Date: Wed, 3 Apr 2002 21:45:30 -0800 From: "Crist J. Clark" <crist.clark@attbi.com> To: Sebastien Petit <spe@selectbourse.net> Cc: Scott Ullrich <sullrich@CRE8.COM>, "'Barney Wolff'" <barney@databus.com>, freebsd-net@FreeBSD.ORG Subject: Re: HUT Project Message-ID: <20020403214530.A57543@blossom.cjclark.org> In-Reply-To: <20020403100144.E148FBADD@sbserv0.intra.selectbourse.net>; from spe@selectbourse.net on Wed, Apr 03, 2002 at 12:06:20PM %2B0200 References: <2F6DCE1EFAB3BC418B5C324F13934C96016C9521@exchange.corp.cre8.com> <20020402170922.G52193@blossom.cjclark.org> <20020403100144.E148FBADD@sbserv0.intra.selectbourse.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Apr 03, 2002 at 12:06:20PM +0200, Sebastien Petit wrote:
[snip]
> Design of freevrrpd cause a problem actually because when a MASTER server
> leave LAN (cable problem), SLAVE take his place and send gratuitous ARP for
> update ARP cache of all hosts on the same LAN.
That's not really accurate. The reason a backup router who becomes
master is required to send a gratuitous ARP is so that the learning
bridges (a.k.a. switches) can learn which port the MAC address is
on. Since the MAC-to-IP relationship never actually changes, there
isn't really any need to update the ARP cache of hosts (that's kinda
the whole idea).
> Normally, I don't need that if
> I can set one ethernet address and one VIP on one alias. This method cause a
> problem when MASTER is living again because it don't send any Gratuitous ARP
> for reupdating all ARP caches of all hosts on the same LAN with his ethernet
> address.
Huh?
> So, my question is simple, is there a mechanism like netgraph or TAP that
> permits me to do that:
>
> xl0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> options=3<rxcsum,txcsum>
> /* Real address of the server on the first LAN 1 */
> inet 172.16.1.1 netmask 0xffff0000 broadcast 172.16.255.255
> ether 00:b0:d0:5e:3a:04
>
> xl1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> options=3<rxcsum,txcsum>
> /* Real address of the server on the LAN 2 */
> inet 10.0.0.1 netmask 0xff000000 broadcast 10.255.255.255
> ether 00:b0:d0:5e:3a:10
>
> /* Alias on xl0 with ethernet address 00:00:5E:00:01:01 because this is the
> VRID 1 */
> xl0:0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> options=3<rxcsum,txcsum>
> inet 172.16.2.1 netmask 0xffff0000 broadcast 172.16.255.255
> ether 00:00:5E:00:01:01
>
> /* Alias on xl1 with ethernet address 00:00:5E:00:01:01 becasue this is the
> VRID 1 on the LAN 2 (not the same as LAN1) */
> xl1:0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
> options=3<rxcsum,txcsum>
> inet 10.0.1.1 netmask 0xff000000 broadcast 10.255.255.255
> ether 00:00:5E:00:01:01
>
> I think that TAP interface cannot permit me to do that because I can't attach
> one tap interface on one physical interface. I can have multiple
> 00:00:5E:00:01:01 MAC addresses on multiple LAN connected on multiple
> physical interfaces of the same host.
> My wish is to implement VRRP as clean as I can but there is some
> limitations...
> Any idea to implement that correctly under FreeBSD ?
One point. I don't see any reason to maintain the separate xl[01]
interfaces with other MAC addresses in this example.
--
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?20020403214530.A57543>