"Chris G. Demetriou" <cgd@postgres.berkeley.edu> wrote:
> however, there have been other changes to the file systems that
> require that newly-newfs'd fs's be used with newer kernel.  (e.g.
> 32-bit uid's and gid's.)  You can still use your old file systems
> with new kernels with no problems, though once you've decided that
> a new kernel is always goign to be running on your machine, you
> should 'fsck -c 2' them.
         ^^^^^^^^^^^

  Well, sometimes I like to play the "what-does-this-button-do" type of
guy. I started up netbsd-current (June 13) in single user mode typed
fsck -c 2 at the prompt and after fsck printed /dev/rwd0a a spontaneous
reset was all that happened.
 The output from script in the next couple of lines shows the current
behaviour of fsck
|Script started on Tue Jun 14 20:38:06 1994
|sheep# fsck
|** /dev/rwd0a
|BAD SUPER BLOCK: VALUES IN SUPER BLOCK DISAGREE WITH THOSE IN FIRST ALTERNATE
|Floating exception (core dumped)
|
|Script done on Tue Jun 14 20:38:13 1994

 Like William said: "To newfs or not to newfs, that's the question".

Onno van der Linden    c/o    vdlinden@fwi.uva.nl (Frank van der Linden)

------------------------------------------------------------------------------