Date: Mon, 5 Feb 2001 10:07:52 -0800 (PST) From: Rich Wales <richw@webcom.com> To: freebsd-stable@freebsd.org Subject: Re: BRIDGE breaks ARP? Message-ID: <20010205175257.40469.richw@wyattearp.stanford.edu>
next in thread | raw e-mail | index | archive | help
Tom Veldhouse wrote:
> > Any chance this problem could be caused by having MROUTING
> > and BRIDGE in the kernel at the same time? . . . I have
> > noticed that I had MROUTING in the kernel and I have these
> > same problems.
Masachika ISHIZUKA replied:
> I do not have MROUTING.
And neither do I.
I noticed something else this morning that might be significant. My
main desktop machine (the one which can't talk directly to my bridge
via its "rl0" interface unless I use an "arp -s" command to hardwire
it with knowledge of the bridge's hardware address) logged a bunch
of unsuccessful attempts by the bridge to supply its =external=
interface's hardware address. For example:
Feb 4 20:23:09 jekyll /kernel: arp: 00:60:97:05:32:cd
attempts to modify permanent entry for 171.66.188.114 on rl0
I had "tcpdump arp" commands (one for each of the two interfaces in
the bridge cluster) running on my bridge at the time, and the bridge
did =not= send out any ARP replies directed to my desktop. It did,
however, send out quite a few ARP replies addressed to my DSL modem
(on the external, "xl0" interface).
Is it possible that these ARP replies were being bridged to the
internal, "rl0" interface, and that my desktop (currently running
4.2-RELEASE) was picking them up and trying to process them, even
though they were intended for another host (my DSL modem)? If so,
this seems buggy to me; my desktop really only wants to know about
=one= hardware address for the bridge (not two).
Rich Wales richw@webcom.com http://www.webcom.com/richw/
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010205175257.40469.richw>