My results are different...
> /* snip - mind you, these were handwritten, so some stuff may be lacking ;) */
>
> emulator [1] e_mapped_timeofday init failed 2
> init: wait for /bin/sh on /etc/rc failed: No such child process; go to single user
>mode
> server_exec_open: emulator "/xMach/etc/emulator" not found: 49154
this line is missing on my screen
> emulator [3] e_mapped_timeofday init failed 2
got this too
> (CORRECT TIME STAMP)
endline missing on this statement - this statement is not there as well.
init: can't get /dev/console for controlling terminal: Operation not permitted
got this too
> server_exec_open: emulator "/xMach/etc/emulator" not found:49154
nope, don't have this.
> (CORRECT TIME STAMP)
don't have this either
init: /bin/sh on /etc/rc terminated abnormally, going to single user mode
got this too
> Enter pathname of shell or RETURN for /bin/sh:
got this too
> /* end snip */
end
> upon hitting return or entering a pathname to a shell, i was quickly greeted
> with the "emulator ... not found" error. problem is that the emulator *WAS*
> there.
well, emulator works for me. the message I get is:
init: single user shell terminated, restarting
Enter pathname of shell or RETURN for sh:
and upon return the same thing again. No shell I tried works. Since this
is a netbsd init, I am not sure what shell it expects. Maybe a netbsd sh
may work. I will try this.
> the other irritating thing was that the #ifdef DEBUG'd printf() statements i
> put into the e_mapped_time.c file didn't show up.
If this is part of the emulator, there may be your problem, it cannot find
the binary, as far as I can tell.
> the extraordinarily irritating thing was that even though the correct time
> was showing up (where "CORRECT TIME STAMP" is above), upon reboot the time on
> my machine returned to being hosed. argh.
This WILL go away soon. Joseph has already committed some fix for the time
issue. I am not sure this particular one is taken care of, but again, it
will.
> so tomorrow it's on to more grepping & debugging. and thinking about how we
> really need our own userland ;-)
I think what we need is our own init. Mach_init is busted, and we don't
know why...
OpenBSD userland is scheduled to make the cut for release 0.2
Good work, ian.
JAn
