Subject: Re: silo and packets: overflowing too-bigness
To: Frederick Bruckman <fredb@fb.sa.enteract.com>
From: Bill Studenmund <wrstuden@loki.stanford.edu>
List: port-mac68k
Date: 07/27/1998 11:23:39
>
> On Sun, 26 Jul 1998, Paul Goyette wrote:
> > On Sat, 25 Jul 1998, Brian Wildasinn wrote:
> > >
> > > /netbsd: ppp0: bad protocol 820
> > > /netbsd: ppp0: bad fcs e262
> > > /netbsd: ppp0: bsd fcs 26f0
> > > /netbsd: zstty0: 3 silo overflows, 0 ibuf floods
>
> Your only recourse is to lower your connection speed. I can get no more
> than 38400 out of my '630--57600 gives silo overflows, so I don't use it,
> even though it hasn't caused any problems like you're haveing.
I agree, though I find it surprising that you always get a protocol 820 -
some byte is systematicly getting dropped (or the other side's doing
something VERY weird).
> > > [can I get the scrolling to stop]
> Obviously, you know about xconsole, since you mentioned it. If you have
> xconsole running, and your screen is still getting messed up, perhaps you
> are starting xdm/X from a root shell? xconsole only grabs the messages to
> the console, not to the root user. You could fix that by editing
> /etc/syslog.conf, but what you should be doing, IMO, is either running
> startx from a user login, or setting the line in /etc/rc.conf to xdm=YES
> and also turning off all gettys in /etc/ttys. Note that in the latter
> case, you can still get a shell in single user mode, when you need it.
The other thing to do is to make sure that you have options INSECURE in
your kernel config. It is set in the GENERIC kernels. Without it, xterm -C
won't work right.
Take care,
Bill