This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


Re: [Xen-users] bonding + vlans ALMOST working, NO ARP

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: Re: [Xen-users] bonding + vlans ALMOST working, NO ARP
From: "Eli Criffield" <elicriffield@xxxxxxxxx>
Date: Fri, 1 Sep 2006 10:19:32 -0500
Delivery-date: Fri, 01 Sep 2006 08:20:17 -0700
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=ukn0iyugkkuxyACoPnY/WPoyX7CNdNJ3RxND90uvTyJ6dJxQVZtz5vg5Vwr8prgRY0naN1MM5e72GCJw27bKjnJ5CPM4trpP9MWnpZ6FkEYYrTBCRYjwQjkKT6JAdixlvz0ayS8CF3+m1lCdd5fb/tPzedhqPhgULrI9m5khly8=
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
In-reply-to: <44F844A2.2040808@xxxxxxxx>
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
References: <18e3f33d0608310859o319e5536p24fef2bd3d2cd004@xxxxxxxxxxxxxx> <44F844A2.2040808@xxxxxxxx>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
Nea all mac's are set static so the switchs and routers with arp cache
don't have to update anything.


On 9/1/06, Tomasz Chmielewski <mangoo@xxxxxxxx> wrote:
Eli Criffield wrote:
> I'm very close to getting bonding + vlans + bridging to work.
> The only part that's not working is ARP.
> Why i have no idea? But if put in an arp entry for the default route
> manually with 'arp -s 00:00:0C:07:0C:C9' everything
> works great.  (Everything that goes though the default route anyway).
> Now if i run tcpdump on any of the interfaces involved in the bridge,
> eth0, eth1, bond0, vlan151 or vlan151-br, vif0,0, or veth0 (i don't
> rename any interfaces) then it works fine, I get arp replays. But if i
> use 'ip' to turn on promisc any of the interfaces it doesn't work,
> just like it doesn't work without promisc off.
> The script is a little diffrent then the last one i posted, I'll atach
> it at the end.
> Basically I'm combining eth0 and eth1 into bond0, add vlans to bond0
> named "vlanXXX"  make a bridge named vlanXXX-br and add vlanXXX to it,
> and then domU's are started with bridge=vlanXXX-br in there xen config
> file to decide what vlan they should be part of.
> So what is tcpdump doing to make arp work???

Doesn't it look like this issue:


3.3. Why does my new domain receive no network traffic until after it
initiates an outgoing connection?

This is an issue that occurs under the following circumstances:

     * You do not specify the domain's MAC address, causing a random MAC
address to be selected at domain creation time
     * The upstream router has a local ARP cache

When a domain is destroyed, the host system's ARP cache is purged of
addresses for the defunct virtual NIC. If the domain is recreated it is
likely to be allocated a different random MAC address. This is no
problem for the host machine, but the external switch/router still knows
the "old" MAC address. The switch/router requires an outbound packet to
the outside world to implicitly eradicate the old ARP-cache entry. This
is not really Xen's fault at all, just a property of the implementation
of ARP resolution.

Tomasz Chmielewski

Xen-users mailing list

<Prev in Thread] Current Thread [Next in Thread>