickstart image. Version shows kickstart version 37.175 and workbench version 37.67. devs:kickstart is 526848 bytes long. Unfortunately, I have no idea what these values were before this debacle started. If someone else out there has a soft-kicked A3000 with WB2.04 installed, could you please check these values against what you have on your machine and e-mail me the results?
In article dated 14 Feb 1996 21:08:23 -0700, Dave Schaumann (d...@CS.Arizona.EDU) wrote:
> Last night, I decided to reinstall the system software in my > Softkicked A3000 (running WB 2.04 with 2meg chip RAM and 4 meg > fast RAM). As far as I could tell, everything went fine with > the installation. However, when I rebooted, the machine guru-ed > (8000 0004) before the startup sequence completed.
> After much investigations, it appears that said crash can be > invoked by any access to the ram: disk. This also occurs when > booting in 1.3 mode.
I saw this (and fixed it!) on a friends Phoenix A1000. Turned out that what he had done was install the fastfilesystem on both his hard drives. To make it even more interesting he had different versions on either drive. One drive had 1.3, the other 2.04. When booting 1.3, it would come up fine, but he had practically no fast ram. 2.04 could be persuaded to boot only by removing all reference to RAM: from his startup sequence.
Do you have multiple hard drives? Do you have the filesystem installed on both drives? Check if the versions and sizes match. Remove the fastfilesystem from one of the drives.
-- _--_|\ John Verhoeven (jo...@acix.DIALix.oz.au or jo...@DIALix.oz.au) / \ >>>> Writing programs in assembler is like digging post <<<< *_.--._/ >>>> holes with a teaspoon: it takes a long time, but <<<< v >>>> you have complete control over all the dirt. <<<<
> On 15-Feb-96 09:08:23, Dave Schaumann wrote this about HELP! Accessing RAM: > crashes my A3000!; > > Last night, I decided to reinstall the system software in my > > Softkicked A3000 (running WB 2.04 with 2meg chip RAM and 4 meg > > fast RAM). As far as I could tell, everything went fine with > > the installation. However, when I rebooted, the machine guru-ed > > (8000 0004) before the startup sequence completed. > > After much investigations, it appears that said crash can be > > invoked by any access to the ram: disk. This also occurs when > > booting in 1.3 mode. > > After removing all references to ram: from my startup sequence, > > I can now boot my machine (in fact, I'm writing this using terminus > > on it right now). > > Can anyone venture a guess as to why accessign RAM: would crash > > my machine when everything else is apparently OK? > An older version of Ramdisk.device maybe ?
Nice try. ramdrive.device is *not* used for the RAM: disk. However, OS 1.x systems need L:RAM-Handler for the RAM: disk, I think.
Try deleting L:RAM-Handler, and you system should work with OS 2.0+
-- Regards,
/ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\ | Rask Ingemann Lambertsen | E-mail: c948...@student.dtu.dk | | Amiga GNU CC README maintainer | WWW: http://www.gbar.dtu.dk/~c948374/ | | Keyboard error: <Ctrl> and <Alt> are stuck - press <Del> to continue |
On 17-Feb-96 07:11:56 d...@CS.Arizona.EDU Dave Schaumann wrote:
> devs:kickstart >is 526848 bytes long. Unfortunately, I have no idea what these values >were before this debacle started.
Last I checked, 512k = 524,288 bytes--2.5k *shorter* than your image.
>Thanks in advance to anyone who replies. >-Dave
The opinions expressed are not necessarily those of the Gateway Amiga Club,Inc. or its members. William F. Maddock Editor in Chief GAC FLAK wmadd...@icon-stl.net The Newsletter of the Gateway Amiga Club, Inc. www.icon-stl.net/~wmaddock
In article <639.6621T1281T1...@icon-stl.net> (William F. Maddock) writes: > On 17-Feb-96 07:11:56 d...@CS.Arizona.EDU Dave Schaumann wrote:
> > devs:kickstart > >is 526848 bytes long. Unfortunately, I have no idea what these values > >were before this debacle started.
That is the correct size for the 2.04 softkick file.
> Last I checked, 512k = 524,288 bytes--2.5k *shorter* than your image.
Yes, but the softboot Kickstart file has an extra chunk of code on the end to identify and validate it to the softboot roms.
-- Cheers, --- + _ ____________ tm Dale Currie ____ ___ _ + | /.\ .. | __ \ / da...@amitrix.com / __[___]__ T tm | | /___\ /\/\ | | |_) | X supp...@amitrix.com / (o.o) | | | / \/ ^^ \ | | | \ | / \ Edmonton AB Canada / `-^-' | | |/ - D E V E L O P M E N T - \ AmiTrix /___ Z O R R O I N K !| + --------------------- Technical Support ---------------- +
Schaumann) wrote: > Last night, I decided to reinstall the system software in my > Softkicked A3000 (running WB 2.04 with 2meg chip RAM and 4 meg > fast RAM). As far as I could tell, everything went fine with > the installation. However, when I rebooted, the machine guru-ed > (8000 0004) before the startup sequence completed.
> After much investigations, it appears that said crash can be > invoked by any access to the ram: disk. This also occurs when > booting in 1.3 mode.
> After removing all references to ram: from my startup sequence, > I can now boot my machine (in fact, I'm writing this using terminus > on it right now).
> Can anyone venture a guess as to why accessign RAM: would crash > my machine when everything else is apparently OK?
> Thanks in advance... > -Dave
I had the exact same symptom, and too thought it was the ramdisk device. However I determined that the RAD device was corrupted. Try replacing your RAD device located in your DEVS: directory.
