Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 10 Apr 2000 00:38:36 -0700
From:      Aaron Smith <aaron-fbsd@mutex.org>
To:        Joe Greco <jgreco@ns.sol.net>
Cc:        smp@freebsd.org
Subject:   Re: Any problems with serial consoles?
Message-ID:  <20000410003836.A88784@gelatinous.com>
In-Reply-To: <200004111252.HAA20496@aurora.sol.net>; from jgreco@ns.sol.net on Tue, Apr 11, 2000 at 07:52:41AM -0500
References:  <200004111252.HAA20496@aurora.sol.net>

next in thread | previous in thread | raw e-mail | index | archive | help
hi joe, 

one of my boxes is almost this exact hardware (well, dual p2-400 on asus
p2b-ds, 256m ram, scsi) and serial console works fine. when i was debugging
serial console early on, i had a problem where the kernel knew the correct
baud rate but the switchover once boot messages were finished resulted in
garbage. you might want to read:

     http://www.mutex.org/aaron/tips/freebsd-serial-console

which is a breakdown of what i did to get serial console working. from your
description it sounds like baud rate problems OR the terminal program
you're using. the bootup process sends a break a couple times; tip is
pretty awful and would hang up on me or get confused. kermit works though.

aaron

On Tue, Apr 11, 2000 at 07:52:41AM -0500, Joe Greco wrote:
> I've a continuing (multi-year) issue with serial consoles on SMP machines.
> 
> The general issue is that, during bootup, the console "loses it" and seems
> to switch baud rates or something, and all user I/O becomes garbage.
> However, kernel messages continue to print out correctly(!).
> 
> The hardware I typically use is ASUS P2B-DS, large RAM (512MB+), Adaptec
> 3940's, and SMC Dual EtherPower 10/100(s).
> 
> The OS revs are basically everything from 3.0 on up, IIRC.
> 
> The system continues to boot just fine, and eventually the getty on ttyd0
> will clear up the situation by resetting the terminal modes.  I infrequently
> see the situation reassert itself after boot, usually after very long
> uptimes.
> 
> The real problem is single-user mode.  The sequence
> 
> # ccdconfig -Cv; mount -a; sh /etc/netstart
> 
> is at least 50% likely to leave the system in the state described, which of
> course means that it cannot hear further commands such as "reboot god damn
> it I don't want to drive to Chicago to reset you".  So I usually end up
> doing something like
> 
> # (sleep 900; reboot)&
> # ccdconfig -Cv; mount -a; sh /etc/netstart; stty sane; reset
> 
> which generally seems not to fail in the described manner.
> 
> I used to be fairly sure it was something in the networking that was
> causing the problem, but recently I ran into the situation when I was
> merely puttering with the disks after doing a "vinum start; mount -a"
> type affair.
> 
> Solutions solicited.  I may be able to provide a box with the described
> syndrome if anyone wants to look at it.
> 
> ... Joe
> 
> -------------------------------------------------------------------------------
> Joe Greco - Systems Administrator			      jgreco@ns.sol.net
> Solaria Public Access UNIX - Milwaukee, WI			   414/342-4847


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-smp" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000410003836.A88784>