Date: Mon, 19 Aug 2002 14:19:19 -0400 From: Barney Wolff <barney@tp.databus.com> To: net@freebsd.org Subject: Re: Bridging when one interface has no carrier Message-ID: <20020819181919.GA9000@tp.databus.com> In-Reply-To: <3D612D6A.9020604@isi.edu> References: <3D61224B.2020902@isi.edu> <20020819102951.A38869@iguana.icir.org> <3D612D6A.9020604@isi.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
I don't recall that FreeBSD has ever had the "strong host model" property and (as I just confirmed by test) it doesn't have it now. If anything, it would make sense to reverse the test, and only allow packets arriving on the "wrong" interface when the right interface is down. But I'd rather see it depend on net.inet.ip.forwarding, or just keep using the loose model. The test ifp->if_flags & IFF_UP would seem to be applied to the interface the frame was received on, not the interface matching the dest IP addr. That makes sense, as it's likely to be garbage if it came on an apparently down interface. Out of idle curiosity, why do interfaces have to have IPs assigned to do bridging? That's not how "real" bridges/switches work. (I should probably go search the archives, as this has surely been answered before.) On Mon, Aug 19, 2002 at 10:39:54AM -0700, Lars Eggert wrote: > > I like the first idea (bypass when bridging) better. Link layers usually > follow the "strong host model" (see RFC 1122), where inbound packets are > dropped unless they match the destination address of the inbound > interface. (Compared to the "weak host model" of IP, which acceptes > inbound packets when their destination address matches *any* local > interface.) Disabling that test altogether might weaken that property? -- Barney Wolff I'm available by contract or FT: http://www.databus.com/bwresume.pdf 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?20020819181919.GA9000>