From owner-freebsd-stable  Wed Dec  2 18:21:26 1998
Return-Path: <owner-freebsd-stable@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id SAA11262
          for freebsd-stable-outgoing; Wed, 2 Dec 1998 18:21:26 -0800 (PST)
          (envelope-from owner-freebsd-stable@FreeBSD.ORG)
Received: from chumbly.math.missouri.edu (chumbly.math.missouri.edu [128.206.72.12])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id SAA11243
          for <freebsd-stable@freebsd.org>; Wed, 2 Dec 1998 18:21:23 -0800 (PST)
          (envelope-from rich@chumbly.math.missouri.edu)
Received: (from rich@localhost)
	by chumbly.math.missouri.edu (8.9.1a/8.9.1) id UAA21832
	for freebsd-stable@freebsd.org; Wed, 2 Dec 1998 20:21:09 -0600 (CST)
From: Rich Winkel <rich@chumbly.math.missouri.edu>
Message-Id: <199812030221.UAA21832@chumbly.math.missouri.edu>
Subject: accessing quiescent fbsd machine via network for maintenance
To: freebsd-stable@FreeBSD.ORG
Date: Wed, 2 Dec 1998 20:21:09 -0600 (CST)
X-Mailer: ELM [version 2.4 PL23]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-stable@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

I'm running FBSD 2.2.7-stable.
For various reasons I'd like to be able to upgrade fbsd machines without
having to be seated in front of them.  Specifically, I want to reboot
into a mode where only a minimal amount of system activity is
occurring, regular users are excluded, but remote root access and nfs
mounts are allowed.  This would allow me to login, mount a remote /usr/src
& /usr/obj, run "make installworld" and install a new kernel.
But this seems to be problematic.  I can do it at the console by
rebooting singleuser and running fsck, swapon, ifconfig and mount, but
I have problems when I try to do it with a modified /etc/rc.
Specifically, in /etc/rc I check for the existence of the file
/etc/netmaintboot immediately after the network_pass1 call.  If the
file exists I execute /etc/rc.netmaint and exit 0.  
/etc/rc.netmaint consists of:

#!/bin/sh
# boot configured for maintenance from network
ldconfig /usr/lib
echo "Down for maintenance, try again later" > /etc/nologin
TERM=cons25 clear
echo "Entering maintenance mode, please don't touch"
/sbin/sshd -b 1024 &
exit 0

But for some reason sshd dies immediately and I can't login.  But if I
then login at the console and run sshd manually it works fine.

Any ideas??  Do you think this kind of capability should be incorporated
into a future release?

Thanks!!
Rich


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