Date: Sun, 24 Feb 2002 10:36:16 -0800 From: "Crist J. Clark" <crist.clark@attbi.com> To: Ruslan Ermilov <ru@FreeBSD.ORG> Cc: net@FreeBSD.ORG Subject: Re: TCP Connections to a Broadcast Address Message-ID: <20020224103616.S16048@blossom.cjclark.org> In-Reply-To: <20020224084702.GB31243@sunbay.com>; from ru@FreeBSD.ORG on Sun, Feb 24, 2002 at 10:47:02AM %2B0200 References: <20020222022626.A83807@blossom.cjclark.org> <20020223115033.GB47437@sunbay.com> <20020223042828.E16048@blossom.cjclark.org> <20020223124535.GB52291@sunbay.com> <20020223072520.J16048@blossom.cjclark.org> <20020224084702.GB31243@sunbay.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Feb 24, 2002 at 10:47:02AM +0200, Ruslan Ermilov wrote:
> On Sat, Feb 23, 2002 at 07:25:20AM -0800, Crist J. Clark wrote:
[snip]
> > If you are correct, and you can cross interfaces, this is a much
> > bigger problem. I didn't mention it in your first mail, but the
> > in_broadcast() function as used in the patch you sent is different
> > than the current implementation. in_broadacast() currently takes an
> > address and an interface. If you are correct, we'd have to loop
> > through the interface list... which makes this uglier.
> >
> No, I can't cross interfaces, /* Check for broadcast addresses. */
> in ip_input() doesn't allow this.
Exactly, I found that code early on. I had considered possibly
dropping all TCP right then and there, but decided that was not the
right place to do it.
> I'm talking about a different
> scenario here, multiple IP networks per interface:
>
> host# netstat -rn -finet | grep -w default
> default 192.168.4.65 UGSc 4 387 rl0
>
> gateway# ifconfig fxp0 inet
> fxp0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> inet 192.168.4.65 netmask 0xffffff00 broadcast 192.168.4.255
> inet 192.168.100.1 netmask 0xffffff00 broadcast 192.168.100.255
Right. But to me, that seems at best suboptimal and at worst slightly
misconfigured. A host shouldn't be routing traffic over one
interface. This is the kind of route that will generate ICMP
redirects.
But none of this really has anything to do with the problem. I am
going to test the final revision I made after reviewing the patches
you presented earlier in this thread and then make the commit.
--
Crist J. Clark | cjclark@alum.mit.edu
| cjclark@jhu.edu
http://people.freebsd.org/~cjc/ | cjc@freebsd.org
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020224103616.S16048>