Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 14 Feb 2007 22:15:22 +0000
From:      "Bruce M. Simpson" <bms@FreeBSD.org>
To:        freebsd-current@freebsd.org
Subject:   Re: panic: sbflush_internal: cc 4294966301 || mb 0 || mbcnt 0
Message-ID:  <45D389FA.2010604@FreeBSD.org>
In-Reply-To: <45D37239.1020100@FreeBSD.org>
References:  <45C64E3A.7050407@kasimir.com> <45D37239.1020100@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
I was able to reproduce this bug, although the circumstances under which 
I did it are weird.

I have two NICs on the same box patched to the same Ethernet broadcast 
domain, but on interconnected switches with no spanning tree. If I stop 
VLAN_HWTAGGING on msk with a vlan interface attached, and then run ping 
-f to a local host via the network configured on fxp, it seems to 
trigger the problem.

sb_cc is always 84 bytes, which is 20 bytes IPv4 header + 64 bytes of 
ICMP data, so it's definitely ping which is triggering the issue in my 
case. I haven't been able to get a clean dump yet as my 6.2-RELEASE 
userland doesn't seem to understand -CURRENT's minidumps.
 
Hope this helps,
BMS



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?45D389FA.2010604>