From owner-freebsd-questions  Sun Oct 22 15:54:51 1995
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id PAA10647
          for questions-outgoing; Sun, 22 Oct 1995 15:54:51 -0700
Received: from irbs.irbs.com (irbs.com [199.182.75.129])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id PAA10641
          for <freebsd-questions@freebsd.org>; Sun, 22 Oct 1995 15:54:46 -0700
Received: (from jc@localhost) by irbs.irbs.com (8.6.12/8.6.6) id SAA06193; Sun, 22 Oct 1995 18:53:35 -0400
From: John Capo <jc@irbs.com>
Message-Id: <199510222253.SAA06193@irbs.irbs.com>
Subject: Re: [ET-users] crash
To: et-users@netrail.net
Date: Sun, 22 Oct 1995 18:53:34 -0400 (EDT)
Cc: freebsd-questions@freebsd.org
In-Reply-To: <Pine.LNX.3.91.951022125005.24849A-100000@netrail.net> from "Nathan Stratton" at Oct 22, 95 12:58:28 pm
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1696      
Sender: owner-questions@freebsd.org
Precedence: bulk

Nathan Stratton writes:
> 
> Oct 22 16:46:09 nr-dc-1 /kernel: arpresolve: can't allocate llinfo
> Oct 22 16:46:40 nr-dc-1 last message repeated 9 times
> Oct 22 16:48:43 nr-dc-1 last message repeated 27 times
> Oct 22 16:58:50 nr-dc-1 last message repeated 88 times
> Oct 22 17:04:25 nr-dc-1 /kernel: FreeBSD 2.0.5-RELEASE #0: Sat Oct 21 21:32:22  1995
> Oct 22 17:04:26 nr-dc-1 /kernel:     root@nr-dc-1:/usr/src/sys/compile/LOCAL
> 
> And then it boots up fine. I have only noticed the actualy panic message 
> once, is ther a way to log this so I can get more info at what is causing 
> this?

This is an old bug.  It was fixed along with a slew of others that
crept in when the new routing code was added.

I was never bitten by this particular bug but various route cloning
bugs did bite me quite often on my router.  I worked around it by
disabling route cloning.

In the function in_addroute() in netinet/in_rmx.c, comment out this
section of code.


        /*
         * For IP, all unicast non-host routes are automatically cloning.
         */
        if(!(rt->rt_flags & (RTF_HOST | RTF_CLONING))) {
                if(!IN_MULTICAST(ntohl(sin->sin_addr.s_addr))) {
                        rt->rt_flags |= RTF_PRCLONING;
                }
        } 

An even better solution is to upgrade to 2.1.

If the system is panicing, it should also be saving a copy of the
running system in the swap space, if dumps are enabled, and you
have enough swap space, and /var/crash exists, and /var/crash has
enough space.

Are there any other arp* messages in the log?  Do you see a "mb_map
full" message in the log?

John Capo
IRBS Engineering                      High performance FreeBSD systems