Re: FIXED: dbootstrap problem + why is init_tab obsolete, and what obsoletes it?
On Wed Jan 12, 2000 at 03:39:59PM -0700, Matt Porter wrote:
> On Wed, Jan 12, 2000 at 04:15:20PM -0500, Ben Collins wrote:
> > First off, I was able to fix the dbootstrap problem. It seems that
> > starting rcS and dbootstrap on the console causes a problem of sorts
> > (control of the terminal is not handed over properly). There is probably a
> > better "fix", but basically I just moved rcS output to tty3. I also moved
> > some things from rcS to inittab (syslogd startup, cat /proc/kmsg, and tail
> > -f /var/log/messages), which seems to work pretty well.
> Yeah, we'll need something else to handle serial console systems that have
> all messages output on /dev/ttyS0.
dbootstrap will still run on whatever /dev/console resolves as, and
at least system messages will show up in /var/log/messages, so serial
console systems will see at least those messages if they choose to look.
The output from rcS will generally not be very critical.
On a related note, we probably should just let syslogd start klogd, so
both system and kernel messages will be logged to /var/log/messages.
With the current scheme, folks on serial consoles won't have access to
kernel message logs...
Erik B. Andersen Web: http://www.xmission.com/~andersen/
--This message was written using 73% post-consumer electrons--