wduke@cogeco.ca wrote:

>Right, worse problems.   How eloquently said and incredibly accurate.   ;)
>
>There's a lot of stuff missing from the installation.   The thing stalled
>while inflating base.tgz and just continued on to the next set, etc.tgz.
>Anyway, to make a long story short, the installation failed a basic sanity
>test. (That cracks me up everytime I think about it.) :)
>
>I think I'm just gonna try a re-installation using the tried and true
>traditional method on a different hard drive.
>
>Thanks,
>
>William
>
>
>  
>
>>From: der Mouse <mouse@Rodents.Montreal.QC.CA>
>>Date: Fri, 14 Oct 2005 14:18:33 -0400 (EDT)
>>To: port-mac68k@NetBSD.org
>>Cc: (in case you're not reading the list:) William Duke <wduke@cogeco.ca>
>>Subject: Re: Error on SCSIRead(), #5
>>
>>    
>>
>>>/sbin/mount -u -w /
>>>      
>>>
>>>mount: Cannot open '/etc/fstab': No such file or directory
>>>mount_ffs: root_device on /: No such file or directory
>>>      
>>>
>>>Hmm, I can't change my file system to 'read write' because 'fstab' is
>>>missing from my 'etc' directory.
>>>      
>>>
>>Not quite.  You can't change it to RW *that way* because /etc/fstab is
>>missing.
>>
>>You see, mount needs to know both the thing being mounted and the place
>>it's being mounted on.  The command you gave specifies only one of
>>those, so mount tries to go off and figure out the other on its own.
>>First, it tries looking in fstab, but you don't have one.  Then it
>>tries looking at the mount table in the kernel, and it finds a mount on
>>/, but the thing mounted there according to the kernel is
>>"root_device", which doesn't exist (hence the second error line).  (The
>>kernel sticks that string in there because it doesn't know what /dev
>>entry corresponds to the root device, though arguably it'd be better to
>>make a guess that'd usually be right rather than not even try because
>>the guess might occasionally be wrong.)
>>    
>>
>
>Again, how eloquently said and incredibly accurate.   It is most certainly
>better to hazard a guess in the absence of definite knowledge than it is to
>simply give up.   Besides, you should be able to use the process of
>elimination to narrow down your options when making the guess.
>
> 
>  
>
>>You probably want something like "/sbin/mount -u -w /dev/sd0a /" - that
>>is, explicitly specify the device as well as the mount point.
>>
>>But you have worse problems.  If whatever you used to install didn't
>>set up fstab, it probably didn't set up various other things too.
>>You'll need to find and fix whichever of them are important to you.
>>
>>    
>>
>
>Right!   It didn't setup a lot of other things, a default terminal setting,
>for example.  I watched the second installation attempt rather closely, so I
>know that the problems can be almost certainly attributed to the portion of
>the installation specific to the base.tgz set.   All other sets show no
>outward indication of having failed during installation.
>
> 
>  
>
>>I don't know what they are or I'd give you a list.  My own list of
>>things to do to a new install is:
>>- set up /etc/rc.conf.local
>>- set up /etc/fstab, make any needed mount points (eg, /kern, /proc)
>>- set up /etc/inetd.conf
>>- set up /etc/ntp.conf
>>- set up /etc/localtime
>>- set up /etc/resolv.conf
>>- fix root crontab
>>- passwd root, add any others (incl. add mouse to groups!)
>>but this is rather heavily tuned for my own setup.
>>
>>/~\ The ASCII    der Mouse
>>\ / Ribbon Campaign
>>X  Against HTML        mouse@rodents.montreal.qc.ca
>>/ \ Email!      7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
>>    
>>
>
>  
>

SCSIRead(), #5 from what i read, mkfs and the installer dont like drives 
about 1gig and up. Try getting base.tgz, etc.tgz and kern*.tgz to 
install first, then cpin the other sets and cd / and type tar -zpxf 
*.tgz Then it should work, but if it doesent, try getting base.tgz and 
kern.tgz installed as much as possible then cpin the other sets (again), 
and hope that tar was installed.

Good luck :) ,
Chuck Findlay