Fw: Re: sysinstall and mfs I'm out of Ideas.

[Mark]

--- On Sat, 2/6/10, Mark <redtick at sbcglobal.net> wrote:

> From: Mark <redtick at sbcglobal.net>
> Subject: Re: sysinstall and mfs I'm out of Ideas.
> To: "Martin McCormick" <martin at dc.cis.okstate.edu>
> Date: Saturday, February 6, 2010, 12:17 AM
> 
> 
> --- On Fri, 2/5/10, Martin McCormick <martin at dc.cis.okstate.edu>
> wrote:
> 
> > From: Martin McCormick <martin at dc.cis.okstate.edu>
> > Subject: Re: sysinstall and mfs I'm out of Ideas.
> > To: "Mark" <redtick at sbcglobal.net>
> > Date: Friday, February 5, 2010, 8:21 PM
> > Mark writes:
> > > I'm sure you have already read this from man
> > sysinstall, but it may give 
> > > a hint to something you may have missed.
> > > 
> > > 
> >    <<<<   Note:
> > Nothing is actually written to disk by this function,
> an
> > >             
> > explicit call to diskPartitionWrite being required for
> that
> > 
> > > to happen. >>>>
> > > 
> > >      diskPartitionWrite
> > >              Causes
> > any pending MBR changes (typically from the
> > >             
> > diskPartitionEditor function) to be written out.
> > > 
> > >             
> > Variables: None
> > > 
> > > 
> > > <<< Note: No file system data is
> actually
> > written to disk until an
> > >             
> > explicit call to diskLabelCommit is made.
> > 
> >     Thank you. The commit is definitely
> > never happening.
> > When it does happen, several messages appear and there
> is a
> > time
> > lag while the newfs takes place. It should happen
> right
> > after
> > one selects the installation media. Instead, it just
> skips
> > that
> > step and ruins mfs by overwriting some of its
> utilities
> > with
> > binaries that should be going to the hard drive. 
> > Again, thanks.
> > 
> > Martin McCormick
> >
> 
> perhaps adding "sleep 120" to halt the script execution for
> 2 minutes to allow the new files system write to complete.
> hth
> 
>