> Put when I tried ppp-up it prints
> (I am on /etc/ppp and I type sh ppp-up ,and I am on single-user mode, and
> I write on the beginnig mount -u /)
> 
> grep: */status no such file or directory
> connecting...
> #
> 
> and nothing !
> Nothing send to my modem ( I have verified it, pluging my PowerBook, on
> the modem-port, with a null-modem cable )
> Is there something wrong

Oh yes...  In fact, I believe the said scripts originated from me, so
I think I know the reason.  Boy, it's been quite a while since I
stopped using these PPP script in favor of newer SLIP scripts written
in Expect (a Tcl-based tool for automated testing of interactive
programs).  No, it doesn't mean the PPP scripts are inferior to my
newer SLIP scripts.  It just is a side effect of being written in
Expect which echoes everything going between my box and the modem, so
that I can see what's going on and know what's wrong with our terminal
servers when something goes wrong with them, which happens frequently
enough...

Anyway, the ppp-* scripts rely on procfs mounted at /proc.  This is
because at the time I wrote those scripts, there was often
disagreement between the kernel and the ps about the process table
format, especially when you downloaded only a new kernel file and not
the rest of the system binary.  So I decided not to use ps for
reliable process detection to kill pppd when shutting down the link.
Instead, I decided to use procfs to do the job.  Therefore, you cannot
use the ppp-* scripts when /proc isn't mounted.

>From the point of view of who designed the scripts, these scripts
aren't designed for a system with semi-permanent or permanent PPP
link.  They are for people like who I used to be, people who just get
on the net at most a few hours a day, while the machine itself is kept
up for days.

So, if you want to establish the ppp link as soon as NetBSD is up, and
keep the link up as long as NetBSD is up, then these scripts are not
appropreate for the job.  You should write your own dial-out script
and so on.

ken